{
  "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": 0.05185335,
      "auxiliary_loss_mlp": 0.02433534,
      "balance_loss_clip": 2.59465408,
      "balance_loss_mlp": 1.87963986,
      "epoch": 0.00012024289063909097,
      "flos": 24932483919360.0,
      "grad_norm": 40.4480724223927,
      "language_loss": 2.58550739,
      "learning_rate": 0.0,
      "loss": 1.9090879,
      "num_input_tokens_seen": 20375,
      "step": 1,
      "time_per_iteration": 15.124518394470215
    },
    {
      "auxiliary_loss_clip": 0.0350525,
      "auxiliary_loss_mlp": 0.01638677,
      "balance_loss_clip": 1.73121238,
      "balance_loss_mlp": 1.2682699,
      "epoch": 0.00024048578127818193,
      "flos": 30664624377600.0,
      "grad_norm": 55.24028104111834,
      "language_loss": 1.8909446,
      "learning_rate": 5.021476677069823e-07,
      "loss": 1.942384,
      "num_input_tokens_seen": 39035,
      "step": 2,
      "time_per_iteration": 2.724127769470215
    },
    {
      "auxiliary_loss_clip": 0.03468621,
      "auxiliary_loss_mlp": 0.01588339,
      "balance_loss_clip": 1.72706831,
      "balance_loss_mlp": 1.22002971,
      "epoch": 0.0003607286719172729,
      "flos": 19026227969280.0,
      "grad_norm": 41.34773921452883,
      "language_loss": 1.6165818,
      "learning_rate": 7.958852231401551e-07,
      "loss": 1.66715133,
      "num_input_tokens_seen": 57600,
      "step": 3,
      "time_per_iteration": 2.657543897628784
    },
    {
      "auxiliary_loss_clip": 0.03496814,
      "auxiliary_loss_mlp": 0.01637831,
      "balance_loss_clip": 1.73247063,
      "balance_loss_mlp": 1.26723325,
      "epoch": 0.00048097156255636386,
      "flos": 19316314206720.0,
      "grad_norm": 37.020231505560474,
      "language_loss": 1.64451838,
      "learning_rate": 1.0042953354139647e-06,
      "loss": 1.6958648,
      "num_input_tokens_seen": 76465,
      "step": 4,
      "time_per_iteration": 2.6132500171661377
    },
    {
      "auxiliary_loss_clip": 0.03469621,
      "auxiliary_loss_mlp": 0.01594352,
      "balance_loss_clip": 1.72772098,
      "balance_loss_mlp": 1.233482,
      "epoch": 0.0006012144531954548,
      "flos": 13991264893440.0,
      "grad_norm": 54.98524326539907,
      "language_loss": 1.93146873,
      "learning_rate": 1.1659507774310057e-06,
      "loss": 1.98210859,
      "num_input_tokens_seen": 94350,
      "step": 5,
      "time_per_iteration": 2.7354953289031982
    },
    {
      "auxiliary_loss_clip": 0.03499077,
      "auxiliary_loss_mlp": 0.01605233,
      "balance_loss_clip": 1.73888242,
      "balance_loss_mlp": 1.24131131,
      "epoch": 0.0007214573438345458,
      "flos": 23148988225920.0,
      "grad_norm": 44.151142915898504,
      "language_loss": 1.6085887,
      "learning_rate": 1.2980328908471373e-06,
      "loss": 1.65963173,
      "num_input_tokens_seen": 114595,
      "step": 6,
      "time_per_iteration": 3.0918002128601074
    },
    {
      "auxiliary_loss_clip": 0.03247097,
      "auxiliary_loss_mlp": 0.01529816,
      "balance_loss_clip": 1.77931428,
      "balance_loss_mlp": 1.16055298,
      "epoch": 0.0008417002344736367,
      "flos": 67663246170240.0,
      "grad_norm": 6.240704648572775,
      "language_loss": 0.81465656,
      "learning_rate": 1.4097067265369432e-06,
      "loss": 0.86242568,
      "num_input_tokens_seen": 179590,
      "step": 7,
      "time_per_iteration": 3.2736217975616455
    },
    {
      "auxiliary_loss_clip": 0.03500811,
      "auxiliary_loss_mlp": 0.01674233,
      "balance_loss_clip": 1.72733915,
      "balance_loss_mlp": 1.29505169,
      "epoch": 0.0009619431251127277,
      "flos": 21281381504640.0,
      "grad_norm": 41.08012040211905,
      "language_loss": 1.58913291,
      "learning_rate": 1.506443003120947e-06,
      "loss": 1.64088345,
      "num_input_tokens_seen": 195090,
      "step": 8,
      "time_per_iteration": 2.8999123573303223
    },
    {
      "auxiliary_loss_clip": 0.03451107,
      "auxiliary_loss_mlp": 0.01623617,
      "balance_loss_clip": 1.73023951,
      "balance_loss_mlp": 1.257406,
      "epoch": 0.0010821860157518186,
      "flos": 23331342597120.0,
      "grad_norm": 17.49076040670353,
      "language_loss": 1.47781503,
      "learning_rate": 1.5917704462803102e-06,
      "loss": 1.52856231,
      "num_input_tokens_seen": 211635,
      "step": 9,
      "time_per_iteration": 2.888622283935547
    },
    {
      "auxiliary_loss_clip": 0.03459074,
      "auxiliary_loss_mlp": 0.01637403,
      "balance_loss_clip": 1.73184085,
      "balance_loss_mlp": 1.27100158,
      "epoch": 0.0012024289063909096,
      "flos": 17010166337280.0,
      "grad_norm": 13.379162811774234,
      "language_loss": 1.5282774,
      "learning_rate": 1.6680984451379884e-06,
      "loss": 1.57924223,
      "num_input_tokens_seen": 224705,
      "step": 10,
      "time_per_iteration": 2.756546974182129
    },
    {
      "auxiliary_loss_clip": 0.03479782,
      "auxiliary_loss_mlp": 0.01619238,
      "balance_loss_clip": 1.73374796,
      "balance_loss_mlp": 1.24692321,
      "epoch": 0.0013226717970300007,
      "flos": 21288133261440.0,
      "grad_norm": 14.39391284296686,
      "language_loss": 1.3245579,
      "learning_rate": 1.7371455188905097e-06,
      "loss": 1.37554812,
      "num_input_tokens_seen": 244635,
      "step": 11,
      "time_per_iteration": 2.78688383102417
    },
    {
      "auxiliary_loss_clip": 0.03472403,
      "auxiliary_loss_mlp": 0.01589809,
      "balance_loss_clip": 1.72860444,
      "balance_loss_mlp": 1.22207201,
      "epoch": 0.0014429146876690916,
      "flos": 27237884935680.0,
      "grad_norm": 10.467330461408412,
      "language_loss": 1.25333619,
      "learning_rate": 1.8001805585541196e-06,
      "loss": 1.3039583,
      "num_input_tokens_seen": 265765,
      "step": 12,
      "time_per_iteration": 2.968977212905884
    },
    {
      "auxiliary_loss_clip": 0.0343064,
      "auxiliary_loss_mlp": 0.01621842,
      "balance_loss_clip": 1.7212956,
      "balance_loss_mlp": 1.25200772,
      "epoch": 0.0015631575783081825,
      "flos": 19062174504960.0,
      "grad_norm": 6.541967633443749,
      "language_loss": 1.28970003,
      "learning_rate": 1.8581671739548328e-06,
      "loss": 1.34022498,
      "num_input_tokens_seen": 283500,
      "step": 13,
      "time_per_iteration": 2.778419256210327
    },
    {
      "auxiliary_loss_clip": 0.03440524,
      "auxiliary_loss_mlp": 0.01579669,
      "balance_loss_clip": 1.72803009,
      "balance_loss_mlp": 1.21574664,
      "epoch": 0.0016834004689472734,
      "flos": 48139473985920.0,
      "grad_norm": 6.231639441799505,
      "language_loss": 1.1346221,
      "learning_rate": 1.9118543942439254e-06,
      "loss": 1.18482399,
      "num_input_tokens_seen": 305685,
      "step": 14,
      "time_per_iteration": 4.086289405822754
    },
    {
      "auxiliary_loss_clip": 0.03415908,
      "auxiliary_loss_mlp": 0.01638095,
      "balance_loss_clip": 1.71943903,
      "balance_loss_mlp": 1.27131164,
      "epoch": 0.0018036433595863645,
      "flos": 34970026314240.0,
      "grad_norm": 5.366564514175447,
      "language_loss": 1.12644029,
      "learning_rate": 1.961836000571161e-06,
      "loss": 1.17698026,
      "num_input_tokens_seen": 327340,
      "step": 15,
      "time_per_iteration": 3.9294369220733643
    },
    {
      "auxiliary_loss_clip": 0.03122865,
      "auxiliary_loss_mlp": 0.01478761,
      "balance_loss_clip": 1.75881386,
      "balance_loss_mlp": 1.12170529,
      "epoch": 0.0019238862502254555,
      "flos": 59768284440960.0,
      "grad_norm": 3.844140140725537,
      "language_loss": 0.64833224,
      "learning_rate": 2.0085906708279293e-06,
      "loss": 0.69434845,
      "num_input_tokens_seen": 382710,
      "step": 16,
      "time_per_iteration": 3.288327693939209
    },
    {
      "auxiliary_loss_clip": 0.03384965,
      "auxiliary_loss_mlp": 0.01584047,
      "balance_loss_clip": 1.71822166,
      "balance_loss_mlp": 1.22088802,
      "epoch": 0.0020441291408645466,
      "flos": 20814543417600.0,
      "grad_norm": 4.289096016947757,
      "language_loss": 1.1584332,
      "learning_rate": 2.0525099325728135e-06,
      "loss": 1.20812321,
      "num_input_tokens_seen": 400890,
      "step": 17,
      "time_per_iteration": 3.0331172943115234
    },
    {
      "auxiliary_loss_clip": 0.03071732,
      "auxiliary_loss_mlp": 0.01458972,
      "balance_loss_clip": 1.74686003,
      "balance_loss_mlp": 1.10649359,
      "epoch": 0.0021643720315036373,
      "flos": 63857001582720.0,
      "grad_norm": 3.604440557412118,
      "language_loss": 0.72346109,
      "learning_rate": 2.0939181139872922e-06,
      "loss": 0.76876813,
      "num_input_tokens_seen": 462605,
      "step": 18,
      "time_per_iteration": 3.3093671798706055
    },
    {
      "auxiliary_loss_clip": 0.0339064,
      "auxiliary_loss_mlp": 0.01601546,
      "balance_loss_clip": 1.71849382,
      "balance_loss_mlp": 1.24925852,
      "epoch": 0.0022846149221427284,
      "flos": 31284981192960.0,
      "grad_norm": 4.8730622052850485,
      "language_loss": 1.01332808,
      "learning_rate": 2.1330868934640175e-06,
      "loss": 1.06324995,
      "num_input_tokens_seen": 483280,
      "step": 19,
      "time_per_iteration": 3.05334210395813
    },
    {
      "auxiliary_loss_clip": 0.03005865,
      "auxiliary_loss_mlp": 0.01438341,
      "balance_loss_clip": 1.73370576,
      "balance_loss_mlp": 1.09196603,
      "epoch": 0.002404857812781819,
      "flos": 51083648161920.0,
      "grad_norm": 3.5596986444807834,
      "language_loss": 0.76430011,
      "learning_rate": 2.170246112844971e-06,
      "loss": 0.80874217,
      "num_input_tokens_seen": 537620,
      "step": 20,
      "time_per_iteration": 3.080176591873169
    },
    {
      "auxiliary_loss_clip": 0.0329226,
      "auxiliary_loss_mlp": 0.01540556,
      "balance_loss_clip": 1.7006886,
      "balance_loss_mlp": 1.18845916,
      "epoch": 0.0025251007034209102,
      "flos": 15815347309440.0,
      "grad_norm": 6.399826225561999,
      "language_loss": 1.01332736,
      "learning_rate": 2.2055919496770983e-06,
      "loss": 1.06165552,
      "num_input_tokens_seen": 555760,
      "step": 21,
      "time_per_iteration": 2.818455696105957
    },
    {
      "auxiliary_loss_clip": 0.03260464,
      "auxiliary_loss_mlp": 0.01506764,
      "balance_loss_clip": 1.69960642,
      "balance_loss_mlp": 1.16553926,
      "epoch": 0.0026453435940600014,
      "flos": 37851857458560.0,
      "grad_norm": 5.118361462888217,
      "language_loss": 0.89574873,
      "learning_rate": 2.2392931865974923e-06,
      "loss": 0.94342101,
      "num_input_tokens_seen": 578450,
      "step": 22,
      "time_per_iteration": 3.0259480476379395
    },
    {
      "auxiliary_loss_clip": 0.03204627,
      "auxiliary_loss_mlp": 0.0151953,
      "balance_loss_clip": 1.68441713,
      "balance_loss_mlp": 1.17048562,
      "epoch": 0.002765586484699092,
      "flos": 21141976821120.0,
      "grad_norm": 4.92583740794861,
      "language_loss": 1.01950443,
      "learning_rate": 2.271496085962064e-06,
      "loss": 1.066746,
      "num_input_tokens_seen": 596145,
      "step": 23,
      "time_per_iteration": 2.886434316635132
    },
    {
      "auxiliary_loss_clip": 0.03203667,
      "auxiliary_loss_mlp": 0.01551569,
      "balance_loss_clip": 1.68738294,
      "balance_loss_mlp": 1.21148896,
      "epoch": 0.002885829375338183,
      "flos": 20667381396480.0,
      "grad_norm": 3.4512426182530604,
      "language_loss": 1.02640259,
      "learning_rate": 2.3023282262611022e-06,
      "loss": 1.07395506,
      "num_input_tokens_seen": 614920,
      "step": 24,
      "time_per_iteration": 2.821049451828003
    },
    {
      "auxiliary_loss_clip": 0.03191176,
      "auxiliary_loss_mlp": 0.01506546,
      "balance_loss_clip": 1.68249726,
      "balance_loss_mlp": 1.17047155,
      "epoch": 0.003006072265977274,
      "flos": 34823869873920.0,
      "grad_norm": 3.4080988107624446,
      "language_loss": 0.9259091,
      "learning_rate": 2.3319015548620114e-06,
      "loss": 0.97288644,
      "num_input_tokens_seen": 636060,
      "step": 25,
      "time_per_iteration": 3.018259286880493
    },
    {
      "auxiliary_loss_clip": 0.03175901,
      "auxiliary_loss_mlp": 0.01514845,
      "balance_loss_clip": 1.67973578,
      "balance_loss_mlp": 1.17209435,
      "epoch": 0.003126315156616365,
      "flos": 24422021118720.0,
      "grad_norm": 2.551703275296398,
      "language_loss": 0.9291147,
      "learning_rate": 2.3603148416618152e-06,
      "loss": 0.97602218,
      "num_input_tokens_seen": 655575,
      "step": 26,
      "time_per_iteration": 2.8973653316497803
    },
    {
      "auxiliary_loss_clip": 0.03138396,
      "auxiliary_loss_mlp": 0.0149702,
      "balance_loss_clip": 1.66958749,
      "balance_loss_mlp": 1.16285288,
      "epoch": 0.003246558047255456,
      "flos": 23622326674560.0,
      "grad_norm": 2.456157128322032,
      "language_loss": 1.01025105,
      "learning_rate": 2.3876556694204647e-06,
      "loss": 1.05660522,
      "num_input_tokens_seen": 675730,
      "step": 27,
      "time_per_iteration": 2.9213926792144775
    },
    {
      "auxiliary_loss_clip": 0.0308387,
      "auxiliary_loss_mlp": 0.01488497,
      "balance_loss_clip": 1.67152953,
      "balance_loss_mlp": 1.15375781,
      "epoch": 0.003366800937894547,
      "flos": 17820275725440.0,
      "grad_norm": 2.4096232349357534,
      "language_loss": 0.90869963,
      "learning_rate": 2.414002061950908e-06,
      "loss": 0.95442331,
      "num_input_tokens_seen": 694605,
      "step": 28,
      "time_per_iteration": 2.9950122833251953
    },
    {
      "auxiliary_loss_clip": 0.03086801,
      "auxiliary_loss_mlp": 0.01489512,
      "balance_loss_clip": 1.66141748,
      "balance_loss_mlp": 1.16297388,
      "epoch": 0.003487043828533638,
      "flos": 24426115269120.0,
      "grad_norm": 2.6213439079831264,
      "language_loss": 1.00017726,
      "learning_rate": 2.4394238264681557e-06,
      "loss": 1.0459404,
      "num_input_tokens_seen": 714340,
      "step": 29,
      "time_per_iteration": 3.02445387840271
    },
    {
      "auxiliary_loss_clip": 0.03056916,
      "auxiliary_loss_mlp": 0.01423809,
      "balance_loss_clip": 1.65792704,
      "balance_loss_mlp": 1.11214781,
      "epoch": 0.003607286719172729,
      "flos": 26140311002880.0,
      "grad_norm": 2.0568852810562825,
      "language_loss": 0.99631858,
      "learning_rate": 2.4639836682781433e-06,
      "loss": 1.04112577,
      "num_input_tokens_seen": 734470,
      "step": 30,
      "time_per_iteration": 2.9185898303985596
    },
    {
      "auxiliary_loss_clip": 0.03020404,
      "auxiliary_loss_mlp": 0.01438776,
      "balance_loss_clip": 1.66035163,
      "balance_loss_mlp": 1.12330055,
      "epoch": 0.00372752960981182,
      "flos": 20593082113920.0,
      "grad_norm": 2.250147371903613,
      "language_loss": 1.00315619,
      "learning_rate": 2.487738122623307e-06,
      "loss": 1.04774785,
      "num_input_tokens_seen": 753380,
      "step": 31,
      "time_per_iteration": 2.8238682746887207
    },
    {
      "auxiliary_loss_clip": 0.02909972,
      "auxiliary_loss_mlp": 0.01405161,
      "balance_loss_clip": 1.62567604,
      "balance_loss_mlp": 1.11238337,
      "epoch": 0.003847772500450911,
      "flos": 22674608282880.0,
      "grad_norm": 3.456212399523975,
      "language_loss": 0.98862624,
      "learning_rate": 2.510738338534912e-06,
      "loss": 1.03177762,
      "num_input_tokens_seen": 772105,
      "step": 32,
      "time_per_iteration": 2.8929226398468018
    },
    {
      "auxiliary_loss_clip": 0.02848385,
      "auxiliary_loss_mlp": 0.01436734,
      "balance_loss_clip": 1.60597122,
      "balance_loss_mlp": 1.12450135,
      "epoch": 0.003968015391090002,
      "flos": 17967796882560.0,
      "grad_norm": 2.344068393941192,
      "language_loss": 1.0257318,
      "learning_rate": 2.5330307420306648e-06,
      "loss": 1.06858301,
      "num_input_tokens_seen": 788955,
      "step": 33,
      "time_per_iteration": 2.7943201065063477
    },
    {
      "auxiliary_loss_clip": 0.02792377,
      "auxiliary_loss_mlp": 0.01416982,
      "balance_loss_clip": 1.59691072,
      "balance_loss_mlp": 1.10951734,
      "epoch": 0.004088258281729093,
      "flos": 27304103658240.0,
      "grad_norm": 2.560357003134473,
      "language_loss": 0.88235676,
      "learning_rate": 2.554657600279796e-06,
      "loss": 0.92445034,
      "num_input_tokens_seen": 810230,
      "step": 34,
      "time_per_iteration": 2.8532426357269287
    },
    {
      "auxiliary_loss_clip": 0.02733954,
      "auxiliary_loss_mlp": 0.01385347,
      "balance_loss_clip": 1.58420229,
      "balance_loss_mlp": 1.08722806,
      "epoch": 0.004208501172368184,
      "flos": 23258587599360.0,
      "grad_norm": 2.093911325986375,
      "language_loss": 1.0344528,
      "learning_rate": 2.5756575039679493e-06,
      "loss": 1.0756458,
      "num_input_tokens_seen": 829780,
      "step": 35,
      "time_per_iteration": 2.988123893737793
    },
    {
      "auxiliary_loss_clip": 0.02735044,
      "auxiliary_loss_mlp": 0.01408631,
      "balance_loss_clip": 1.5803895,
      "balance_loss_mlp": 1.11489916,
      "epoch": 0.0043287440630072746,
      "flos": 17312104062720.0,
      "grad_norm": 2.061206029369176,
      "language_loss": 0.95139968,
      "learning_rate": 2.5960657816942747e-06,
      "loss": 0.99283636,
      "num_input_tokens_seen": 848695,
      "step": 36,
      "time_per_iteration": 2.8112614154815674
    },
    {
      "auxiliary_loss_clip": 0.02238624,
      "auxiliary_loss_mlp": 0.01335062,
      "balance_loss_clip": 1.53190207,
      "balance_loss_mlp": 1.06192935,
      "epoch": 0.004448986953646365,
      "flos": 53092491160320.0,
      "grad_norm": 1.4277416072482807,
      "language_loss": 0.61032999,
      "learning_rate": 2.6159148575788668e-06,
      "loss": 0.64606678,
      "num_input_tokens_seen": 906730,
      "step": 37,
      "time_per_iteration": 3.436404228210449
    },
    {
      "auxiliary_loss_clip": 0.02629954,
      "auxiliary_loss_mlp": 0.01370801,
      "balance_loss_clip": 1.55763412,
      "balance_loss_mlp": 1.09804964,
      "epoch": 0.004569229844285457,
      "flos": 13444165866240.0,
      "grad_norm": 2.4946954402368133,
      "language_loss": 0.98737931,
      "learning_rate": 2.635234561171e-06,
      "loss": 1.02738678,
      "num_input_tokens_seen": 925125,
      "step": 38,
      "time_per_iteration": 2.8793773651123047
    },
    {
      "auxiliary_loss_clip": 0.02597857,
      "auxiliary_loss_mlp": 0.01371571,
      "balance_loss_clip": 1.54921222,
      "balance_loss_mlp": 1.09119105,
      "epoch": 0.0046894727349245475,
      "flos": 16209609966720.0,
      "grad_norm": 4.024938973739427,
      "language_loss": 0.93987781,
      "learning_rate": 2.6540523970949877e-06,
      "loss": 0.97957206,
      "num_input_tokens_seen": 939970,
      "step": 39,
      "time_per_iteration": 3.034435272216797
    },
    {
      "auxiliary_loss_clip": 0.02548622,
      "auxiliary_loss_mlp": 0.01362675,
      "balance_loss_clip": 1.54095733,
      "balance_loss_mlp": 1.08973384,
      "epoch": 0.004809715625563638,
      "flos": 23914244505600.0,
      "grad_norm": 2.5883133214017384,
      "language_loss": 0.9252919,
      "learning_rate": 2.6723937805519533e-06,
      "loss": 0.96440482,
      "num_input_tokens_seen": 957470,
      "step": 40,
      "time_per_iteration": 3.7699296474456787
    },
    {
      "auxiliary_loss_clip": 0.02528984,
      "auxiliary_loss_mlp": 0.01340672,
      "balance_loss_clip": 1.52901721,
      "balance_loss_mlp": 1.08089066,
      "epoch": 0.00492995851620273,
      "flos": 20773030273920.0,
      "grad_norm": 2.29589790401655,
      "language_loss": 0.92849833,
      "learning_rate": 2.690282243737839e-06,
      "loss": 0.96719486,
      "num_input_tokens_seen": 976405,
      "step": 41,
      "time_per_iteration": 4.826427459716797
    },
    {
      "auxiliary_loss_clip": 0.02516023,
      "auxiliary_loss_mlp": 0.0132808,
      "balance_loss_clip": 1.51832938,
      "balance_loss_mlp": 1.06562912,
      "epoch": 0.0050502014068418205,
      "flos": 20338655103360.0,
      "grad_norm": 2.678993185822361,
      "language_loss": 0.99334121,
      "learning_rate": 2.7077396173840807e-06,
      "loss": 1.03178227,
      "num_input_tokens_seen": 994690,
      "step": 42,
      "time_per_iteration": 2.8256630897521973
    },
    {
      "auxiliary_loss_clip": 0.02490952,
      "auxiliary_loss_mlp": 0.01341936,
      "balance_loss_clip": 1.5156579,
      "balance_loss_mlp": 1.07891214,
      "epoch": 0.005170444297480911,
      "flos": 25994872834560.0,
      "grad_norm": 2.4240305145624,
      "language_loss": 0.9267841,
      "learning_rate": 2.7247861909342594e-06,
      "loss": 0.96511292,
      "num_input_tokens_seen": 1015615,
      "step": 43,
      "time_per_iteration": 3.077012062072754
    },
    {
      "auxiliary_loss_clip": 0.02424027,
      "auxiliary_loss_mlp": 0.01355532,
      "balance_loss_clip": 1.49722731,
      "balance_loss_mlp": 1.09765887,
      "epoch": 0.005290687188120003,
      "flos": 20954055841920.0,
      "grad_norm": 2.1686668542059713,
      "language_loss": 0.8290838,
      "learning_rate": 2.7414408543044743e-06,
      "loss": 0.86687934,
      "num_input_tokens_seen": 1031255,
      "step": 44,
      "time_per_iteration": 3.0378284454345703
    },
    {
      "auxiliary_loss_clip": 0.02410129,
      "auxiliary_loss_mlp": 0.0132397,
      "balance_loss_clip": 1.49380207,
      "balance_loss_mlp": 1.07239115,
      "epoch": 0.005410930078759093,
      "flos": 15851401585920.0,
      "grad_norm": 7.028687941224897,
      "language_loss": 0.79304945,
      "learning_rate": 2.7577212237113157e-06,
      "loss": 0.83039045,
      "num_input_tokens_seen": 1048295,
      "step": 45,
      "time_per_iteration": 2.830261707305908
    },
    {
      "auxiliary_loss_clip": 0.02388573,
      "auxiliary_loss_mlp": 0.01321429,
      "balance_loss_clip": 1.4874568,
      "balance_loss_mlp": 1.06908679,
      "epoch": 0.005531172969398184,
      "flos": 21104988791040.0,
      "grad_norm": 2.264610562022804,
      "language_loss": 1.0418762,
      "learning_rate": 2.7736437536690466e-06,
      "loss": 1.07897627,
      "num_input_tokens_seen": 1067925,
      "step": 46,
      "time_per_iteration": 2.8702497482299805
    },
    {
      "auxiliary_loss_clip": 0.02365727,
      "auxiliary_loss_mlp": 0.01296018,
      "balance_loss_clip": 1.48145151,
      "balance_loss_mlp": 1.06141376,
      "epoch": 0.005651415860037276,
      "flos": 20844887431680.0,
      "grad_norm": 1.9491775411636278,
      "language_loss": 1.07772648,
      "learning_rate": 2.789223836941131e-06,
      "loss": 1.11434388,
      "num_input_tokens_seen": 1088060,
      "step": 47,
      "time_per_iteration": 2.9034242630004883
    },
    {
      "auxiliary_loss_clip": 0.02329709,
      "auxiliary_loss_mlp": 0.01287748,
      "balance_loss_clip": 1.47227085,
      "balance_loss_mlp": 1.05962896,
      "epoch": 0.005771658750676366,
      "flos": 13260195383040.0,
      "grad_norm": 2.298104600796841,
      "language_loss": 1.0869168,
      "learning_rate": 2.8044758939680847e-06,
      "loss": 1.12309146,
      "num_input_tokens_seen": 1104130,
      "step": 48,
      "time_per_iteration": 2.8259263038635254
    },
    {
      "auxiliary_loss_clip": 0.02288665,
      "auxiliary_loss_mlp": 0.01296505,
      "balance_loss_clip": 1.46353304,
      "balance_loss_mlp": 1.06533456,
      "epoch": 0.005891901641315457,
      "flos": 24425396997120.0,
      "grad_norm": 3.3437307150582796,
      "language_loss": 1.01959813,
      "learning_rate": 2.8194134530738863e-06,
      "loss": 1.05544972,
      "num_input_tokens_seen": 1122900,
      "step": 49,
      "time_per_iteration": 2.914057493209839
    },
    {
      "auxiliary_loss_clip": 0.02263423,
      "auxiliary_loss_mlp": 0.01294982,
      "balance_loss_clip": 1.45742893,
      "balance_loss_mlp": 1.07258475,
      "epoch": 0.006012144531954548,
      "flos": 23076197314560.0,
      "grad_norm": 2.539005595426923,
      "language_loss": 0.9024263,
      "learning_rate": 2.834049222568994e-06,
      "loss": 0.93801039,
      "num_input_tokens_seen": 1140250,
      "step": 50,
      "time_per_iteration": 2.8540079593658447
    },
    {
      "auxiliary_loss_clip": 0.02280104,
      "auxiliary_loss_mlp": 0.01269549,
      "balance_loss_clip": 1.46038389,
      "balance_loss_mlp": 1.06527221,
      "epoch": 0.006132387422593639,
      "flos": 22528775064960.0,
      "grad_norm": 2.121873560301535,
      "language_loss": 0.9251225,
      "learning_rate": 2.848395155712969e-06,
      "loss": 0.96061903,
      "num_input_tokens_seen": 1160470,
      "step": 51,
      "time_per_iteration": 2.8759188652038574
    },
    {
      "auxiliary_loss_clip": 0.02246212,
      "auxiliary_loss_mlp": 0.01270323,
      "balance_loss_clip": 1.45885146,
      "balance_loss_mlp": 1.06432915,
      "epoch": 0.00625263031323273,
      "flos": 27628340751360.0,
      "grad_norm": 2.553519185204845,
      "language_loss": 0.97582018,
      "learning_rate": 2.8624625093687977e-06,
      "loss": 1.01098549,
      "num_input_tokens_seen": 1177605,
      "step": 52,
      "time_per_iteration": 2.7934319972991943
    },
    {
      "auxiliary_loss_clip": 0.0224054,
      "auxiliary_loss_mlp": 0.0126221,
      "balance_loss_clip": 1.45193815,
      "balance_loss_mlp": 1.05297363,
      "epoch": 0.006372873203871821,
      "flos": 23110671392640.0,
      "grad_norm": 2.2530380833144013,
      "language_loss": 0.88830531,
      "learning_rate": 2.876261897070029e-06,
      "loss": 0.92333281,
      "num_input_tokens_seen": 1197735,
      "step": 53,
      "time_per_iteration": 3.0498597621917725
    },
    {
      "auxiliary_loss_clip": 0.0222182,
      "auxiliary_loss_mlp": 0.01261327,
      "balance_loss_clip": 1.44919372,
      "balance_loss_mlp": 1.06105566,
      "epoch": 0.006493116094510912,
      "flos": 22856028900480.0,
      "grad_norm": 2.401777300318519,
      "language_loss": 0.92243266,
      "learning_rate": 2.889803337127447e-06,
      "loss": 0.95726418,
      "num_input_tokens_seen": 1216335,
      "step": 54,
      "time_per_iteration": 2.7963247299194336
    },
    {
      "auxiliary_loss_clip": 0.02206187,
      "auxiliary_loss_mlp": 0.01266015,
      "balance_loss_clip": 1.44370282,
      "balance_loss_mlp": 1.05964017,
      "epoch": 0.006613358985150003,
      "flos": 23071708114560.0,
      "grad_norm": 2.3859300876222624,
      "language_loss": 0.84876549,
      "learning_rate": 2.903096296321516e-06,
      "loss": 0.88348752,
      "num_input_tokens_seen": 1234480,
      "step": 55,
      "time_per_iteration": 2.797780752182007
    },
    {
      "auxiliary_loss_clip": 0.02172939,
      "auxiliary_loss_mlp": 0.01254272,
      "balance_loss_clip": 1.43711352,
      "balance_loss_mlp": 1.05886436,
      "epoch": 0.006733601875789094,
      "flos": 26537662229760.0,
      "grad_norm": 2.1001634179739392,
      "language_loss": 0.91604102,
      "learning_rate": 2.9161497296578907e-06,
      "loss": 0.95031315,
      "num_input_tokens_seen": 1253870,
      "step": 56,
      "time_per_iteration": 2.9598331451416016
    },
    {
      "auxiliary_loss_clip": 0.02164301,
      "auxiliary_loss_mlp": 0.01258346,
      "balance_loss_clip": 1.43020535,
      "balance_loss_mlp": 1.06522655,
      "epoch": 0.006853844766428185,
      "flos": 15523178083200.0,
      "grad_norm": 2.215194976434327,
      "language_loss": 0.8603946,
      "learning_rate": 2.928972116604173e-06,
      "loss": 0.89462101,
      "num_input_tokens_seen": 1270145,
      "step": 57,
      "time_per_iteration": 2.871432065963745
    },
    {
      "auxiliary_loss_clip": 0.0214085,
      "auxiliary_loss_mlp": 0.0125286,
      "balance_loss_clip": 1.42077827,
      "balance_loss_mlp": 1.06307864,
      "epoch": 0.006974087657067276,
      "flos": 24243760897920.0,
      "grad_norm": 2.0156335405022316,
      "language_loss": 1.01923656,
      "learning_rate": 2.9415714941751377e-06,
      "loss": 1.05317366,
      "num_input_tokens_seen": 1291365,
      "step": 58,
      "time_per_iteration": 2.8330562114715576
    },
    {
      "auxiliary_loss_clip": 0.02137264,
      "auxiliary_loss_mlp": 0.01234336,
      "balance_loss_clip": 1.42289662,
      "balance_loss_mlp": 1.05275667,
      "epoch": 0.007094330547706367,
      "flos": 25772513690880.0,
      "grad_norm": 1.8938591915910568,
      "language_loss": 0.93560141,
      "learning_rate": 2.9539554871897396e-06,
      "loss": 0.96931744,
      "num_input_tokens_seen": 1311535,
      "step": 59,
      "time_per_iteration": 2.9713799953460693
    },
    {
      "auxiliary_loss_clip": 0.02098636,
      "auxiliary_loss_mlp": 0.01247756,
      "balance_loss_clip": 1.41418004,
      "balance_loss_mlp": 1.06436443,
      "epoch": 0.007214573438345458,
      "flos": 21319015979520.0,
      "grad_norm": 2.1353651673232483,
      "language_loss": 0.97507596,
      "learning_rate": 2.9661313359851253e-06,
      "loss": 1.00853992,
      "num_input_tokens_seen": 1329420,
      "step": 60,
      "time_per_iteration": 2.9260361194610596
    },
    {
      "auxiliary_loss_clip": 0.02085896,
      "auxiliary_loss_mlp": 0.01243967,
      "balance_loss_clip": 1.40890956,
      "balance_loss_mlp": 1.07068479,
      "epoch": 0.007334816328984549,
      "flos": 24937088192640.0,
      "grad_norm": 2.1055018418407947,
      "language_loss": 0.93890011,
      "learning_rate": 2.978105921839922e-06,
      "loss": 0.97219872,
      "num_input_tokens_seen": 1349965,
      "step": 61,
      "time_per_iteration": 3.2079946994781494
    },
    {
      "auxiliary_loss_clip": 0.02068824,
      "auxiliary_loss_mlp": 0.0122212,
      "balance_loss_clip": 1.40309405,
      "balance_loss_mlp": 1.0543685,
      "epoch": 0.00745505921962364,
      "flos": 18510586277760.0,
      "grad_norm": 2.362703955112609,
      "language_loss": 0.72291899,
      "learning_rate": 2.9898857903302893e-06,
      "loss": 0.75582838,
      "num_input_tokens_seen": 1368915,
      "step": 62,
      "time_per_iteration": 2.854692220687866
    },
    {
      "auxiliary_loss_clip": 0.02053978,
      "auxiliary_loss_mlp": 0.01230505,
      "balance_loss_clip": 1.39704847,
      "balance_loss_mlp": 1.06199086,
      "epoch": 0.007575302110262731,
      "flos": 18477656484480.0,
      "grad_norm": 2.6033688101864447,
      "language_loss": 0.87621224,
      "learning_rate": 3.001477172817253e-06,
      "loss": 0.90905702,
      "num_input_tokens_seen": 1386805,
      "step": 63,
      "time_per_iteration": 2.839292287826538
    },
    {
      "auxiliary_loss_clip": 0.02035088,
      "auxiliary_loss_mlp": 0.01219528,
      "balance_loss_clip": 1.3924197,
      "balance_loss_mlp": 1.05826211,
      "epoch": 0.007695545000901822,
      "flos": 24973178382720.0,
      "grad_norm": 2.4802408981364645,
      "language_loss": 0.96154487,
      "learning_rate": 3.012886006241894e-06,
      "loss": 0.99409103,
      "num_input_tokens_seen": 1406190,
      "step": 64,
      "time_per_iteration": 2.898923635482788
    },
    {
      "auxiliary_loss_clip": 0.02033275,
      "auxiliary_loss_mlp": 0.01213702,
      "balance_loss_clip": 1.39344919,
      "balance_loss_mlp": 1.05663168,
      "epoch": 0.007815787891540913,
      "flos": 21324223451520.0,
      "grad_norm": 2.2258345459190374,
      "language_loss": 0.88132226,
      "learning_rate": 3.0241179513858383e-06,
      "loss": 0.91379201,
      "num_input_tokens_seen": 1425500,
      "step": 65,
      "time_per_iteration": 2.930673837661743
    },
    {
      "auxiliary_loss_clip": 0.02013862,
      "auxiliary_loss_mlp": 0.01228352,
      "balance_loss_clip": 1.38246441,
      "balance_loss_mlp": 1.06336641,
      "epoch": 0.007936030782180003,
      "flos": 21575777374080.0,
      "grad_norm": 2.1350493325315916,
      "language_loss": 0.87790543,
      "learning_rate": 3.035178409737647e-06,
      "loss": 0.91032761,
      "num_input_tokens_seen": 1442950,
      "step": 66,
      "time_per_iteration": 3.8758981227874756
    },
    {
      "auxiliary_loss_clip": 0.01997916,
      "auxiliary_loss_mlp": 0.0119593,
      "balance_loss_clip": 1.37966132,
      "balance_loss_mlp": 1.05011344,
      "epoch": 0.008056273672819095,
      "flos": 20120785159680.0,
      "grad_norm": 2.3731468169460297,
      "language_loss": 0.88661158,
      "learning_rate": 3.046072539090907e-06,
      "loss": 0.91855001,
      "num_input_tokens_seen": 1460915,
      "step": 67,
      "time_per_iteration": 4.797903299331665
    },
    {
      "auxiliary_loss_clip": 0.01976797,
      "auxiliary_loss_mlp": 0.01202246,
      "balance_loss_clip": 1.37254548,
      "balance_loss_mlp": 1.06157875,
      "epoch": 0.008176516563458186,
      "flos": 18333116156160.0,
      "grad_norm": 3.662888951917216,
      "language_loss": 1.04483604,
      "learning_rate": 3.056805267986779e-06,
      "loss": 1.07662642,
      "num_input_tokens_seen": 1478385,
      "step": 68,
      "time_per_iteration": 3.6564290523529053
    },
    {
      "auxiliary_loss_clip": 0.01963282,
      "auxiliary_loss_mlp": 0.01206784,
      "balance_loss_clip": 1.36568475,
      "balance_loss_mlp": 1.06220698,
      "epoch": 0.008296759454097276,
      "flos": 21872076664320.0,
      "grad_norm": 2.0516703415677027,
      "language_loss": 0.95429122,
      "learning_rate": 3.0673813091022194e-06,
      "loss": 0.98599184,
      "num_input_tokens_seen": 1497605,
      "step": 69,
      "time_per_iteration": 2.877664566040039
    },
    {
      "auxiliary_loss_clip": 0.01783168,
      "auxiliary_loss_mlp": 0.01193475,
      "balance_loss_clip": 1.4093554,
      "balance_loss_mlp": 1.05690885,
      "epoch": 0.008417002344736368,
      "flos": 63408228036480.0,
      "grad_norm": 1.3112540293831965,
      "language_loss": 0.62117922,
      "learning_rate": 3.0778051716749317e-06,
      "loss": 0.65094566,
      "num_input_tokens_seen": 1561150,
      "step": 70,
      "time_per_iteration": 3.382751941680908
    },
    {
      "auxiliary_loss_clip": 0.01938984,
      "auxiliary_loss_mlp": 0.01209784,
      "balance_loss_clip": 1.35544503,
      "balance_loss_mlp": 1.06978416,
      "epoch": 0.008537245235375458,
      "flos": 22966454286720.0,
      "grad_norm": 2.283432386402717,
      "language_loss": 0.90425324,
      "learning_rate": 3.0880811730470094e-06,
      "loss": 0.93574101,
      "num_input_tokens_seen": 1580605,
      "step": 71,
      "time_per_iteration": 2.8299033641815186
    },
    {
      "auxiliary_loss_clip": 0.01752322,
      "auxiliary_loss_mlp": 0.01160332,
      "balance_loss_clip": 1.39341569,
      "balance_loss_mlp": 1.0313952,
      "epoch": 0.008657488126014549,
      "flos": 61984046712960.0,
      "grad_norm": 1.171437747189355,
      "language_loss": 0.58690393,
      "learning_rate": 3.098213449401257e-06,
      "loss": 0.61603051,
      "num_input_tokens_seen": 1647535,
      "step": 72,
      "time_per_iteration": 3.305851697921753
    },
    {
      "auxiliary_loss_clip": 0.01907767,
      "auxiliary_loss_mlp": 0.01181609,
      "balance_loss_clip": 1.34569395,
      "balance_loss_mlp": 1.05286348,
      "epoch": 0.00877773101665364,
      "flos": 30296791152000.0,
      "grad_norm": 2.33621893584454,
      "language_loss": 0.98638523,
      "learning_rate": 3.1082059657570015e-06,
      "loss": 1.01727891,
      "num_input_tokens_seen": 1666770,
      "step": 73,
      "time_per_iteration": 2.9069511890411377
    },
    {
      "auxiliary_loss_clip": 0.01881045,
      "auxiliary_loss_mlp": 0.01189698,
      "balance_loss_clip": 1.33891714,
      "balance_loss_mlp": 1.06705499,
      "epoch": 0.00889797390729273,
      "flos": 23514056104320.0,
      "grad_norm": 4.091807571584644,
      "language_loss": 0.96691203,
      "learning_rate": 3.1180625252858496e-06,
      "loss": 0.99761951,
      "num_input_tokens_seen": 1685200,
      "step": 74,
      "time_per_iteration": 2.854128122329712
    },
    {
      "auxiliary_loss_clip": 0.01866239,
      "auxiliary_loss_mlp": 0.0117165,
      "balance_loss_clip": 1.33517671,
      "balance_loss_mlp": 1.04900789,
      "epoch": 0.009018216797931822,
      "flos": 23075838178560.0,
      "grad_norm": 3.0707094594819253,
      "language_loss": 0.80149853,
      "learning_rate": 3.1277867780021663e-06,
      "loss": 0.83187747,
      "num_input_tokens_seen": 1701835,
      "step": 75,
      "time_per_iteration": 2.7745285034179688
    },
    {
      "auxiliary_loss_clip": 0.01848283,
      "auxiliary_loss_mlp": 0.01177624,
      "balance_loss_clip": 1.3272078,
      "balance_loss_mlp": 1.05717504,
      "epoch": 0.009138459688570914,
      "flos": 15918877284480.0,
      "grad_norm": 1.947709545683033,
      "language_loss": 0.95480663,
      "learning_rate": 3.1373822288779824e-06,
      "loss": 0.98506576,
      "num_input_tokens_seen": 1718415,
      "step": 76,
      "time_per_iteration": 2.8145854473114014
    },
    {
      "auxiliary_loss_clip": 0.01854701,
      "auxiliary_loss_mlp": 0.01177368,
      "balance_loss_clip": 1.33191991,
      "balance_loss_mlp": 1.06521583,
      "epoch": 0.009258702579210003,
      "flos": 27016531372800.0,
      "grad_norm": 1.9834128787942806,
      "language_loss": 0.79539251,
      "learning_rate": 3.1468522454274533e-06,
      "loss": 0.82571316,
      "num_input_tokens_seen": 1738770,
      "step": 77,
      "time_per_iteration": 2.856318235397339
    },
    {
      "auxiliary_loss_clip": 0.01843248,
      "auxiliary_loss_mlp": 0.01169944,
      "balance_loss_clip": 1.32464731,
      "balance_loss_mlp": 1.05187941,
      "epoch": 0.009378945469849095,
      "flos": 26903196984960.0,
      "grad_norm": 2.869358897674171,
      "language_loss": 0.91812158,
      "learning_rate": 3.15620006480197e-06,
      "loss": 0.94825351,
      "num_input_tokens_seen": 1758040,
      "step": 78,
      "time_per_iteration": 2.7537238597869873
    },
    {
      "auxiliary_loss_clip": 0.01844362,
      "auxiliary_loss_mlp": 0.01166852,
      "balance_loss_clip": 1.32906866,
      "balance_loss_mlp": 1.0563693,
      "epoch": 0.009499188360488187,
      "flos": 35694236327040.0,
      "grad_norm": 15.050915885869903,
      "language_loss": 0.74670494,
      "learning_rate": 3.1654288004333087e-06,
      "loss": 0.77681714,
      "num_input_tokens_seen": 1776705,
      "step": 79,
      "time_per_iteration": 2.8749284744262695
    },
    {
      "auxiliary_loss_clip": 0.01805544,
      "auxiliary_loss_mlp": 0.01165912,
      "balance_loss_clip": 1.31124902,
      "balance_loss_mlp": 1.05681217,
      "epoch": 0.009619431251127276,
      "flos": 21503201944320.0,
      "grad_norm": 2.3154383860474352,
      "language_loss": 0.75886965,
      "learning_rate": 3.1745414482589353e-06,
      "loss": 0.78858423,
      "num_input_tokens_seen": 1795915,
      "step": 80,
      "time_per_iteration": 2.6829047203063965
    },
    {
      "auxiliary_loss_clip": 0.01807058,
      "auxiliary_loss_mlp": 0.01162513,
      "balance_loss_clip": 1.3155309,
      "balance_loss_mlp": 1.05770373,
      "epoch": 0.009739674141766368,
      "flos": 17421056991360.0,
      "grad_norm": 2.457638570543815,
      "language_loss": 0.86934531,
      "learning_rate": 3.1835408925606204e-06,
      "loss": 0.89904106,
      "num_input_tokens_seen": 1814055,
      "step": 81,
      "time_per_iteration": 2.681938648223877
    },
    {
      "auxiliary_loss_clip": 0.01789336,
      "auxiliary_loss_mlp": 0.01155422,
      "balance_loss_clip": 1.30708194,
      "balance_loss_mlp": 1.05314064,
      "epoch": 0.00985991703240546,
      "flos": 27527109246720.0,
      "grad_norm": 2.28596350467388,
      "language_loss": 0.8905437,
      "learning_rate": 3.1924299114448214e-06,
      "loss": 0.91999137,
      "num_input_tokens_seen": 1834535,
      "step": 82,
      "time_per_iteration": 2.969802141189575
    },
    {
      "auxiliary_loss_clip": 0.01781989,
      "auxiliary_loss_mlp": 0.01151799,
      "balance_loss_clip": 1.30663323,
      "balance_loss_mlp": 1.05061412,
      "epoch": 0.00998015992304455,
      "flos": 13808084509440.0,
      "grad_norm": 2.1136109592028736,
      "language_loss": 0.83584112,
      "learning_rate": 3.2012111819909055e-06,
      "loss": 0.86517894,
      "num_input_tokens_seen": 1851865,
      "step": 83,
      "time_per_iteration": 2.7572309970855713
    },
    {
      "auxiliary_loss_clip": 0.01776569,
      "auxiliary_loss_mlp": 0.01158084,
      "balance_loss_clip": 1.30134261,
      "balance_loss_mlp": 1.06128633,
      "epoch": 0.010100402813683641,
      "flos": 20191385341440.0,
      "grad_norm": 2.2702170252346567,
      "language_loss": 0.9495101,
      "learning_rate": 3.2098872850910627e-06,
      "loss": 0.97885668,
      "num_input_tokens_seen": 1868540,
      "step": 84,
      "time_per_iteration": 2.8883330821990967
    },
    {
      "auxiliary_loss_clip": 0.01785461,
      "auxiliary_loss_mlp": 0.01163535,
      "balance_loss_clip": 1.30605602,
      "balance_loss_mlp": 1.06778574,
      "epoch": 0.010220645704322733,
      "flos": 17201642762880.0,
      "grad_norm": 2.0581303258922254,
      "language_loss": 0.89431781,
      "learning_rate": 3.2184607100038194e-06,
      "loss": 0.92380774,
      "num_input_tokens_seen": 1887180,
      "step": 85,
      "time_per_iteration": 2.8665943145751953
    },
    {
      "auxiliary_loss_clip": 0.01765957,
      "auxiliary_loss_mlp": 0.01150984,
      "balance_loss_clip": 1.2992996,
      "balance_loss_mlp": 1.06396151,
      "epoch": 0.010340888594961822,
      "flos": 21470415805440.0,
      "grad_norm": 2.0774094340109954,
      "language_loss": 0.9318859,
      "learning_rate": 3.2269338586412414e-06,
      "loss": 0.96105534,
      "num_input_tokens_seen": 1904765,
      "step": 86,
      "time_per_iteration": 2.816697835922241
    },
    {
      "auxiliary_loss_clip": 0.01743224,
      "auxiliary_loss_mlp": 0.0114959,
      "balance_loss_clip": 1.28645968,
      "balance_loss_mlp": 1.05679739,
      "epoch": 0.010461131485600914,
      "flos": 23002831785600.0,
      "grad_norm": 2.572931391505403,
      "language_loss": 0.96277022,
      "learning_rate": 3.2353090496083106e-06,
      "loss": 0.99169838,
      "num_input_tokens_seen": 1922600,
      "step": 87,
      "time_per_iteration": 2.7933943271636963
    },
    {
      "auxiliary_loss_clip": 0.0173763,
      "auxiliary_loss_mlp": 0.0114468,
      "balance_loss_clip": 1.28482556,
      "balance_loss_mlp": 1.05713248,
      "epoch": 0.010581374376240005,
      "flos": 33546850571520.0,
      "grad_norm": 2.0404347768577282,
      "language_loss": 0.81148064,
      "learning_rate": 3.2435885220114572e-06,
      "loss": 0.84030378,
      "num_input_tokens_seen": 1943950,
      "step": 88,
      "time_per_iteration": 2.949205160140991
    },
    {
      "auxiliary_loss_clip": 0.01725969,
      "auxiliary_loss_mlp": 0.01138088,
      "balance_loss_clip": 1.27929282,
      "balance_loss_mlp": 1.05626285,
      "epoch": 0.010701617266879095,
      "flos": 21763087822080.0,
      "grad_norm": 2.0617378156859574,
      "language_loss": 0.93719208,
      "learning_rate": 3.2517744390519113e-06,
      "loss": 0.96583265,
      "num_input_tokens_seen": 1962815,
      "step": 89,
      "time_per_iteration": 2.825634241104126
    },
    {
      "auxiliary_loss_clip": 0.01720225,
      "auxiliary_loss_mlp": 0.01133944,
      "balance_loss_clip": 1.27932894,
      "balance_loss_mlp": 1.05154669,
      "epoch": 0.010821860157518187,
      "flos": 19060199256960.0,
      "grad_norm": 2.0125441191829574,
      "language_loss": 0.74850976,
      "learning_rate": 3.259868891418298e-06,
      "loss": 0.77705145,
      "num_input_tokens_seen": 1980580,
      "step": 90,
      "time_per_iteration": 2.8090717792510986
    },
    {
      "auxiliary_loss_clip": 0.01719674,
      "auxiliary_loss_mlp": 0.01140068,
      "balance_loss_clip": 1.27796435,
      "balance_loss_mlp": 1.0583384,
      "epoch": 0.010942103048157278,
      "flos": 25447378757760.0,
      "grad_norm": 1.982475465063501,
      "language_loss": 0.84916496,
      "learning_rate": 3.2678739004917757e-06,
      "loss": 0.87776244,
      "num_input_tokens_seen": 2000315,
      "step": 91,
      "time_per_iteration": 2.842331647872925
    },
    {
      "auxiliary_loss_clip": 0.01709546,
      "auxiliary_loss_mlp": 0.01129463,
      "balance_loss_clip": 1.27631843,
      "balance_loss_mlp": 1.05417073,
      "epoch": 0.011062345938796368,
      "flos": 27493928058240.0,
      "grad_norm": 1.7759373949911144,
      "language_loss": 0.92175782,
      "learning_rate": 3.275791421376029e-06,
      "loss": 0.95014787,
      "num_input_tokens_seen": 2023760,
      "step": 92,
      "time_per_iteration": 3.946227550506592
    },
    {
      "auxiliary_loss_clip": 0.01698145,
      "auxiliary_loss_mlp": 0.01136776,
      "balance_loss_clip": 1.26935482,
      "balance_loss_mlp": 1.06443989,
      "epoch": 0.01118258882943546,
      "flos": 16071210864000.0,
      "grad_norm": 2.4656475965603013,
      "language_loss": 0.95927751,
      "learning_rate": 3.2836233457634622e-06,
      "loss": 0.98762679,
      "num_input_tokens_seen": 2041895,
      "step": 93,
      "time_per_iteration": 2.8196961879730225
    },
    {
      "auxiliary_loss_clip": 0.01700361,
      "auxiliary_loss_mlp": 0.0113281,
      "balance_loss_clip": 1.26682377,
      "balance_loss_mlp": 1.05937696,
      "epoch": 0.011302831720074551,
      "flos": 20668602458880.0,
      "grad_norm": 2.2017472758015124,
      "language_loss": 0.8547945,
      "learning_rate": 3.2913715046481135e-06,
      "loss": 0.88312614,
      "num_input_tokens_seen": 2061640,
      "step": 94,
      "time_per_iteration": 4.754944324493408
    },
    {
      "auxiliary_loss_clip": 0.01681891,
      "auxiliary_loss_mlp": 0.01115872,
      "balance_loss_clip": 1.26045489,
      "balance_loss_mlp": 1.04506171,
      "epoch": 0.011423074610713641,
      "flos": 13072238490240.0,
      "grad_norm": 2.3088488900392665,
      "language_loss": 0.88925433,
      "learning_rate": 3.299037670895023e-06,
      "loss": 0.91723192,
      "num_input_tokens_seen": 2078255,
      "step": 95,
      "time_per_iteration": 2.8999264240264893
    },
    {
      "auxiliary_loss_clip": 0.01679649,
      "auxiliary_loss_mlp": 0.01116165,
      "balance_loss_clip": 1.25485086,
      "balance_loss_mlp": 1.04945564,
      "epoch": 0.011543317501352733,
      "flos": 30335646689280.0,
      "grad_norm": 2.0562111507023673,
      "language_loss": 0.80218911,
      "learning_rate": 3.3066235616750667e-06,
      "loss": 0.83014727,
      "num_input_tokens_seen": 2099490,
      "step": 96,
      "time_per_iteration": 2.9678542613983154
    },
    {
      "auxiliary_loss_clip": 0.01657413,
      "auxiliary_loss_mlp": 0.01113427,
      "balance_loss_clip": 1.25130641,
      "balance_loss_mlp": 1.04857683,
      "epoch": 0.011663560391991824,
      "flos": 15522962601600.0,
      "grad_norm": 2.352959931248121,
      "language_loss": 0.92383903,
      "learning_rate": 3.3141308407736276e-06,
      "loss": 0.95154744,
      "num_input_tokens_seen": 2116125,
      "step": 97,
      "time_per_iteration": 2.9431073665618896
    },
    {
      "auxiliary_loss_clip": 0.0165958,
      "auxiliary_loss_mlp": 0.01128718,
      "balance_loss_clip": 1.24800241,
      "balance_loss_mlp": 1.06410646,
      "epoch": 0.011783803282630914,
      "flos": 19902125116800.0,
      "grad_norm": 2.014775478477277,
      "language_loss": 0.86724949,
      "learning_rate": 3.321561120780869e-06,
      "loss": 0.89513242,
      "num_input_tokens_seen": 2134835,
      "step": 98,
      "time_per_iteration": 2.887434959411621
    },
    {
      "auxiliary_loss_clip": 0.01650592,
      "auxiliary_loss_mlp": 0.01119005,
      "balance_loss_clip": 1.24637866,
      "balance_loss_mlp": 1.05787432,
      "epoch": 0.011904046173270006,
      "flos": 22340674517760.0,
      "grad_norm": 2.2020641853907517,
      "language_loss": 1.01383877,
      "learning_rate": 3.3289159651708192e-06,
      "loss": 1.04153466,
      "num_input_tokens_seen": 2152410,
      "step": 99,
      "time_per_iteration": 2.9292523860931396
    },
    {
      "auxiliary_loss_clip": 0.01649363,
      "auxiliary_loss_mlp": 0.0111427,
      "balance_loss_clip": 1.24714684,
      "balance_loss_mlp": 1.05361629,
      "epoch": 0.012024289063909096,
      "flos": 19100060375040.0,
      "grad_norm": 2.179403670663143,
      "language_loss": 0.97495639,
      "learning_rate": 3.3361968902759768e-06,
      "loss": 1.00259268,
      "num_input_tokens_seen": 2172090,
      "step": 100,
      "time_per_iteration": 2.917595863342285
    },
    {
      "auxiliary_loss_clip": 0.01628827,
      "auxiliary_loss_mlp": 0.01114397,
      "balance_loss_clip": 1.23518527,
      "balance_loss_mlp": 1.05488825,
      "epoch": 0.012144531954548187,
      "flos": 15012205159680.0,
      "grad_norm": 2.3609805264819257,
      "language_loss": 0.93646479,
      "learning_rate": 3.343405367163663e-06,
      "loss": 0.96389699,
      "num_input_tokens_seen": 2189020,
      "step": 101,
      "time_per_iteration": 3.0340323448181152
    },
    {
      "auxiliary_loss_clip": 0.01618673,
      "auxiliary_loss_mlp": 0.01107611,
      "balance_loss_clip": 1.23096919,
      "balance_loss_mlp": 1.05379951,
      "epoch": 0.012264774845187279,
      "flos": 15122020014720.0,
      "grad_norm": 3.844212241035829,
      "language_loss": 0.80961734,
      "learning_rate": 3.350542823419951e-06,
      "loss": 0.83688015,
      "num_input_tokens_seen": 2205620,
      "step": 102,
      "time_per_iteration": 2.863964080810547
    },
    {
      "auxiliary_loss_clip": 0.0161837,
      "auxiliary_loss_mlp": 0.01118917,
      "balance_loss_clip": 1.23053753,
      "balance_loss_mlp": 1.06331778,
      "epoch": 0.012385017735826368,
      "flos": 13949248959360.0,
      "grad_norm": 3.3718106306784406,
      "language_loss": 0.87442303,
      "learning_rate": 3.3576106448465615e-06,
      "loss": 0.90179592,
      "num_input_tokens_seen": 2219000,
      "step": 103,
      "time_per_iteration": 3.0117757320404053
    },
    {
      "auxiliary_loss_clip": 0.0161988,
      "auxiliary_loss_mlp": 0.01118772,
      "balance_loss_clip": 1.2264924,
      "balance_loss_mlp": 1.06312466,
      "epoch": 0.01250526062646546,
      "flos": 23623260428160.0,
      "grad_norm": 2.1873757980323822,
      "language_loss": 0.8826037,
      "learning_rate": 3.3646101770757797e-06,
      "loss": 0.90999019,
      "num_input_tokens_seen": 2237790,
      "step": 104,
      "time_per_iteration": 3.094442367553711
    },
    {
      "auxiliary_loss_clip": 0.01607943,
      "auxiliary_loss_mlp": 0.01109079,
      "balance_loss_clip": 1.22631788,
      "balance_loss_mlp": 1.05631709,
      "epoch": 0.012625503517104552,
      "flos": 34640078958720.0,
      "grad_norm": 1.9907310622882193,
      "language_loss": 0.85675466,
      "learning_rate": 3.371542727108104e-06,
      "loss": 0.88392484,
      "num_input_tokens_seen": 2259965,
      "step": 105,
      "time_per_iteration": 3.135592222213745
    },
    {
      "auxiliary_loss_clip": 0.01609238,
      "auxiliary_loss_mlp": 0.01112593,
      "balance_loss_clip": 1.22191572,
      "balance_loss_mlp": 1.05992675,
      "epoch": 0.012745746407743641,
      "flos": 17821891837440.0,
      "grad_norm": 8.5359428410745,
      "language_loss": 0.89892548,
      "learning_rate": 3.3784095647770114e-06,
      "loss": 0.92614377,
      "num_input_tokens_seen": 2278610,
      "step": 106,
      "time_per_iteration": 2.9435110092163086
    },
    {
      "auxiliary_loss_clip": 0.01597149,
      "auxiliary_loss_mlp": 0.01111246,
      "balance_loss_clip": 1.21515703,
      "balance_loss_mlp": 1.05619562,
      "epoch": 0.012865989298382733,
      "flos": 20595057361920.0,
      "grad_norm": 1.9533495183186578,
      "language_loss": 0.88561028,
      "learning_rate": 3.3852119241449547e-06,
      "loss": 0.91269422,
      "num_input_tokens_seen": 2297730,
      "step": 107,
      "time_per_iteration": 2.9650747776031494
    },
    {
      "auxiliary_loss_clip": 0.01591744,
      "auxiliary_loss_mlp": 0.01107736,
      "balance_loss_clip": 1.21159911,
      "balance_loss_mlp": 1.06110132,
      "epoch": 0.012986232189021825,
      "flos": 23948969978880.0,
      "grad_norm": 3.471173636601977,
      "language_loss": 0.96278369,
      "learning_rate": 3.3919510048344295e-06,
      "loss": 0.98977852,
      "num_input_tokens_seen": 2315740,
      "step": 108,
      "time_per_iteration": 2.9390201568603516
    },
    {
      "auxiliary_loss_clip": 0.01576125,
      "auxiliary_loss_mlp": 0.01103272,
      "balance_loss_clip": 1.20501542,
      "balance_loss_mlp": 1.05315685,
      "epoch": 0.013106475079660914,
      "flos": 23725425686400.0,
      "grad_norm": 3.083095720731453,
      "language_loss": 0.86733073,
      "learning_rate": 3.3986279732976907e-06,
      "loss": 0.89412475,
      "num_input_tokens_seen": 2334215,
      "step": 109,
      "time_per_iteration": 2.8686563968658447
    },
    {
      "auxiliary_loss_clip": 0.0157439,
      "auxiliary_loss_mlp": 0.01078175,
      "balance_loss_clip": 1.20360136,
      "balance_loss_mlp": 1.03320932,
      "epoch": 0.013226717970300006,
      "flos": 21102438925440.0,
      "grad_norm": 2.0767055122512605,
      "language_loss": 0.95307249,
      "learning_rate": 3.4052439640284983e-06,
      "loss": 0.97959816,
      "num_input_tokens_seen": 2353130,
      "step": 110,
      "time_per_iteration": 2.7682225704193115
    },
    {
      "auxiliary_loss_clip": 0.01566163,
      "auxiliary_loss_mlp": 0.01090116,
      "balance_loss_clip": 1.20040119,
      "balance_loss_mlp": 1.04743886,
      "epoch": 0.013346960860939098,
      "flos": 24863902231680.0,
      "grad_norm": 1.8833394320244652,
      "language_loss": 0.81225985,
      "learning_rate": 3.4118000807190217e-06,
      "loss": 0.8388226,
      "num_input_tokens_seen": 2374010,
      "step": 111,
      "time_per_iteration": 2.8246755599975586
    },
    {
      "auxiliary_loss_clip": 0.01569366,
      "auxiliary_loss_mlp": 0.01106493,
      "balance_loss_clip": 1.19975197,
      "balance_loss_mlp": 1.05921423,
      "epoch": 0.013467203751578187,
      "flos": 28181940140160.0,
      "grad_norm": 1.6462364361892563,
      "language_loss": 0.76067609,
      "learning_rate": 3.4182973973648723e-06,
      "loss": 0.78743464,
      "num_input_tokens_seen": 2395220,
      "step": 112,
      "time_per_iteration": 2.7980284690856934
    },
    {
      "auxiliary_loss_clip": 0.01558801,
      "auxiliary_loss_mlp": 0.01104865,
      "balance_loss_clip": 1.19457853,
      "balance_loss_mlp": 1.05863571,
      "epoch": 0.013587446642217279,
      "flos": 18916233546240.0,
      "grad_norm": 3.394787604953982,
      "language_loss": 0.94863075,
      "learning_rate": 3.424736959321014e-06,
      "loss": 0.97526741,
      "num_input_tokens_seen": 2413025,
      "step": 113,
      "time_per_iteration": 2.8256382942199707
    },
    {
      "auxiliary_loss_clip": 0.01559872,
      "auxiliary_loss_mlp": 0.01098674,
      "balance_loss_clip": 1.19667745,
      "balance_loss_mlp": 1.05499601,
      "epoch": 0.01370768953285637,
      "flos": 23988615615360.0,
      "grad_norm": 1.901973391649418,
      "language_loss": 0.88773286,
      "learning_rate": 3.431119784311155e-06,
      "loss": 0.91431832,
      "num_input_tokens_seen": 2432700,
      "step": 114,
      "time_per_iteration": 2.8701696395874023
    },
    {
      "auxiliary_loss_clip": 0.01545032,
      "auxiliary_loss_mlp": 0.01079572,
      "balance_loss_clip": 1.18943,
      "balance_loss_mlp": 1.04044747,
      "epoch": 0.01382793242349546,
      "flos": 39202565512320.0,
      "grad_norm": 2.1527019786214994,
      "language_loss": 0.77658367,
      "learning_rate": 3.43744686339307e-06,
      "loss": 0.80282974,
      "num_input_tokens_seen": 2455020,
      "step": 115,
      "time_per_iteration": 2.93011212348938
    },
    {
      "auxiliary_loss_clip": 0.01547355,
      "auxiliary_loss_mlp": 0.01082829,
      "balance_loss_clip": 1.18514788,
      "balance_loss_mlp": 1.0420835,
      "epoch": 0.013948175314134552,
      "flos": 41353506714240.0,
      "grad_norm": 2.155507109342344,
      "language_loss": 0.90846664,
      "learning_rate": 3.44371916188212e-06,
      "loss": 0.9347685,
      "num_input_tokens_seen": 2475775,
      "step": 116,
      "time_per_iteration": 3.031212329864502
    },
    {
      "auxiliary_loss_clip": 0.01540892,
      "auxiliary_loss_mlp": 0.01091971,
      "balance_loss_clip": 1.18246925,
      "balance_loss_mlp": 1.05020022,
      "epoch": 0.014068418204773643,
      "flos": 22453542028800.0,
      "grad_norm": 2.515884576721947,
      "language_loss": 0.86120546,
      "learning_rate": 3.449937620235143e-06,
      "loss": 0.88753408,
      "num_input_tokens_seen": 2496370,
      "step": 117,
      "time_per_iteration": 2.75055193901062
    },
    {
      "auxiliary_loss_clip": 0.01532086,
      "auxiliary_loss_mlp": 0.01097726,
      "balance_loss_clip": 1.18155766,
      "balance_loss_mlp": 1.05829155,
      "epoch": 0.014188661095412733,
      "flos": 23805147922560.0,
      "grad_norm": 1.6152434685571957,
      "language_loss": 0.89283013,
      "learning_rate": 3.456103154896722e-06,
      "loss": 0.91912818,
      "num_input_tokens_seen": 2517645,
      "step": 118,
      "time_per_iteration": 3.817178726196289
    },
    {
      "auxiliary_loss_clip": 0.01523611,
      "auxiliary_loss_mlp": 0.01097407,
      "balance_loss_clip": 1.17816901,
      "balance_loss_mlp": 1.05787742,
      "epoch": 0.014308903986051825,
      "flos": 23660248458240.0,
      "grad_norm": 2.0804984762725893,
      "language_loss": 0.92290831,
      "learning_rate": 3.462216659109757e-06,
      "loss": 0.94911849,
      "num_input_tokens_seen": 2537825,
      "step": 119,
      "time_per_iteration": 3.8003664016723633
    },
    {
      "auxiliary_loss_clip": 0.01525992,
      "auxiliary_loss_mlp": 0.01088293,
      "balance_loss_clip": 1.17881882,
      "balance_loss_mlp": 1.05195808,
      "epoch": 0.014429146876690916,
      "flos": 20667991927680.0,
      "grad_norm": 2.7768114635749925,
      "language_loss": 0.85283077,
      "learning_rate": 3.4682790036921077e-06,
      "loss": 0.8789736,
      "num_input_tokens_seen": 2556485,
      "step": 120,
      "time_per_iteration": 3.7602312564849854
    },
    {
      "auxiliary_loss_clip": 0.01509532,
      "auxiliary_loss_mlp": 0.01087468,
      "balance_loss_clip": 1.16937089,
      "balance_loss_mlp": 1.05232525,
      "epoch": 0.014549389767330006,
      "flos": 20229199384320.0,
      "grad_norm": 2.126694838565762,
      "language_loss": 0.83170217,
      "learning_rate": 3.4742910377810193e-06,
      "loss": 0.85767221,
      "num_input_tokens_seen": 2573945,
      "step": 121,
      "time_per_iteration": 2.8198482990264893
    },
    {
      "auxiliary_loss_clip": 0.01513981,
      "auxiliary_loss_mlp": 0.01075418,
      "balance_loss_clip": 1.17171133,
      "balance_loss_mlp": 1.03891611,
      "epoch": 0.014669632657969098,
      "flos": 18004174381440.0,
      "grad_norm": 2.345661585606919,
      "language_loss": 0.88761133,
      "learning_rate": 3.4802535895469042e-06,
      "loss": 0.91350532,
      "num_input_tokens_seen": 2592695,
      "step": 122,
      "time_per_iteration": 2.8230888843536377
    },
    {
      "auxiliary_loss_clip": 0.01510159,
      "auxiliary_loss_mlp": 0.01084426,
      "balance_loss_clip": 1.16765797,
      "balance_loss_mlp": 1.04744697,
      "epoch": 0.01478987554860819,
      "flos": 22741796672640.0,
      "grad_norm": 2.5498199937514014,
      "language_loss": 0.89572775,
      "learning_rate": 3.4861674668779934e-06,
      "loss": 0.9216736,
      "num_input_tokens_seen": 2610925,
      "step": 123,
      "time_per_iteration": 2.7821788787841797
    },
    {
      "auxiliary_loss_clip": 0.01500901,
      "auxiliary_loss_mlp": 0.01069206,
      "balance_loss_clip": 1.16391468,
      "balance_loss_mlp": 1.03506422,
      "epoch": 0.01491011843924728,
      "flos": 17198590106880.0,
      "grad_norm": 1.9648674555996732,
      "language_loss": 0.84062564,
      "learning_rate": 3.492033458037272e-06,
      "loss": 0.86632669,
      "num_input_tokens_seen": 2629495,
      "step": 124,
      "time_per_iteration": 3.1374223232269287
    },
    {
      "auxiliary_loss_clip": 0.0150294,
      "auxiliary_loss_mlp": 0.01071043,
      "balance_loss_clip": 1.16474628,
      "balance_loss_mlp": 1.03673458,
      "epoch": 0.01503036132988637,
      "flos": 17673867889920.0,
      "grad_norm": 3.3861179220679785,
      "language_loss": 0.8692565,
      "learning_rate": 3.497852332293018e-06,
      "loss": 0.89499629,
      "num_input_tokens_seen": 2645070,
      "step": 125,
      "time_per_iteration": 2.884185552597046
    },
    {
      "auxiliary_loss_clip": 0.01501221,
      "auxiliary_loss_mlp": 0.0107896,
      "balance_loss_clip": 1.16382384,
      "balance_loss_mlp": 1.04803681,
      "epoch": 0.015150604220525462,
      "flos": 18878239935360.0,
      "grad_norm": 1.9272317732018271,
      "language_loss": 0.96497774,
      "learning_rate": 3.5036248405242356e-06,
      "loss": 0.99077952,
      "num_input_tokens_seen": 2663825,
      "step": 126,
      "time_per_iteration": 2.827526092529297
    },
    {
      "auxiliary_loss_clip": 0.01501179,
      "auxiliary_loss_mlp": 0.01081524,
      "balance_loss_clip": 1.16387129,
      "balance_loss_mlp": 1.04778731,
      "epoch": 0.015270847111164552,
      "flos": 39420184060800.0,
      "grad_norm": 1.9059687342879121,
      "language_loss": 0.82785332,
      "learning_rate": 3.509351715802146e-06,
      "loss": 0.85368031,
      "num_input_tokens_seen": 2684710,
      "step": 127,
      "time_per_iteration": 2.9764604568481445
    },
    {
      "auxiliary_loss_clip": 0.01488506,
      "auxiliary_loss_mlp": 0.01086115,
      "balance_loss_clip": 1.1569519,
      "balance_loss_mlp": 1.0524025,
      "epoch": 0.015391090001803644,
      "flos": 43762466286720.0,
      "grad_norm": 2.1243832330950214,
      "language_loss": 0.78334767,
      "learning_rate": 3.5150336739488763e-06,
      "loss": 0.80909389,
      "num_input_tokens_seen": 2706995,
      "step": 128,
      "time_per_iteration": 3.010080099105835
    },
    {
      "auxiliary_loss_clip": 0.01489312,
      "auxiliary_loss_mlp": 0.01069716,
      "balance_loss_clip": 1.16090679,
      "balance_loss_mlp": 1.041273,
      "epoch": 0.015511332892442733,
      "flos": 18916341287040.0,
      "grad_norm": 2.281662691310887,
      "language_loss": 0.83838475,
      "learning_rate": 3.5206714140744143e-06,
      "loss": 0.86397499,
      "num_input_tokens_seen": 2727050,
      "step": 129,
      "time_per_iteration": 2.8404948711395264
    },
    {
      "auxiliary_loss_clip": 0.01489672,
      "auxiliary_loss_mlp": 0.01081931,
      "balance_loss_clip": 1.16307259,
      "balance_loss_mlp": 1.0519855,
      "epoch": 0.015631575783081827,
      "flos": 24535283679360.0,
      "grad_norm": 4.512470047939525,
      "language_loss": 0.87657356,
      "learning_rate": 3.5262656190928208e-06,
      "loss": 0.90228963,
      "num_input_tokens_seen": 2745350,
      "step": 130,
      "time_per_iteration": 2.8411309719085693
    },
    {
      "auxiliary_loss_clip": 0.01405221,
      "auxiliary_loss_mlp": 0.01096591,
      "balance_loss_clip": 1.17936254,
      "balance_loss_mlp": 1.06588256,
      "epoch": 0.015751818673720917,
      "flos": 62328536098560.0,
      "grad_norm": 1.0470112473331754,
      "language_loss": 0.71513855,
      "learning_rate": 3.5318169562186737e-06,
      "loss": 0.74015665,
      "num_input_tokens_seen": 2814195,
      "step": 131,
      "time_per_iteration": 3.3645803928375244
    },
    {
      "auxiliary_loss_clip": 0.01480737,
      "auxiliary_loss_mlp": 0.01077277,
      "balance_loss_clip": 1.15788054,
      "balance_loss_mlp": 1.05138469,
      "epoch": 0.015872061564360006,
      "flos": 23878549365120.0,
      "grad_norm": 7.354892763364748,
      "language_loss": 0.81998622,
      "learning_rate": 3.5373260774446292e-06,
      "loss": 0.84556633,
      "num_input_tokens_seen": 2834645,
      "step": 132,
      "time_per_iteration": 2.8588204383850098
    },
    {
      "auxiliary_loss_clip": 0.01472199,
      "auxiliary_loss_mlp": 0.01067898,
      "balance_loss_clip": 1.15078843,
      "balance_loss_mlp": 1.04170752,
      "epoch": 0.0159923044549991,
      "flos": 23367899664000.0,
      "grad_norm": 1.9327874635236526,
      "language_loss": 0.90098774,
      "learning_rate": 3.542793620000961e-06,
      "loss": 0.92638874,
      "num_input_tokens_seen": 2854120,
      "step": 133,
      "time_per_iteration": 2.8278417587280273
    },
    {
      "auxiliary_loss_clip": 0.01470067,
      "auxiliary_loss_mlp": 0.01085602,
      "balance_loss_clip": 1.15295649,
      "balance_loss_mlp": 1.05700421,
      "epoch": 0.01611254734563819,
      "flos": 17858305249920.0,
      "grad_norm": 3.230339817463091,
      "language_loss": 0.86880374,
      "learning_rate": 3.5482202067978894e-06,
      "loss": 0.89436042,
      "num_input_tokens_seen": 2871330,
      "step": 134,
      "time_per_iteration": 2.8220551013946533
    },
    {
      "auxiliary_loss_clip": 0.01470042,
      "auxiliary_loss_mlp": 0.01076111,
      "balance_loss_clip": 1.15019131,
      "balance_loss_mlp": 1.05086291,
      "epoch": 0.01623279023627728,
      "flos": 20954774113920.0,
      "grad_norm": 1.941718955997415,
      "language_loss": 0.76157939,
      "learning_rate": 3.553606446851471e-06,
      "loss": 0.78704095,
      "num_input_tokens_seen": 2888070,
      "step": 135,
      "time_per_iteration": 2.7709388732910156
    },
    {
      "auxiliary_loss_clip": 0.01474535,
      "auxiliary_loss_mlp": 0.01073724,
      "balance_loss_clip": 1.15253222,
      "balance_loss_mlp": 1.04929805,
      "epoch": 0.016353033126916373,
      "flos": 15742412743680.0,
      "grad_norm": 2.1694246647825075,
      "language_loss": 0.83607149,
      "learning_rate": 3.5589529356937613e-06,
      "loss": 0.86155409,
      "num_input_tokens_seen": 2906465,
      "step": 136,
      "time_per_iteration": 2.808544397354126
    },
    {
      "auxiliary_loss_clip": 0.01465771,
      "auxiliary_loss_mlp": 0.01083064,
      "balance_loss_clip": 1.14579844,
      "balance_loss_mlp": 1.05627775,
      "epoch": 0.016473276017555463,
      "flos": 18807280617600.0,
      "grad_norm": 2.0041320258243296,
      "language_loss": 0.77109492,
      "learning_rate": 3.5642602557679627e-06,
      "loss": 0.79658329,
      "num_input_tokens_seen": 2924915,
      "step": 137,
      "time_per_iteration": 2.855867862701416
    },
    {
      "auxiliary_loss_clip": 0.0146943,
      "auxiliary_loss_mlp": 0.01081235,
      "balance_loss_clip": 1.15673423,
      "balance_loss_mlp": 1.05791807,
      "epoch": 0.016593518908194552,
      "flos": 24352641999360.0,
      "grad_norm": 2.2122183361332857,
      "language_loss": 0.84191215,
      "learning_rate": 3.569528976809202e-06,
      "loss": 0.86741883,
      "num_input_tokens_seen": 2942130,
      "step": 138,
      "time_per_iteration": 2.8571856021881104
    },
    {
      "auxiliary_loss_clip": 0.01461878,
      "auxiliary_loss_mlp": 0.01069839,
      "balance_loss_clip": 1.14551651,
      "balance_loss_mlp": 1.04577065,
      "epoch": 0.016713761798833646,
      "flos": 22346133384960.0,
      "grad_norm": 1.6381033341548359,
      "language_loss": 0.89973754,
      "learning_rate": 3.5747596562115522e-06,
      "loss": 0.92505473,
      "num_input_tokens_seen": 2962745,
      "step": 139,
      "time_per_iteration": 2.8138585090637207
    },
    {
      "auxiliary_loss_clip": 0.01465311,
      "auxiliary_loss_mlp": 0.01072614,
      "balance_loss_clip": 1.14912438,
      "balance_loss_mlp": 1.04620934,
      "epoch": 0.016834004689472735,
      "flos": 17821820010240.0,
      "grad_norm": 2.7676590601189366,
      "language_loss": 0.90954685,
      "learning_rate": 3.5799528393819138e-06,
      "loss": 0.93492603,
      "num_input_tokens_seen": 2981825,
      "step": 140,
      "time_per_iteration": 2.9601049423217773
    },
    {
      "auxiliary_loss_clip": 0.01458671,
      "auxiliary_loss_mlp": 0.0105828,
      "balance_loss_clip": 1.14633071,
      "balance_loss_mlp": 1.03489137,
      "epoch": 0.016954247580111825,
      "flos": 20519501103360.0,
      "grad_norm": 1.910920404567093,
      "language_loss": 0.8799212,
      "learning_rate": 3.585109060081286e-06,
      "loss": 0.90509075,
      "num_input_tokens_seen": 3001625,
      "step": 141,
      "time_per_iteration": 2.7530276775360107
    },
    {
      "auxiliary_loss_clip": 0.01452697,
      "auxiliary_loss_mlp": 0.01071222,
      "balance_loss_clip": 1.140885,
      "balance_loss_mlp": 1.0459137,
      "epoch": 0.017074490470750915,
      "flos": 22088869200000.0,
      "grad_norm": 1.924673396026974,
      "language_loss": 0.78502619,
      "learning_rate": 3.590228840753992e-06,
      "loss": 0.81026542,
      "num_input_tokens_seen": 3022055,
      "step": 142,
      "time_per_iteration": 2.771705389022827
    },
    {
      "auxiliary_loss_clip": 0.01450587,
      "auxiliary_loss_mlp": 0.01068869,
      "balance_loss_clip": 1.14321744,
      "balance_loss_mlp": 1.04685104,
      "epoch": 0.01719473336139001,
      "flos": 15997270717440.0,
      "grad_norm": 2.4113029719914483,
      "language_loss": 0.87520206,
      "learning_rate": 3.5953126928453423e-06,
      "loss": 0.90039665,
      "num_input_tokens_seen": 3039605,
      "step": 143,
      "time_per_iteration": 2.724475860595703
    },
    {
      "auxiliary_loss_clip": 0.01450361,
      "auxiliary_loss_mlp": 0.01078619,
      "balance_loss_clip": 1.14039803,
      "balance_loss_mlp": 1.05749464,
      "epoch": 0.017314976252029098,
      "flos": 22492038430080.0,
      "grad_norm": 2.1893314532138235,
      "language_loss": 0.80578887,
      "learning_rate": 3.600361117108239e-06,
      "loss": 0.83107865,
      "num_input_tokens_seen": 3059405,
      "step": 144,
      "time_per_iteration": 3.7564456462860107
    },
    {
      "auxiliary_loss_clip": 0.01446321,
      "auxiliary_loss_mlp": 0.01072348,
      "balance_loss_clip": 1.13807535,
      "balance_loss_mlp": 1.05041409,
      "epoch": 0.017435219142668188,
      "flos": 22018053536640.0,
      "grad_norm": 2.8018449633652587,
      "language_loss": 0.9730289,
      "learning_rate": 3.6053746038991616e-06,
      "loss": 0.99821568,
      "num_input_tokens_seen": 3078490,
      "step": 145,
      "time_per_iteration": 2.800995111465454
    },
    {
      "auxiliary_loss_clip": 0.01354881,
      "auxiliary_loss_mlp": 0.01027184,
      "balance_loss_clip": 1.14543271,
      "balance_loss_mlp": 1.00725198,
      "epoch": 0.01755546203330728,
      "flos": 72240526149120.0,
      "grad_norm": 1.0524961718906605,
      "language_loss": 0.58465552,
      "learning_rate": 3.6103536334639843e-06,
      "loss": 0.60847616,
      "num_input_tokens_seen": 3131755,
      "step": 146,
      "time_per_iteration": 4.970015287399292
    },
    {
      "auxiliary_loss_clip": 0.01437244,
      "auxiliary_loss_mlp": 0.01066177,
      "balance_loss_clip": 1.13468313,
      "balance_loss_mlp": 1.04539835,
      "epoch": 0.01767570492394637,
      "flos": 25337061112320.0,
      "grad_norm": 2.1058019384327302,
      "language_loss": 0.85418963,
      "learning_rate": 3.615298676214041e-06,
      "loss": 0.87922388,
      "num_input_tokens_seen": 3152035,
      "step": 147,
      "time_per_iteration": 3.699542284011841
    },
    {
      "auxiliary_loss_clip": 0.01440453,
      "auxiliary_loss_mlp": 0.01070402,
      "balance_loss_clip": 1.13720393,
      "balance_loss_mlp": 1.04976666,
      "epoch": 0.01779594781458546,
      "flos": 20449188230400.0,
      "grad_norm": 2.046749943377627,
      "language_loss": 0.89099574,
      "learning_rate": 3.6202101929928317e-06,
      "loss": 0.91610432,
      "num_input_tokens_seen": 3170625,
      "step": 148,
      "time_per_iteration": 2.800687313079834
    },
    {
      "auxiliary_loss_clip": 0.01441264,
      "auxiliary_loss_mlp": 0.01064125,
      "balance_loss_clip": 1.13609171,
      "balance_loss_mlp": 1.04478979,
      "epoch": 0.017916190705224554,
      "flos": 16253601148800.0,
      "grad_norm": 2.1072054355984555,
      "language_loss": 0.88344449,
      "learning_rate": 3.6250886353337413e-06,
      "loss": 0.90849835,
      "num_input_tokens_seen": 3188155,
      "step": 149,
      "time_per_iteration": 2.749575614929199
    },
    {
      "auxiliary_loss_clip": 0.01437429,
      "auxiliary_loss_mlp": 0.0105909,
      "balance_loss_clip": 1.135849,
      "balance_loss_mlp": 1.03925323,
      "epoch": 0.018036433595863644,
      "flos": 23330588411520.0,
      "grad_norm": 1.9737595196182178,
      "language_loss": 0.86308873,
      "learning_rate": 3.6299344457091488e-06,
      "loss": 0.88805389,
      "num_input_tokens_seen": 3209015,
      "step": 150,
      "time_per_iteration": 2.790248394012451
    },
    {
      "auxiliary_loss_clip": 0.01439488,
      "auxiliary_loss_mlp": 0.01078192,
      "balance_loss_clip": 1.13671196,
      "balance_loss_mlp": 1.05781925,
      "epoch": 0.018156676486502734,
      "flos": 18588010043520.0,
      "grad_norm": 2.238573141532112,
      "language_loss": 0.94059765,
      "learning_rate": 3.634748057771256e-06,
      "loss": 0.96577442,
      "num_input_tokens_seen": 3224955,
      "step": 151,
      "time_per_iteration": 2.7550957202911377
    },
    {
      "auxiliary_loss_clip": 0.01427599,
      "auxiliary_loss_mlp": 0.0105804,
      "balance_loss_clip": 1.1266582,
      "balance_loss_mlp": 1.03970575,
      "epoch": 0.018276919377141827,
      "flos": 25448707560960.0,
      "grad_norm": 1.5915521445320684,
      "language_loss": 0.85643578,
      "learning_rate": 3.639529896584965e-06,
      "loss": 0.88129216,
      "num_input_tokens_seen": 3246330,
      "step": 152,
      "time_per_iteration": 2.820107936859131
    },
    {
      "auxiliary_loss_clip": 0.01430669,
      "auxiliary_loss_mlp": 0.01059875,
      "balance_loss_clip": 1.13322699,
      "balance_loss_mlp": 1.0415405,
      "epoch": 0.018397162267780917,
      "flos": 20047311889920.0,
      "grad_norm": 2.769260137492558,
      "language_loss": 0.88744026,
      "learning_rate": 3.6442803788531233e-06,
      "loss": 0.91234577,
      "num_input_tokens_seen": 3264290,
      "step": 153,
      "time_per_iteration": 2.7013626098632812
    },
    {
      "auxiliary_loss_clip": 0.01431323,
      "auxiliary_loss_mlp": 0.01061977,
      "balance_loss_clip": 1.1296134,
      "balance_loss_mlp": 1.04326081,
      "epoch": 0.018517405158420007,
      "flos": 27565282425600.0,
      "grad_norm": 3.258863247413859,
      "language_loss": 0.95886743,
      "learning_rate": 3.6489999131344357e-06,
      "loss": 0.98380041,
      "num_input_tokens_seen": 3287065,
      "step": 154,
      "time_per_iteration": 2.830514907836914
    },
    {
      "auxiliary_loss_clip": 0.0142769,
      "auxiliary_loss_mlp": 0.01065261,
      "balance_loss_clip": 1.134233,
      "balance_loss_mlp": 1.04895306,
      "epoch": 0.0186376480490591,
      "flos": 19354056422400.0,
      "grad_norm": 1.9841251633011139,
      "language_loss": 0.907013,
      "learning_rate": 3.653688900054313e-06,
      "loss": 0.93194246,
      "num_input_tokens_seen": 3305595,
      "step": 155,
      "time_per_iteration": 2.7411725521087646
    },
    {
      "auxiliary_loss_clip": 0.01423214,
      "auxiliary_loss_mlp": 0.01060383,
      "balance_loss_clip": 1.12524366,
      "balance_loss_mlp": 1.04120255,
      "epoch": 0.01875789093969819,
      "flos": 26687840993280.0,
      "grad_norm": 2.690586605004093,
      "language_loss": 0.75845289,
      "learning_rate": 3.6583477325089526e-06,
      "loss": 0.78328884,
      "num_input_tokens_seen": 3326135,
      "step": 156,
      "time_per_iteration": 2.8148012161254883
    },
    {
      "auxiliary_loss_clip": 0.01421321,
      "auxiliary_loss_mlp": 0.01059233,
      "balance_loss_clip": 1.12792981,
      "balance_loss_mlp": 1.04157853,
      "epoch": 0.01887813383033728,
      "flos": 24353001135360.0,
      "grad_norm": 2.352809953029022,
      "language_loss": 1.04187012,
      "learning_rate": 3.6629767958628916e-06,
      "loss": 1.06667566,
      "num_input_tokens_seen": 3343510,
      "step": 157,
      "time_per_iteration": 2.8264880180358887
    },
    {
      "auxiliary_loss_clip": 0.01421375,
      "auxiliary_loss_mlp": 0.01059967,
      "balance_loss_clip": 1.1273632,
      "balance_loss_mlp": 1.04277694,
      "epoch": 0.018998376720976373,
      "flos": 14647532330880.0,
      "grad_norm": 2.1713064446657246,
      "language_loss": 0.854738,
      "learning_rate": 3.667576468140291e-06,
      "loss": 0.87955135,
      "num_input_tokens_seen": 3361325,
      "step": 158,
      "time_per_iteration": 2.7602081298828125
    },
    {
      "auxiliary_loss_clip": 0.01414862,
      "auxiliary_loss_mlp": 0.01051776,
      "balance_loss_clip": 1.12403047,
      "balance_loss_mlp": 1.03544462,
      "epoch": 0.019118619611615463,
      "flos": 29305261146240.0,
      "grad_norm": 2.6060873271372436,
      "language_loss": 0.88996333,
      "learning_rate": 3.672147120210184e-06,
      "loss": 0.9146297,
      "num_input_tokens_seen": 3377925,
      "step": 159,
      "time_per_iteration": 2.8499960899353027
    },
    {
      "auxiliary_loss_clip": 0.01412016,
      "auxiliary_loss_mlp": 0.01060039,
      "balance_loss_clip": 1.12487054,
      "balance_loss_mlp": 1.04268193,
      "epoch": 0.019238862502254553,
      "flos": 20886723797760.0,
      "grad_norm": 1.9559202045059678,
      "language_loss": 0.86516941,
      "learning_rate": 3.6766891159659177e-06,
      "loss": 0.88988996,
      "num_input_tokens_seen": 3396335,
      "step": 160,
      "time_per_iteration": 2.767106533050537
    },
    {
      "auxiliary_loss_clip": 0.01411421,
      "auxiliary_loss_mlp": 0.01061497,
      "balance_loss_clip": 1.12399995,
      "balance_loss_mlp": 1.04452109,
      "epoch": 0.019359105392893646,
      "flos": 21360672777600.0,
      "grad_norm": 2.81894709315836,
      "language_loss": 0.87793481,
      "learning_rate": 3.6812028124990075e-06,
      "loss": 0.90266395,
      "num_input_tokens_seen": 3413605,
      "step": 161,
      "time_per_iteration": 2.7439303398132324
    },
    {
      "auxiliary_loss_clip": 0.01411097,
      "auxiliary_loss_mlp": 0.01052106,
      "balance_loss_clip": 1.1228044,
      "balance_loss_mlp": 1.03483212,
      "epoch": 0.019479348283532736,
      "flos": 16283729681280.0,
      "grad_norm": 2.8074200011211827,
      "language_loss": 0.81778455,
      "learning_rate": 3.6856885602676016e-06,
      "loss": 0.84241652,
      "num_input_tokens_seen": 3429640,
      "step": 162,
      "time_per_iteration": 2.6884756088256836
    },
    {
      "auxiliary_loss_clip": 0.01413507,
      "auxiliary_loss_mlp": 0.01050099,
      "balance_loss_clip": 1.12342739,
      "balance_loss_mlp": 1.03429198,
      "epoch": 0.019599591174171826,
      "flos": 22091239497600.0,
      "grad_norm": 2.2076073146283215,
      "language_loss": 0.94065458,
      "learning_rate": 3.6901467032597733e-06,
      "loss": 0.96529067,
      "num_input_tokens_seen": 3448125,
      "step": 163,
      "time_per_iteration": 2.842823028564453
    },
    {
      "auxiliary_loss_clip": 0.01410813,
      "auxiliary_loss_mlp": 0.01052435,
      "balance_loss_clip": 1.12379646,
      "balance_loss_mlp": 1.03503036,
      "epoch": 0.01971983406481092,
      "flos": 19609668581760.0,
      "grad_norm": 2.1869274119184716,
      "language_loss": 0.87030613,
      "learning_rate": 3.694577579151804e-06,
      "loss": 0.89493859,
      "num_input_tokens_seen": 3466535,
      "step": 164,
      "time_per_iteration": 2.736877202987671
    },
    {
      "auxiliary_loss_clip": 0.01406862,
      "auxiliary_loss_mlp": 0.0104722,
      "balance_loss_clip": 1.12166619,
      "balance_loss_mlp": 1.03223562,
      "epoch": 0.01984007695545001,
      "flos": 19099342103040.0,
      "grad_norm": 2.4404651327452744,
      "language_loss": 0.73719752,
      "learning_rate": 3.6989815194616703e-06,
      "loss": 0.76173836,
      "num_input_tokens_seen": 3483730,
      "step": 165,
      "time_per_iteration": 2.7721731662750244
    },
    {
      "auxiliary_loss_clip": 0.01408646,
      "auxiliary_loss_mlp": 0.01055466,
      "balance_loss_clip": 1.11953568,
      "balance_loss_mlp": 1.03801417,
      "epoch": 0.0199603198460891,
      "flos": 20848406964480.0,
      "grad_norm": 4.857659726631675,
      "language_loss": 0.8002311,
      "learning_rate": 3.703358849697888e-06,
      "loss": 0.82487226,
      "num_input_tokens_seen": 3503640,
      "step": 166,
      "time_per_iteration": 2.7634975910186768
    },
    {
      "auxiliary_loss_clip": 0.01406838,
      "auxiliary_loss_mlp": 0.01052504,
      "balance_loss_clip": 1.12337053,
      "balance_loss_mlp": 1.03816307,
      "epoch": 0.020080562736728192,
      "flos": 21870747861120.0,
      "grad_norm": 2.1875892594009785,
      "language_loss": 0.82706326,
      "learning_rate": 3.7077098895038803e-06,
      "loss": 0.85165668,
      "num_input_tokens_seen": 3523010,
      "step": 167,
      "time_per_iteration": 2.749976873397827
    },
    {
      "auxiliary_loss_clip": 0.01406452,
      "auxiliary_loss_mlp": 0.01048675,
      "balance_loss_clip": 1.12160575,
      "balance_loss_mlp": 1.03306484,
      "epoch": 0.020200805627367282,
      "flos": 21688788539520.0,
      "grad_norm": 2.4741749559642185,
      "language_loss": 0.96855152,
      "learning_rate": 3.712034952798045e-06,
      "loss": 0.99310279,
      "num_input_tokens_seen": 3541125,
      "step": 168,
      "time_per_iteration": 2.7482552528381348
    },
    {
      "auxiliary_loss_clip": 0.01406778,
      "auxiliary_loss_mlp": 0.01057899,
      "balance_loss_clip": 1.12008226,
      "balance_loss_mlp": 1.03841996,
      "epoch": 0.02032104851800637,
      "flos": 33543043729920.0,
      "grad_norm": 3.279659660358365,
      "language_loss": 0.84822339,
      "learning_rate": 3.7163343479096656e-06,
      "loss": 0.87287021,
      "num_input_tokens_seen": 3562700,
      "step": 169,
      "time_per_iteration": 2.939336061477661
    },
    {
      "auxiliary_loss_clip": 0.01401406,
      "auxiliary_loss_mlp": 0.01058633,
      "balance_loss_clip": 1.12103868,
      "balance_loss_mlp": 1.0429213,
      "epoch": 0.020441291408645465,
      "flos": 31686965274240.0,
      "grad_norm": 2.339435141336895,
      "language_loss": 0.83053195,
      "learning_rate": 3.720608377710802e-06,
      "loss": 0.85513234,
      "num_input_tokens_seen": 3582790,
      "step": 170,
      "time_per_iteration": 2.875774383544922
    },
    {
      "auxiliary_loss_clip": 0.0139467,
      "auxiliary_loss_mlp": 0.01046045,
      "balance_loss_clip": 1.11312723,
      "balance_loss_mlp": 1.02862847,
      "epoch": 0.020561534299284555,
      "flos": 20886687884160.0,
      "grad_norm": 2.4586957232903157,
      "language_loss": 0.86451703,
      "learning_rate": 3.7248573397443277e-06,
      "loss": 0.88892424,
      "num_input_tokens_seen": 3601715,
      "step": 171,
      "time_per_iteration": 3.7218985557556152
    },
    {
      "auxiliary_loss_clip": 0.01399487,
      "auxiliary_loss_mlp": 0.01068393,
      "balance_loss_clip": 1.11956608,
      "balance_loss_mlp": 1.05225229,
      "epoch": 0.020681777189923645,
      "flos": 20996610480000.0,
      "grad_norm": 2.3911632665998077,
      "language_loss": 0.97546595,
      "learning_rate": 3.729081526348224e-06,
      "loss": 1.00014472,
      "num_input_tokens_seen": 3620245,
      "step": 172,
      "time_per_iteration": 4.51881217956543
    },
    {
      "auxiliary_loss_clip": 0.01397606,
      "auxiliary_loss_mlp": 0.01056997,
      "balance_loss_clip": 1.11667407,
      "balance_loss_mlp": 1.03977156,
      "epoch": 0.020802020080562738,
      "flos": 28257532312320.0,
      "grad_norm": 2.3935862268171713,
      "language_loss": 0.84945768,
      "learning_rate": 3.7332812247762777e-06,
      "loss": 0.87400365,
      "num_input_tokens_seen": 3641545,
      "step": 173,
      "time_per_iteration": 3.668985605239868
    },
    {
      "auxiliary_loss_clip": 0.01398223,
      "auxiliary_loss_mlp": 0.01062681,
      "balance_loss_clip": 1.11892092,
      "balance_loss_mlp": 1.04707646,
      "epoch": 0.020922262971201828,
      "flos": 19681274344320.0,
      "grad_norm": 2.677966740313008,
      "language_loss": 0.9557116,
      "learning_rate": 3.737456717315293e-06,
      "loss": 0.98032069,
      "num_input_tokens_seen": 3660510,
      "step": 174,
      "time_per_iteration": 2.7798876762390137
    },
    {
      "auxiliary_loss_clip": 0.01389692,
      "auxiliary_loss_mlp": 0.01061838,
      "balance_loss_clip": 1.11586213,
      "balance_loss_mlp": 1.04701412,
      "epoch": 0.021042505861840918,
      "flos": 15666353694720.0,
      "grad_norm": 2.584713260868514,
      "language_loss": 0.90972525,
      "learning_rate": 3.7416082813989552e-06,
      "loss": 0.93424058,
      "num_input_tokens_seen": 3677505,
      "step": 175,
      "time_per_iteration": 2.7605347633361816
    },
    {
      "auxiliary_loss_clip": 0.01396952,
      "auxiliary_loss_mlp": 0.0105995,
      "balance_loss_clip": 1.11871052,
      "balance_loss_mlp": 1.04347563,
      "epoch": 0.02116274875248001,
      "flos": 21142012734720.0,
      "grad_norm": 2.2019067971469384,
      "language_loss": 0.89728659,
      "learning_rate": 3.745736189718439e-06,
      "loss": 0.92185569,
      "num_input_tokens_seen": 3696760,
      "step": 176,
      "time_per_iteration": 2.8059604167938232
    },
    {
      "auxiliary_loss_clip": 0.01386524,
      "auxiliary_loss_mlp": 0.01057998,
      "balance_loss_clip": 1.11310744,
      "balance_loss_mlp": 1.04185688,
      "epoch": 0.0212829916431191,
      "flos": 24715770543360.0,
      "grad_norm": 5.083658055104306,
      "language_loss": 0.72672617,
      "learning_rate": 3.749840710329894e-06,
      "loss": 0.75117135,
      "num_input_tokens_seen": 3717465,
      "step": 177,
      "time_per_iteration": 2.905709743499756
    },
    {
      "auxiliary_loss_clip": 0.01400805,
      "auxiliary_loss_mlp": 0.01062535,
      "balance_loss_clip": 1.1187886,
      "balance_loss_mlp": 1.04449844,
      "epoch": 0.02140323453375819,
      "flos": 16645493508480.0,
      "grad_norm": 3.2270729314001505,
      "language_loss": 0.98196453,
      "learning_rate": 3.7539221067588938e-06,
      "loss": 1.006598,
      "num_input_tokens_seen": 3731440,
      "step": 178,
      "time_per_iteration": 2.9785890579223633
    },
    {
      "auxiliary_loss_clip": 0.01393212,
      "auxiliary_loss_mlp": 0.0106083,
      "balance_loss_clip": 1.11594725,
      "balance_loss_mlp": 1.0431633,
      "epoch": 0.021523477424397284,
      "flos": 20299332689280.0,
      "grad_norm": 3.0130927826453724,
      "language_loss": 0.9399097,
      "learning_rate": 3.757980638101964e-06,
      "loss": 0.96445012,
      "num_input_tokens_seen": 3744935,
      "step": 179,
      "time_per_iteration": 2.7836740016937256
    },
    {
      "auxiliary_loss_clip": 0.01396677,
      "auxiliary_loss_mlp": 0.01055619,
      "balance_loss_clip": 1.11702013,
      "balance_loss_mlp": 1.03856063,
      "epoch": 0.021643720315036374,
      "flos": 26104005331200.0,
      "grad_norm": 2.3515788374465996,
      "language_loss": 0.89665902,
      "learning_rate": 3.7620165591252806e-06,
      "loss": 0.92118204,
      "num_input_tokens_seen": 3763035,
      "step": 180,
      "time_per_iteration": 2.808241605758667
    },
    {
      "auxiliary_loss_clip": 0.01390258,
      "auxiliary_loss_mlp": 0.01049499,
      "balance_loss_clip": 1.11372483,
      "balance_loss_mlp": 1.03439546,
      "epoch": 0.021763963205675464,
      "flos": 24787663614720.0,
      "grad_norm": 2.061116092992959,
      "language_loss": 0.94357014,
      "learning_rate": 3.766030120360636e-06,
      "loss": 0.96796769,
      "num_input_tokens_seen": 3782665,
      "step": 181,
      "time_per_iteration": 2.8364601135253906
    },
    {
      "auxiliary_loss_clip": 0.01383954,
      "auxiliary_loss_mlp": 0.01059256,
      "balance_loss_clip": 1.11025405,
      "balance_loss_mlp": 1.04348516,
      "epoch": 0.021884206096314557,
      "flos": 25813559957760.0,
      "grad_norm": 2.1774312005120016,
      "language_loss": 0.90386599,
      "learning_rate": 3.7700215681987578e-06,
      "loss": 0.92829812,
      "num_input_tokens_seen": 3802435,
      "step": 182,
      "time_per_iteration": 2.8307816982269287
    },
    {
      "auxiliary_loss_clip": 0.01388904,
      "auxiliary_loss_mlp": 0.01062942,
      "balance_loss_clip": 1.11349559,
      "balance_loss_mlp": 1.04580021,
      "epoch": 0.022004448986953647,
      "flos": 20082719721600.0,
      "grad_norm": 2.106698207036957,
      "language_loss": 0.82442737,
      "learning_rate": 3.7739911449800767e-06,
      "loss": 0.84894586,
      "num_input_tokens_seen": 3822490,
      "step": 183,
      "time_per_iteration": 2.771064519882202
    },
    {
      "auxiliary_loss_clip": 0.01385641,
      "auxiliary_loss_mlp": 0.01063684,
      "balance_loss_clip": 1.11253738,
      "balance_loss_mlp": 1.04848528,
      "epoch": 0.022124691877592736,
      "flos": 20480609652480.0,
      "grad_norm": 10.01806930537304,
      "language_loss": 0.80700278,
      "learning_rate": 3.7779390890830114e-06,
      "loss": 0.831496,
      "num_input_tokens_seen": 3841140,
      "step": 184,
      "time_per_iteration": 2.750239610671997
    },
    {
      "auxiliary_loss_clip": 0.01387648,
      "auxiliary_loss_mlp": 0.01059842,
      "balance_loss_clip": 1.11061227,
      "balance_loss_mlp": 1.04320073,
      "epoch": 0.02224493476823183,
      "flos": 23586847015680.0,
      "grad_norm": 3.0381593078692735,
      "language_loss": 0.85912728,
      "learning_rate": 3.7818656350098723e-06,
      "loss": 0.88360214,
      "num_input_tokens_seen": 3862090,
      "step": 185,
      "time_per_iteration": 2.874722957611084
    },
    {
      "auxiliary_loss_clip": 0.01391617,
      "auxiliary_loss_mlp": 0.01054138,
      "balance_loss_clip": 1.11451161,
      "balance_loss_mlp": 1.03827143,
      "epoch": 0.02236517765887092,
      "flos": 16909940413440.0,
      "grad_norm": 2.6183044213235505,
      "language_loss": 0.77457958,
      "learning_rate": 3.7857710134704447e-06,
      "loss": 0.7990371,
      "num_input_tokens_seen": 3881025,
      "step": 186,
      "time_per_iteration": 2.7487235069274902
    },
    {
      "auxiliary_loss_clip": 0.01381016,
      "auxiliary_loss_mlp": 0.01063469,
      "balance_loss_clip": 1.11160409,
      "balance_loss_mlp": 1.04805529,
      "epoch": 0.02248542054951001,
      "flos": 43508182930560.0,
      "grad_norm": 2.3155219040959443,
      "language_loss": 0.79675031,
      "learning_rate": 3.7896554514633234e-06,
      "loss": 0.82119519,
      "num_input_tokens_seen": 3905310,
      "step": 187,
      "time_per_iteration": 2.924100637435913
    },
    {
      "auxiliary_loss_clip": 0.01384638,
      "auxiliary_loss_mlp": 0.01053662,
      "balance_loss_clip": 1.11265886,
      "balance_loss_mlp": 1.03777134,
      "epoch": 0.022605663440149103,
      "flos": 23367648268800.0,
      "grad_norm": 2.030597847857301,
      "language_loss": 0.84090066,
      "learning_rate": 3.7935191723550955e-06,
      "loss": 0.86528367,
      "num_input_tokens_seen": 3924265,
      "step": 188,
      "time_per_iteration": 2.7614407539367676
    },
    {
      "auxiliary_loss_clip": 0.01380448,
      "auxiliary_loss_mlp": 0.01061818,
      "balance_loss_clip": 1.1108582,
      "balance_loss_mlp": 1.04704189,
      "epoch": 0.022725906330788193,
      "flos": 29019915504000.0,
      "grad_norm": 2.4174044874421847,
      "language_loss": 0.88843203,
      "learning_rate": 3.797362395957408e-06,
      "loss": 0.91285467,
      "num_input_tokens_seen": 3944830,
      "step": 189,
      "time_per_iteration": 2.827993154525757
    },
    {
      "auxiliary_loss_clip": 0.01385462,
      "auxiliary_loss_mlp": 0.01066293,
      "balance_loss_clip": 1.11246538,
      "balance_loss_mlp": 1.04990768,
      "epoch": 0.022846149221427282,
      "flos": 24496176746880.0,
      "grad_norm": 2.820205565799056,
      "language_loss": 0.7828964,
      "learning_rate": 3.8011853386020055e-06,
      "loss": 0.80741394,
      "num_input_tokens_seen": 3965735,
      "step": 190,
      "time_per_iteration": 2.816617488861084
    },
    {
      "auxiliary_loss_clip": 0.01377878,
      "auxiliary_loss_mlp": 0.01055378,
      "balance_loss_clip": 1.10769558,
      "balance_loss_mlp": 1.0389626,
      "epoch": 0.022966392112066376,
      "flos": 15523537219200.0,
      "grad_norm": 4.205736982046081,
      "language_loss": 0.89230353,
      "learning_rate": 3.804988213213804e-06,
      "loss": 0.91663611,
      "num_input_tokens_seen": 3983975,
      "step": 191,
      "time_per_iteration": 2.7268142700195312
    },
    {
      "auxiliary_loss_clip": 0.01332765,
      "auxiliary_loss_mlp": 0.010204,
      "balance_loss_clip": 1.13905621,
      "balance_loss_mlp": 1.009004,
      "epoch": 0.023086635002705466,
      "flos": 55650408433920.0,
      "grad_norm": 1.0193065220079842,
      "language_loss": 0.63183057,
      "learning_rate": 3.808771229382049e-06,
      "loss": 0.65536225,
      "num_input_tokens_seen": 4043440,
      "step": 192,
      "time_per_iteration": 3.234945058822632
    },
    {
      "auxiliary_loss_clip": 0.01376066,
      "auxiliary_loss_mlp": 0.01059947,
      "balance_loss_clip": 1.10488367,
      "balance_loss_mlp": 1.04489684,
      "epoch": 0.023206877893344555,
      "flos": 19313441118720.0,
      "grad_norm": 2.2923417103502497,
      "language_loss": 0.84590209,
      "learning_rate": 3.8125345934296324e-06,
      "loss": 0.87026227,
      "num_input_tokens_seen": 4061750,
      "step": 193,
      "time_per_iteration": 2.8023629188537598
    },
    {
      "auxiliary_loss_clip": 0.01384401,
      "auxiliary_loss_mlp": 0.01062732,
      "balance_loss_clip": 1.11157298,
      "balance_loss_mlp": 1.04557776,
      "epoch": 0.02332712078398365,
      "flos": 23072965090560.0,
      "grad_norm": 2.6432770401468173,
      "language_loss": 0.87767673,
      "learning_rate": 3.81627850848061e-06,
      "loss": 0.90214807,
      "num_input_tokens_seen": 4082345,
      "step": 194,
      "time_per_iteration": 2.8351337909698486
    },
    {
      "auxiliary_loss_clip": 0.01375835,
      "auxiliary_loss_mlp": 0.01057839,
      "balance_loss_clip": 1.10816979,
      "balance_loss_mlp": 1.0424614,
      "epoch": 0.02344736367462274,
      "flos": 24425971614720.0,
      "grad_norm": 2.4317739756573546,
      "language_loss": 0.86197197,
      "learning_rate": 3.820003174525994e-06,
      "loss": 0.88630867,
      "num_input_tokens_seen": 4101770,
      "step": 195,
      "time_per_iteration": 2.7413034439086914
    },
    {
      "auxiliary_loss_clip": 0.01376602,
      "auxiliary_loss_mlp": 0.01067832,
      "balance_loss_clip": 1.10962224,
      "balance_loss_mlp": 1.05214441,
      "epoch": 0.02356760656526183,
      "flos": 21579799697280.0,
      "grad_norm": 2.2088585876215014,
      "language_loss": 0.82884306,
      "learning_rate": 3.823708788487851e-06,
      "loss": 0.85328746,
      "num_input_tokens_seen": 4118770,
      "step": 196,
      "time_per_iteration": 2.7688775062561035
    },
    {
      "auxiliary_loss_clip": 0.01374153,
      "auxiliary_loss_mlp": 0.0105394,
      "balance_loss_clip": 1.10866654,
      "balance_loss_mlp": 1.03936028,
      "epoch": 0.02368784945590092,
      "flos": 25193598192000.0,
      "grad_norm": 2.2775931692604146,
      "language_loss": 0.84494966,
      "learning_rate": 3.827395544281781e-06,
      "loss": 0.86923057,
      "num_input_tokens_seen": 4141110,
      "step": 197,
      "time_per_iteration": 3.7828962802886963
    },
    {
      "auxiliary_loss_clip": 0.013715,
      "auxiliary_loss_mlp": 0.01062923,
      "balance_loss_clip": 1.10606265,
      "balance_loss_mlp": 1.04691291,
      "epoch": 0.02380809234654001,
      "flos": 27562481164800.0,
      "grad_norm": 2.1097556387909506,
      "language_loss": 0.79050046,
      "learning_rate": 3.831063632877802e-06,
      "loss": 0.81484467,
      "num_input_tokens_seen": 4161430,
      "step": 198,
      "time_per_iteration": 3.693183660507202
    },
    {
      "auxiliary_loss_clip": 0.01372649,
      "auxiliary_loss_mlp": 0.01057332,
      "balance_loss_clip": 1.1128931,
      "balance_loss_mlp": 1.04405248,
      "epoch": 0.0239283352371791,
      "flos": 18259786540800.0,
      "grad_norm": 2.7748710098979927,
      "language_loss": 0.75861305,
      "learning_rate": 3.834713242359712e-06,
      "loss": 0.78291291,
      "num_input_tokens_seen": 4179260,
      "step": 199,
      "time_per_iteration": 4.531643390655518
    },
    {
      "auxiliary_loss_clip": 0.01374863,
      "auxiliary_loss_mlp": 0.01052781,
      "balance_loss_clip": 1.1081984,
      "balance_loss_mlp": 1.03691423,
      "epoch": 0.02404857812781819,
      "flos": 21395110942080.0,
      "grad_norm": 2.1194168769978687,
      "language_loss": 0.87429619,
      "learning_rate": 3.838344557982959e-06,
      "loss": 0.89857262,
      "num_input_tokens_seen": 4200640,
      "step": 200,
      "time_per_iteration": 2.777866840362549
    },
    {
      "auxiliary_loss_clip": 0.01367756,
      "auxiliary_loss_mlp": 0.01052818,
      "balance_loss_clip": 1.10328841,
      "balance_loss_mlp": 1.03816676,
      "epoch": 0.024168821018457284,
      "flos": 16654256426880.0,
      "grad_norm": 3.0054558149711457,
      "language_loss": 0.84460413,
      "learning_rate": 3.841957762231063e-06,
      "loss": 0.86880988,
      "num_input_tokens_seen": 4218170,
      "step": 201,
      "time_per_iteration": 2.683396339416504
    },
    {
      "auxiliary_loss_clip": 0.01368141,
      "auxiliary_loss_mlp": 0.01059586,
      "balance_loss_clip": 1.10404384,
      "balance_loss_mlp": 1.04553103,
      "epoch": 0.024289063909096374,
      "flos": 22820872464000.0,
      "grad_norm": 2.3682594129790293,
      "language_loss": 0.87701333,
      "learning_rate": 3.8455530348706454e-06,
      "loss": 0.9012906,
      "num_input_tokens_seen": 4237770,
      "step": 202,
      "time_per_iteration": 2.7738115787506104
    },
    {
      "auxiliary_loss_clip": 0.01371453,
      "auxiliary_loss_mlp": 0.01053554,
      "balance_loss_clip": 1.10883713,
      "balance_loss_mlp": 1.03942215,
      "epoch": 0.024409306799735464,
      "flos": 17748598135680.0,
      "grad_norm": 2.221813409638455,
      "language_loss": 0.77522755,
      "learning_rate": 3.849130553005099e-06,
      "loss": 0.79947764,
      "num_input_tokens_seen": 4255985,
      "step": 203,
      "time_per_iteration": 2.752946615219116
    },
    {
      "auxiliary_loss_clip": 0.01365549,
      "auxiliary_loss_mlp": 0.01049086,
      "balance_loss_clip": 1.10420012,
      "balance_loss_mlp": 1.03451836,
      "epoch": 0.024529549690374557,
      "flos": 21616213109760.0,
      "grad_norm": 1.870271482861934,
      "language_loss": 0.83964068,
      "learning_rate": 3.852690491126933e-06,
      "loss": 0.86378706,
      "num_input_tokens_seen": 4276035,
      "step": 204,
      "time_per_iteration": 2.697665214538574
    },
    {
      "auxiliary_loss_clip": 0.01370377,
      "auxiliary_loss_mlp": 0.01060337,
      "balance_loss_clip": 1.10610855,
      "balance_loss_mlp": 1.04486394,
      "epoch": 0.024649792581013647,
      "flos": 25551662918400.0,
      "grad_norm": 4.4223544878559276,
      "language_loss": 0.91339481,
      "learning_rate": 3.856233021168845e-06,
      "loss": 0.93770194,
      "num_input_tokens_seen": 4295730,
      "step": 205,
      "time_per_iteration": 2.8299307823181152
    },
    {
      "auxiliary_loss_clip": 0.01360187,
      "auxiliary_loss_mlp": 0.01061447,
      "balance_loss_clip": 1.10306025,
      "balance_loss_mlp": 1.04707015,
      "epoch": 0.024770035471652737,
      "flos": 34495574544000.0,
      "grad_norm": 2.3332945254978066,
      "language_loss": 0.91160154,
      "learning_rate": 3.859758312553544e-06,
      "loss": 0.9358179,
      "num_input_tokens_seen": 4317950,
      "step": 206,
      "time_per_iteration": 2.800576686859131
    },
    {
      "auxiliary_loss_clip": 0.01368615,
      "auxiliary_loss_mlp": 0.01053305,
      "balance_loss_clip": 1.10833526,
      "balance_loss_mlp": 1.03942871,
      "epoch": 0.02489027836229183,
      "flos": 21505428587520.0,
      "grad_norm": 2.0325659393015827,
      "language_loss": 0.91972327,
      "learning_rate": 3.8632665322423735e-06,
      "loss": 0.94394255,
      "num_input_tokens_seen": 4337605,
      "step": 207,
      "time_per_iteration": 2.7743022441864014
    },
    {
      "auxiliary_loss_clip": 0.01360453,
      "auxiliary_loss_mlp": 0.01053911,
      "balance_loss_clip": 1.103719,
      "balance_loss_mlp": 1.04061317,
      "epoch": 0.02501052125293092,
      "flos": 23219013790080.0,
      "grad_norm": 2.257265616818119,
      "language_loss": 0.85757107,
      "learning_rate": 3.866757844782762e-06,
      "loss": 0.8817147,
      "num_input_tokens_seen": 4358110,
      "step": 208,
      "time_per_iteration": 2.7585670948028564
    },
    {
      "auxiliary_loss_clip": 0.01364872,
      "auxiliary_loss_mlp": 0.01052165,
      "balance_loss_clip": 1.10511982,
      "balance_loss_mlp": 1.03812778,
      "epoch": 0.02513076414357001,
      "flos": 26388920010240.0,
      "grad_norm": 3.00746894247213,
      "language_loss": 0.91636527,
      "learning_rate": 3.870232412354527e-06,
      "loss": 0.94053555,
      "num_input_tokens_seen": 4374955,
      "step": 209,
      "time_per_iteration": 2.7799816131591797
    },
    {
      "auxiliary_loss_clip": 0.01362747,
      "auxiliary_loss_mlp": 0.01050501,
      "balance_loss_clip": 1.10177195,
      "balance_loss_mlp": 1.03611851,
      "epoch": 0.025251007034209103,
      "flos": 13590430047360.0,
      "grad_norm": 2.361205924004267,
      "language_loss": 0.92321646,
      "learning_rate": 3.873690394815086e-06,
      "loss": 0.94734895,
      "num_input_tokens_seen": 4391535,
      "step": 210,
      "time_per_iteration": 2.6885297298431396
    },
    {
      "auxiliary_loss_clip": 0.01362161,
      "auxiliary_loss_mlp": 0.01056806,
      "balance_loss_clip": 1.10361063,
      "balance_loss_mlp": 1.04296613,
      "epoch": 0.025371249924848193,
      "flos": 15049229103360.0,
      "grad_norm": 2.7122818308156673,
      "language_loss": 0.91295362,
      "learning_rate": 3.877131949743587e-06,
      "loss": 0.93714333,
      "num_input_tokens_seen": 4408400,
      "step": 211,
      "time_per_iteration": 2.734980344772339
    },
    {
      "auxiliary_loss_clip": 0.0136257,
      "auxiliary_loss_mlp": 0.01058233,
      "balance_loss_clip": 1.10355306,
      "balance_loss_mlp": 1.04359412,
      "epoch": 0.025491492815487283,
      "flos": 25553853648000.0,
      "grad_norm": 1.9471941622752686,
      "language_loss": 0.77953184,
      "learning_rate": 3.880557232483993e-06,
      "loss": 0.80373985,
      "num_input_tokens_seen": 4427840,
      "step": 212,
      "time_per_iteration": 2.8809571266174316
    },
    {
      "auxiliary_loss_clip": 0.01358599,
      "auxiliary_loss_mlp": 0.01059711,
      "balance_loss_clip": 1.09979248,
      "balance_loss_mlp": 1.0455606,
      "epoch": 0.025611735706126376,
      "flos": 20630752502400.0,
      "grad_norm": 2.058100426221792,
      "language_loss": 0.86874056,
      "learning_rate": 3.883966396187164e-06,
      "loss": 0.89292371,
      "num_input_tokens_seen": 4447110,
      "step": 213,
      "time_per_iteration": 2.7649528980255127
    },
    {
      "auxiliary_loss_clip": 0.01361408,
      "auxiliary_loss_mlp": 0.01056279,
      "balance_loss_clip": 1.10471392,
      "balance_loss_mlp": 1.04271269,
      "epoch": 0.025731978596765466,
      "flos": 19062282245760.0,
      "grad_norm": 2.06182492731019,
      "language_loss": 0.9002409,
      "learning_rate": 3.887359591851937e-06,
      "loss": 0.92441773,
      "num_input_tokens_seen": 4464715,
      "step": 214,
      "time_per_iteration": 2.7615795135498047
    },
    {
      "auxiliary_loss_clip": 0.01359557,
      "auxiliary_loss_mlp": 0.01059355,
      "balance_loss_clip": 1.10376525,
      "balance_loss_mlp": 1.04664099,
      "epoch": 0.025852221487404556,
      "flos": 22163814927360.0,
      "grad_norm": 2.2948977226564065,
      "language_loss": 0.92339492,
      "learning_rate": 3.890736968365265e-06,
      "loss": 0.94758397,
      "num_input_tokens_seen": 4485030,
      "step": 215,
      "time_per_iteration": 2.831876754760742
    },
    {
      "auxiliary_loss_clip": 0.01356212,
      "auxiliary_loss_mlp": 0.01051982,
      "balance_loss_clip": 1.10108805,
      "balance_loss_mlp": 1.03895843,
      "epoch": 0.02597246437804365,
      "flos": 26541971861760.0,
      "grad_norm": 1.9002254306811344,
      "language_loss": 0.84894013,
      "learning_rate": 3.894098672541412e-06,
      "loss": 0.87302208,
      "num_input_tokens_seen": 4505935,
      "step": 216,
      "time_per_iteration": 2.769833564758301
    },
    {
      "auxiliary_loss_clip": 0.01360296,
      "auxiliary_loss_mlp": 0.01043609,
      "balance_loss_clip": 1.10322666,
      "balance_loss_mlp": 1.02880335,
      "epoch": 0.02609270726868274,
      "flos": 32671671696000.0,
      "grad_norm": 2.983336770809252,
      "language_loss": 0.7539891,
      "learning_rate": 3.89744484916025e-06,
      "loss": 0.77802813,
      "num_input_tokens_seen": 4527045,
      "step": 217,
      "time_per_iteration": 2.8319971561431885
    },
    {
      "auxiliary_loss_clip": 0.01357317,
      "auxiliary_loss_mlp": 0.0105667,
      "balance_loss_clip": 1.1006856,
      "balance_loss_mlp": 1.04200745,
      "epoch": 0.02621295015932183,
      "flos": 26243553669120.0,
      "grad_norm": 2.0334399927727214,
      "language_loss": 0.87341565,
      "learning_rate": 3.900775641004673e-06,
      "loss": 0.89755559,
      "num_input_tokens_seen": 4546360,
      "step": 218,
      "time_per_iteration": 2.747948408126831
    },
    {
      "auxiliary_loss_clip": 0.01364569,
      "auxiliary_loss_mlp": 0.01058721,
      "balance_loss_clip": 1.10413694,
      "balance_loss_mlp": 1.04421937,
      "epoch": 0.026333193049960922,
      "flos": 42921402353280.0,
      "grad_norm": 2.819072398591503,
      "language_loss": 0.73976684,
      "learning_rate": 3.904091188897156e-06,
      "loss": 0.7639997,
      "num_input_tokens_seen": 4565495,
      "step": 219,
      "time_per_iteration": 2.894177198410034
    },
    {
      "auxiliary_loss_clip": 0.01360033,
      "auxiliary_loss_mlp": 0.01051442,
      "balance_loss_clip": 1.10461283,
      "balance_loss_mlp": 1.03708887,
      "epoch": 0.026453435940600012,
      "flos": 17963846386560.0,
      "grad_norm": 2.0996588794038673,
      "language_loss": 0.8175354,
      "learning_rate": 3.90739163173548e-06,
      "loss": 0.84165013,
      "num_input_tokens_seen": 4583330,
      "step": 220,
      "time_per_iteration": 2.7752740383148193
    },
    {
      "auxiliary_loss_clip": 0.01357105,
      "auxiliary_loss_mlp": 0.01045676,
      "balance_loss_clip": 1.10279036,
      "balance_loss_mlp": 1.03192544,
      "epoch": 0.026573678831239102,
      "flos": 18984319776000.0,
      "grad_norm": 2.3464482690171358,
      "language_loss": 0.88298255,
      "learning_rate": 3.910677106527646e-06,
      "loss": 0.90701038,
      "num_input_tokens_seen": 4600520,
      "step": 221,
      "time_per_iteration": 2.787811040878296
    },
    {
      "auxiliary_loss_clip": 0.01351518,
      "auxiliary_loss_mlp": 0.01053075,
      "balance_loss_clip": 1.09929323,
      "balance_loss_mlp": 1.03961086,
      "epoch": 0.026693921721878195,
      "flos": 29241448634880.0,
      "grad_norm": 3.253100233402835,
      "language_loss": 0.8435421,
      "learning_rate": 3.913947748426004e-06,
      "loss": 0.86758804,
      "num_input_tokens_seen": 4617340,
      "step": 222,
      "time_per_iteration": 2.812692165374756
    },
    {
      "auxiliary_loss_clip": 0.0135301,
      "auxiliary_loss_mlp": 0.01053136,
      "balance_loss_clip": 1.10340643,
      "balance_loss_mlp": 1.03986788,
      "epoch": 0.026814164612517285,
      "flos": 14128083797760.0,
      "grad_norm": 2.9640195959748294,
      "language_loss": 0.76391083,
      "learning_rate": 3.9172036907606136e-06,
      "loss": 0.78797233,
      "num_input_tokens_seen": 4630820,
      "step": 223,
      "time_per_iteration": 3.6649186611175537
    },
    {
      "auxiliary_loss_clip": 0.01351957,
      "auxiliary_loss_mlp": 0.01050038,
      "balance_loss_clip": 1.0982511,
      "balance_loss_mlp": 1.03568554,
      "epoch": 0.026934407503156375,
      "flos": 23511973115520.0,
      "grad_norm": 1.751343768477567,
      "language_loss": 0.94717634,
      "learning_rate": 3.920445065071855e-06,
      "loss": 0.97119629,
      "num_input_tokens_seen": 4651985,
      "step": 224,
      "time_per_iteration": 2.7335052490234375
    },
    {
      "auxiliary_loss_clip": 0.01360616,
      "auxiliary_loss_mlp": 0.01050029,
      "balance_loss_clip": 1.10528016,
      "balance_loss_mlp": 1.0361886,
      "epoch": 0.027054650393795468,
      "flos": 28950356816640.0,
      "grad_norm": 2.30077936299829,
      "language_loss": 0.79769337,
      "learning_rate": 3.923672001142322e-06,
      "loss": 0.82179981,
      "num_input_tokens_seen": 4672295,
      "step": 225,
      "time_per_iteration": 4.62237024307251
    },
    {
      "auxiliary_loss_clip": 0.01352292,
      "auxiliary_loss_mlp": 0.0105772,
      "balance_loss_clip": 1.10008276,
      "balance_loss_mlp": 1.04387426,
      "epoch": 0.027174893284434558,
      "flos": 31431568596480.0,
      "grad_norm": 1.9024021131716646,
      "language_loss": 0.84345257,
      "learning_rate": 3.926884627027996e-06,
      "loss": 0.86755276,
      "num_input_tokens_seen": 4696065,
      "step": 226,
      "time_per_iteration": 2.83705997467041
    },
    {
      "auxiliary_loss_clip": 0.01352244,
      "auxiliary_loss_mlp": 0.01053726,
      "balance_loss_clip": 1.09980464,
      "balance_loss_mlp": 1.04023778,
      "epoch": 0.027295136175073648,
      "flos": 22054466949120.0,
      "grad_norm": 2.9542252370744757,
      "language_loss": 0.77466595,
      "learning_rate": 3.930083069088744e-06,
      "loss": 0.79872561,
      "num_input_tokens_seen": 4716065,
      "step": 227,
      "time_per_iteration": 3.6264679431915283
    },
    {
      "auxiliary_loss_clip": 0.01306465,
      "auxiliary_loss_mlp": 0.01013097,
      "balance_loss_clip": 1.1239388,
      "balance_loss_mlp": 1.00470495,
      "epoch": 0.02741537906571274,
      "flos": 60800752972800.0,
      "grad_norm": 0.9902577423682047,
      "language_loss": 0.5929786,
      "learning_rate": 3.933267452018137e-06,
      "loss": 0.61617422,
      "num_input_tokens_seen": 4775860,
      "step": 228,
      "time_per_iteration": 3.2040202617645264
    },
    {
      "auxiliary_loss_clip": 0.01350931,
      "auxiliary_loss_mlp": 0.01051655,
      "balance_loss_clip": 1.10382247,
      "balance_loss_mlp": 1.03811252,
      "epoch": 0.02753562195635183,
      "flos": 24606278910720.0,
      "grad_norm": 2.2743680103587933,
      "language_loss": 0.84586841,
      "learning_rate": 3.936437898872622e-06,
      "loss": 0.86989427,
      "num_input_tokens_seen": 4795835,
      "step": 229,
      "time_per_iteration": 2.7399940490722656
    },
    {
      "auxiliary_loss_clip": 0.0134869,
      "auxiliary_loss_mlp": 0.01045317,
      "balance_loss_clip": 1.09896278,
      "balance_loss_mlp": 1.03119087,
      "epoch": 0.02765586484699092,
      "flos": 34094236907520.0,
      "grad_norm": 2.3971266980341017,
      "language_loss": 0.79809111,
      "learning_rate": 3.9395945311000525e-06,
      "loss": 0.8220312,
      "num_input_tokens_seen": 4817460,
      "step": 230,
      "time_per_iteration": 2.787911891937256
    },
    {
      "auxiliary_loss_clip": 0.01354364,
      "auxiliary_loss_mlp": 0.01061729,
      "balance_loss_clip": 1.10214245,
      "balance_loss_mlp": 1.04802585,
      "epoch": 0.027776107737630014,
      "flos": 14829922615680.0,
      "grad_norm": 2.4284518178037535,
      "language_loss": 0.91098785,
      "learning_rate": 3.942737468567608e-06,
      "loss": 0.93514878,
      "num_input_tokens_seen": 4835475,
      "step": 231,
      "time_per_iteration": 2.765449285507202
    },
    {
      "auxiliary_loss_clip": 0.01348988,
      "auxiliary_loss_mlp": 0.01047904,
      "balance_loss_clip": 1.09852827,
      "balance_loss_mlp": 1.0341239,
      "epoch": 0.027896350628269104,
      "flos": 47920347066240.0,
      "grad_norm": 5.3465684806328975,
      "language_loss": 0.86013812,
      "learning_rate": 3.9458668295891026e-06,
      "loss": 0.88410699,
      "num_input_tokens_seen": 4857760,
      "step": 232,
      "time_per_iteration": 2.9893720149993896
    },
    {
      "auxiliary_loss_clip": 0.01348164,
      "auxiliary_loss_mlp": 0.01047751,
      "balance_loss_clip": 1.09742785,
      "balance_loss_mlp": 1.03361297,
      "epoch": 0.028016593518908194,
      "flos": 21684550734720.0,
      "grad_norm": 2.3889118383210137,
      "language_loss": 0.86683476,
      "learning_rate": 3.948982730951712e-06,
      "loss": 0.89079392,
      "num_input_tokens_seen": 4875855,
      "step": 233,
      "time_per_iteration": 2.7209746837615967
    },
    {
      "auxiliary_loss_clip": 0.01355515,
      "auxiliary_loss_mlp": 0.01047735,
      "balance_loss_clip": 1.10196698,
      "balance_loss_mlp": 1.03400862,
      "epoch": 0.028136836409547287,
      "flos": 18439483305600.0,
      "grad_norm": 2.3153398062930006,
      "language_loss": 0.82083392,
      "learning_rate": 3.9520852879421254e-06,
      "loss": 0.84486639,
      "num_input_tokens_seen": 4893200,
      "step": 234,
      "time_per_iteration": 2.7447385787963867
    },
    {
      "auxiliary_loss_clip": 0.0134852,
      "auxiliary_loss_mlp": 0.01049668,
      "balance_loss_clip": 1.10179472,
      "balance_loss_mlp": 1.03695416,
      "epoch": 0.028257079300186377,
      "flos": 31576934937600.0,
      "grad_norm": 2.6035341777533425,
      "language_loss": 0.81793094,
      "learning_rate": 3.955174614372137e-06,
      "loss": 0.84191281,
      "num_input_tokens_seen": 4912965,
      "step": 235,
      "time_per_iteration": 2.9367239475250244
    },
    {
      "auxiliary_loss_clip": 0.01351472,
      "auxiliary_loss_mlp": 0.01051407,
      "balance_loss_clip": 1.10115337,
      "balance_loss_mlp": 1.03737617,
      "epoch": 0.028377322190825467,
      "flos": 23513337832320.0,
      "grad_norm": 3.0205303310042786,
      "language_loss": 0.84392649,
      "learning_rate": 3.9582508226037045e-06,
      "loss": 0.86795521,
      "num_input_tokens_seen": 4933105,
      "step": 236,
      "time_per_iteration": 2.8009884357452393
    },
    {
      "auxiliary_loss_clip": 0.01355882,
      "auxiliary_loss_mlp": 0.01046938,
      "balance_loss_clip": 1.10224128,
      "balance_loss_mlp": 1.0320245,
      "epoch": 0.02849756508146456,
      "flos": 20479604071680.0,
      "grad_norm": 3.126810420876177,
      "language_loss": 0.94031394,
      "learning_rate": 3.9613140235734636e-06,
      "loss": 0.96434212,
      "num_input_tokens_seen": 4950085,
      "step": 237,
      "time_per_iteration": 2.818848133087158
    },
    {
      "auxiliary_loss_clip": 0.01345716,
      "auxiliary_loss_mlp": 0.01051909,
      "balance_loss_clip": 1.09649253,
      "balance_loss_mlp": 1.03852165,
      "epoch": 0.02861780797210365,
      "flos": 14283362292480.0,
      "grad_norm": 1.9492820369422654,
      "language_loss": 0.81114453,
      "learning_rate": 3.96436432681674e-06,
      "loss": 0.83512074,
      "num_input_tokens_seen": 4968075,
      "step": 238,
      "time_per_iteration": 2.7809040546417236
    },
    {
      "auxiliary_loss_clip": 0.01351098,
      "auxiliary_loss_mlp": 0.01059997,
      "balance_loss_clip": 1.09921718,
      "balance_loss_mlp": 1.0455842,
      "epoch": 0.02873805086274274,
      "flos": 25808532053760.0,
      "grad_norm": 2.896015817043678,
      "language_loss": 0.88993847,
      "learning_rate": 3.967401840491044e-06,
      "loss": 0.91404939,
      "num_input_tokens_seen": 4987355,
      "step": 239,
      "time_per_iteration": 2.804818868637085
    },
    {
      "auxiliary_loss_clip": 0.01342768,
      "auxiliary_loss_mlp": 0.01055283,
      "balance_loss_clip": 1.09838605,
      "balance_loss_mlp": 1.04143643,
      "epoch": 0.028858293753381833,
      "flos": 17304238984320.0,
      "grad_norm": 2.4441988796828573,
      "language_loss": 0.8776691,
      "learning_rate": 3.97042667139909e-06,
      "loss": 0.90164959,
      "num_input_tokens_seen": 5004680,
      "step": 240,
      "time_per_iteration": 2.754904270172119
    },
    {
      "auxiliary_loss_clip": 0.01345238,
      "auxiliary_loss_mlp": 0.0105009,
      "balance_loss_clip": 1.09929109,
      "balance_loss_mlp": 1.03543878,
      "epoch": 0.028978536644020923,
      "flos": 23038347358080.0,
      "grad_norm": 6.068383854857364,
      "language_loss": 0.87527883,
      "learning_rate": 3.973438925011327e-06,
      "loss": 0.89923203,
      "num_input_tokens_seen": 5022965,
      "step": 241,
      "time_per_iteration": 2.7754368782043457
    },
    {
      "auxiliary_loss_clip": 0.01345657,
      "auxiliary_loss_mlp": 0.01051036,
      "balance_loss_clip": 1.09687865,
      "balance_loss_mlp": 1.03799999,
      "epoch": 0.029098779534660012,
      "flos": 28329712692480.0,
      "grad_norm": 2.669159319658,
      "language_loss": 0.91498315,
      "learning_rate": 3.976438705488002e-06,
      "loss": 0.93895006,
      "num_input_tokens_seen": 5042625,
      "step": 242,
      "time_per_iteration": 2.8563334941864014
    },
    {
      "auxiliary_loss_clip": 0.01345993,
      "auxiliary_loss_mlp": 0.01052138,
      "balance_loss_clip": 1.10155296,
      "balance_loss_mlp": 1.03893578,
      "epoch": 0.029219022425299106,
      "flos": 13881665520000.0,
      "grad_norm": 2.503728396137688,
      "language_loss": 0.93030715,
      "learning_rate": 3.9794261157007744e-06,
      "loss": 0.95428848,
      "num_input_tokens_seen": 5060380,
      "step": 243,
      "time_per_iteration": 2.77337384223938
    },
    {
      "auxiliary_loss_clip": 0.01350233,
      "auxiliary_loss_mlp": 0.01060041,
      "balance_loss_clip": 1.09979486,
      "balance_loss_mlp": 1.04515171,
      "epoch": 0.029339265315938196,
      "flos": 19422501788160.0,
      "grad_norm": 2.5069334635064298,
      "language_loss": 0.84689534,
      "learning_rate": 3.982401257253887e-06,
      "loss": 0.87099808,
      "num_input_tokens_seen": 5078720,
      "step": 244,
      "time_per_iteration": 2.765263795852661
    },
    {
      "auxiliary_loss_clip": 0.01340168,
      "auxiliary_loss_mlp": 0.01048887,
      "balance_loss_clip": 1.09528947,
      "balance_loss_mlp": 1.03563106,
      "epoch": 0.029459508206577285,
      "flos": 15669550005120.0,
      "grad_norm": 2.685864990034809,
      "language_loss": 0.9006784,
      "learning_rate": 3.985364230504893e-06,
      "loss": 0.92456895,
      "num_input_tokens_seen": 5096605,
      "step": 245,
      "time_per_iteration": 2.804231882095337
    },
    {
      "auxiliary_loss_clip": 0.01344364,
      "auxiliary_loss_mlp": 0.01057639,
      "balance_loss_clip": 1.09699297,
      "balance_loss_mlp": 1.04313135,
      "epoch": 0.02957975109721638,
      "flos": 28220975245440.0,
      "grad_norm": 1.8932610035630282,
      "language_loss": 0.84514695,
      "learning_rate": 3.988315134584976e-06,
      "loss": 0.86916697,
      "num_input_tokens_seen": 5116285,
      "step": 246,
      "time_per_iteration": 2.8952322006225586
    },
    {
      "auxiliary_loss_clip": 0.01343434,
      "auxiliary_loss_mlp": 0.01051519,
      "balance_loss_clip": 1.0947144,
      "balance_loss_mlp": 1.03574777,
      "epoch": 0.02969999398785547,
      "flos": 24315869450880.0,
      "grad_norm": 3.821309773843259,
      "language_loss": 0.80428189,
      "learning_rate": 3.991254067418851e-06,
      "loss": 0.82823145,
      "num_input_tokens_seen": 5136825,
      "step": 247,
      "time_per_iteration": 3.032799482345581
    },
    {
      "auxiliary_loss_clip": 0.01339841,
      "auxiliary_loss_mlp": 0.01051495,
      "balance_loss_clip": 1.09616446,
      "balance_loss_mlp": 1.0384717,
      "epoch": 0.02982023687849456,
      "flos": 35078584193280.0,
      "grad_norm": 2.1582477880021402,
      "language_loss": 0.83072686,
      "learning_rate": 3.994181125744254e-06,
      "loss": 0.85464025,
      "num_input_tokens_seen": 5158630,
      "step": 248,
      "time_per_iteration": 2.971174955368042
    },
    {
      "auxiliary_loss_clip": 0.01340289,
      "auxiliary_loss_mlp": 0.01049199,
      "balance_loss_clip": 1.09623718,
      "balance_loss_mlp": 1.03635406,
      "epoch": 0.02994047976913365,
      "flos": 26177155378560.0,
      "grad_norm": 2.088632066253113,
      "language_loss": 0.73913443,
      "learning_rate": 3.99709640513106e-06,
      "loss": 0.76302928,
      "num_input_tokens_seen": 5179510,
      "step": 249,
      "time_per_iteration": 2.8281075954437256
    },
    {
      "auxiliary_loss_clip": 0.01348462,
      "auxiliary_loss_mlp": 0.01051117,
      "balance_loss_clip": 1.09823823,
      "balance_loss_mlp": 1.03733075,
      "epoch": 0.03006072265977274,
      "flos": 25625028447360.0,
      "grad_norm": 6.816670829178058,
      "language_loss": 0.8588227,
      "learning_rate": 4e-06,
      "loss": 0.88281846,
      "num_input_tokens_seen": 5199345,
      "step": 250,
      "time_per_iteration": 3.8108372688293457
    },
    {
      "auxiliary_loss_clip": 0.01341821,
      "auxiliary_loss_mlp": 0.010688,
      "balance_loss_clip": 1.09857023,
      "balance_loss_mlp": 1.05453646,
      "epoch": 0.03018096555041183,
      "flos": 22127078292480.0,
      "grad_norm": 3.125625763282027,
      "language_loss": 0.88394189,
      "learning_rate": 3.999999848300794e-06,
      "loss": 0.90804815,
      "num_input_tokens_seen": 5218330,
      "step": 251,
      "time_per_iteration": 3.801835060119629
    },
    {
      "auxiliary_loss_clip": 0.01335552,
      "auxiliary_loss_mlp": 0.01049475,
      "balance_loss_clip": 1.09202397,
      "balance_loss_mlp": 1.03714848,
      "epoch": 0.030301208441050925,
      "flos": 30188197359360.0,
      "grad_norm": 2.86659690833024,
      "language_loss": 0.89084959,
      "learning_rate": 3.999999393203203e-06,
      "loss": 0.91469985,
      "num_input_tokens_seen": 5240740,
      "step": 252,
      "time_per_iteration": 3.948835611343384
    },
    {
      "auxiliary_loss_clip": 0.01335395,
      "auxiliary_loss_mlp": 0.01054395,
      "balance_loss_clip": 1.09271145,
      "balance_loss_mlp": 1.0413233,
      "epoch": 0.030421451331690014,
      "flos": 23621392920960.0,
      "grad_norm": 1.9779316473184654,
      "language_loss": 0.84849322,
      "learning_rate": 3.999998634707293e-06,
      "loss": 0.8723911,
      "num_input_tokens_seen": 5260290,
      "step": 253,
      "time_per_iteration": 3.857210636138916
    },
    {
      "auxiliary_loss_clip": 0.01347754,
      "auxiliary_loss_mlp": 0.01050184,
      "balance_loss_clip": 1.10188508,
      "balance_loss_mlp": 1.03646266,
      "epoch": 0.030541694222329104,
      "flos": 27928446883200.0,
      "grad_norm": 4.733068461318254,
      "language_loss": 0.96485597,
      "learning_rate": 3.999997572813182e-06,
      "loss": 0.98883545,
      "num_input_tokens_seen": 5278100,
      "step": 254,
      "time_per_iteration": 2.930072069168091
    },
    {
      "auxiliary_loss_clip": 0.01332245,
      "auxiliary_loss_mlp": 0.01052874,
      "balance_loss_clip": 1.09131348,
      "balance_loss_mlp": 1.04056013,
      "epoch": 0.030661937112968194,
      "flos": 18588441006720.0,
      "grad_norm": 2.074594538834533,
      "language_loss": 0.8743943,
      "learning_rate": 3.999996207521028e-06,
      "loss": 0.89824545,
      "num_input_tokens_seen": 5296810,
      "step": 255,
      "time_per_iteration": 2.823927402496338
    },
    {
      "auxiliary_loss_clip": 0.01342606,
      "auxiliary_loss_mlp": 0.01058079,
      "balance_loss_clip": 1.09483993,
      "balance_loss_mlp": 1.04451942,
      "epoch": 0.030782180003607287,
      "flos": 12969139478400.0,
      "grad_norm": 2.6900103329889355,
      "language_loss": 0.82006526,
      "learning_rate": 3.999994538831039e-06,
      "loss": 0.8440721,
      "num_input_tokens_seen": 5313395,
      "step": 256,
      "time_per_iteration": 2.78749942779541
    },
    {
      "auxiliary_loss_clip": 0.013415,
      "auxiliary_loss_mlp": 0.01044121,
      "balance_loss_clip": 1.09565234,
      "balance_loss_mlp": 1.03155029,
      "epoch": 0.030902422894246377,
      "flos": 23335364920320.0,
      "grad_norm": 3.1255889585630223,
      "language_loss": 0.85788071,
      "learning_rate": 3.99999256674347e-06,
      "loss": 0.88173693,
      "num_input_tokens_seen": 5333545,
      "step": 257,
      "time_per_iteration": 2.9167113304138184
    },
    {
      "auxiliary_loss_clip": 0.01287033,
      "auxiliary_loss_mlp": 0.01009573,
      "balance_loss_clip": 1.11401916,
      "balance_loss_mlp": 1.00194359,
      "epoch": 0.031022665784885467,
      "flos": 55094151438720.0,
      "grad_norm": 1.0133523503063164,
      "language_loss": 0.53469801,
      "learning_rate": 3.999990291258618e-06,
      "loss": 0.55766404,
      "num_input_tokens_seen": 5392235,
      "step": 258,
      "time_per_iteration": 3.3357157707214355
    },
    {
      "auxiliary_loss_clip": 0.01340913,
      "auxiliary_loss_mlp": 0.01047525,
      "balance_loss_clip": 1.09750795,
      "balance_loss_mlp": 1.0347513,
      "epoch": 0.03114290867552456,
      "flos": 19317786664320.0,
      "grad_norm": 2.5352565504313773,
      "language_loss": 0.86768931,
      "learning_rate": 3.999987712376829e-06,
      "loss": 0.89157367,
      "num_input_tokens_seen": 5410555,
      "step": 259,
      "time_per_iteration": 2.8712048530578613
    },
    {
      "auxiliary_loss_clip": 0.01339515,
      "auxiliary_loss_mlp": 0.01045031,
      "balance_loss_clip": 1.09789824,
      "balance_loss_mlp": 1.03243625,
      "epoch": 0.031263151566163654,
      "flos": 20959442881920.0,
      "grad_norm": 2.149696148376434,
      "language_loss": 0.81988156,
      "learning_rate": 3.999984830098494e-06,
      "loss": 0.84372699,
      "num_input_tokens_seen": 5430135,
      "step": 260,
      "time_per_iteration": 2.8163013458251953
    },
    {
      "auxiliary_loss_clip": 0.01336853,
      "auxiliary_loss_mlp": 0.01053589,
      "balance_loss_clip": 1.09521079,
      "balance_loss_mlp": 1.04053593,
      "epoch": 0.03138339445680274,
      "flos": 14793006412800.0,
      "grad_norm": 10.197211473278525,
      "language_loss": 0.97925705,
      "learning_rate": 3.999981644424051e-06,
      "loss": 1.00316143,
      "num_input_tokens_seen": 5444935,
      "step": 261,
      "time_per_iteration": 2.8005497455596924
    },
    {
      "auxiliary_loss_clip": 0.01337548,
      "auxiliary_loss_mlp": 0.01055254,
      "balance_loss_clip": 1.09510875,
      "balance_loss_mlp": 1.0418545,
      "epoch": 0.03150363734744183,
      "flos": 11655599022720.0,
      "grad_norm": 2.5538454581829955,
      "language_loss": 0.86118734,
      "learning_rate": 3.999978155353982e-06,
      "loss": 0.88511544,
      "num_input_tokens_seen": 5462080,
      "step": 262,
      "time_per_iteration": 2.778538942337036
    },
    {
      "auxiliary_loss_clip": 0.01338932,
      "auxiliary_loss_mlp": 0.01052875,
      "balance_loss_clip": 1.09458613,
      "balance_loss_mlp": 1.03946364,
      "epoch": 0.03162388023808092,
      "flos": 33727732485120.0,
      "grad_norm": 4.031965289812195,
      "language_loss": 0.8030324,
      "learning_rate": 3.9999743628888186e-06,
      "loss": 0.82695049,
      "num_input_tokens_seen": 5483870,
      "step": 263,
      "time_per_iteration": 2.9718143939971924
    },
    {
      "auxiliary_loss_clip": 0.01336032,
      "auxiliary_loss_mlp": 0.01041891,
      "balance_loss_clip": 1.09265518,
      "balance_loss_mlp": 1.02889729,
      "epoch": 0.03174412312872001,
      "flos": 20810952057600.0,
      "grad_norm": 3.17582198277961,
      "language_loss": 0.89755172,
      "learning_rate": 3.999970267029133e-06,
      "loss": 0.92133093,
      "num_input_tokens_seen": 5502830,
      "step": 264,
      "time_per_iteration": 2.877045154571533
    },
    {
      "auxiliary_loss_clip": 0.01335376,
      "auxiliary_loss_mlp": 0.01042721,
      "balance_loss_clip": 1.09457791,
      "balance_loss_mlp": 1.03043103,
      "epoch": 0.0318643660193591,
      "flos": 23727939638400.0,
      "grad_norm": 3.4862208557705006,
      "language_loss": 0.80065787,
      "learning_rate": 3.999965867775548e-06,
      "loss": 0.82443881,
      "num_input_tokens_seen": 5523225,
      "step": 265,
      "time_per_iteration": 2.843622922897339
    },
    {
      "auxiliary_loss_clip": 0.01329444,
      "auxiliary_loss_mlp": 0.01051522,
      "balance_loss_clip": 1.09027636,
      "balance_loss_mlp": 1.03839076,
      "epoch": 0.0319846089099982,
      "flos": 13917863450880.0,
      "grad_norm": 3.8454419625070932,
      "language_loss": 0.86982346,
      "learning_rate": 3.9999611651287315e-06,
      "loss": 0.89363313,
      "num_input_tokens_seen": 5541380,
      "step": 266,
      "time_per_iteration": 2.7257847785949707
    },
    {
      "auxiliary_loss_clip": 0.013381,
      "auxiliary_loss_mlp": 0.01049526,
      "balance_loss_clip": 1.09426999,
      "balance_loss_mlp": 1.03616261,
      "epoch": 0.03210485180063729,
      "flos": 14753253035520.0,
      "grad_norm": 2.5900079616428258,
      "language_loss": 0.78845614,
      "learning_rate": 3.999956159089396e-06,
      "loss": 0.81233239,
      "num_input_tokens_seen": 5558830,
      "step": 267,
      "time_per_iteration": 2.782325267791748
    },
    {
      "auxiliary_loss_clip": 0.01332773,
      "auxiliary_loss_mlp": 0.01052993,
      "balance_loss_clip": 1.09255803,
      "balance_loss_mlp": 1.03933704,
      "epoch": 0.03222509469127638,
      "flos": 28913153304960.0,
      "grad_norm": 2.546387430530761,
      "language_loss": 0.79494566,
      "learning_rate": 3.999950849658302e-06,
      "loss": 0.81880331,
      "num_input_tokens_seen": 5577750,
      "step": 268,
      "time_per_iteration": 2.7528021335601807
    },
    {
      "auxiliary_loss_clip": 0.01338098,
      "auxiliary_loss_mlp": 0.01053685,
      "balance_loss_clip": 1.09447348,
      "balance_loss_mlp": 1.04033995,
      "epoch": 0.03234533758191547,
      "flos": 16946389739520.0,
      "grad_norm": 2.4056250199516427,
      "language_loss": 0.84279263,
      "learning_rate": 3.999945236836254e-06,
      "loss": 0.86671042,
      "num_input_tokens_seen": 5596715,
      "step": 269,
      "time_per_iteration": 2.662045478820801
    },
    {
      "auxiliary_loss_clip": 0.0134117,
      "auxiliary_loss_mlp": 0.01047767,
      "balance_loss_clip": 1.09677863,
      "balance_loss_mlp": 1.03439212,
      "epoch": 0.03246558047255456,
      "flos": 18989096284800.0,
      "grad_norm": 7.269938419316769,
      "language_loss": 0.94754583,
      "learning_rate": 3.999939320624103e-06,
      "loss": 0.97143525,
      "num_input_tokens_seen": 5611865,
      "step": 270,
      "time_per_iteration": 2.6919608116149902
    },
    {
      "auxiliary_loss_clip": 0.01338612,
      "auxiliary_loss_mlp": 0.01053971,
      "balance_loss_clip": 1.0957396,
      "balance_loss_mlp": 1.04126334,
      "epoch": 0.03258582336319365,
      "flos": 23728334688000.0,
      "grad_norm": 1.9794311798870219,
      "language_loss": 0.89741373,
      "learning_rate": 3.999933101022749e-06,
      "loss": 0.92133951,
      "num_input_tokens_seen": 5632270,
      "step": 271,
      "time_per_iteration": 2.7683424949645996
    },
    {
      "auxiliary_loss_clip": 0.0133512,
      "auxiliary_loss_mlp": 0.01051519,
      "balance_loss_clip": 1.09472823,
      "balance_loss_mlp": 1.03762519,
      "epoch": 0.032706066253832745,
      "flos": 27670823562240.0,
      "grad_norm": 5.93741361288679,
      "language_loss": 0.86896449,
      "learning_rate": 3.999926578033132e-06,
      "loss": 0.89283085,
      "num_input_tokens_seen": 5652085,
      "step": 272,
      "time_per_iteration": 2.735405445098877
    },
    {
      "auxiliary_loss_clip": 0.01335904,
      "auxiliary_loss_mlp": 0.01052581,
      "balance_loss_clip": 1.09321272,
      "balance_loss_mlp": 1.03800726,
      "epoch": 0.032826309144471835,
      "flos": 45624685968000.0,
      "grad_norm": 2.163945435116917,
      "language_loss": 0.63048041,
      "learning_rate": 3.999919751656244e-06,
      "loss": 0.6543653,
      "num_input_tokens_seen": 5678985,
      "step": 273,
      "time_per_iteration": 2.951040506362915
    },
    {
      "auxiliary_loss_clip": 0.01337855,
      "auxiliary_loss_mlp": 0.01051075,
      "balance_loss_clip": 1.09649622,
      "balance_loss_mlp": 1.03689516,
      "epoch": 0.032946552035110925,
      "flos": 25812374808960.0,
      "grad_norm": 2.6761635615992465,
      "language_loss": 0.76008534,
      "learning_rate": 3.9999126218931195e-06,
      "loss": 0.78397465,
      "num_input_tokens_seen": 5697020,
      "step": 274,
      "time_per_iteration": 2.782667398452759
    },
    {
      "auxiliary_loss_clip": 0.01336813,
      "auxiliary_loss_mlp": 0.01051471,
      "balance_loss_clip": 1.09756875,
      "balance_loss_mlp": 1.03808951,
      "epoch": 0.033066794925750015,
      "flos": 15121984101120.0,
      "grad_norm": 2.762134296438335,
      "language_loss": 0.89619279,
      "learning_rate": 3.99990518874484e-06,
      "loss": 0.92007565,
      "num_input_tokens_seen": 5713460,
      "step": 275,
      "time_per_iteration": 2.68269419670105
    },
    {
      "auxiliary_loss_clip": 0.01337846,
      "auxiliary_loss_mlp": 0.01064409,
      "balance_loss_clip": 1.09644914,
      "balance_loss_mlp": 1.05105162,
      "epoch": 0.033187037816389105,
      "flos": 22776593973120.0,
      "grad_norm": 2.6973177973036084,
      "language_loss": 0.92412651,
      "learning_rate": 3.999897452212534e-06,
      "loss": 0.94814909,
      "num_input_tokens_seen": 5730790,
      "step": 276,
      "time_per_iteration": 3.7675364017486572
    },
    {
      "auxiliary_loss_clip": 0.01335095,
      "auxiliary_loss_mlp": 0.01054433,
      "balance_loss_clip": 1.09305429,
      "balance_loss_mlp": 1.03920376,
      "epoch": 0.033307280707028195,
      "flos": 23331414424320.0,
      "grad_norm": 5.853511870832606,
      "language_loss": 1.0015769,
      "learning_rate": 3.999889412297374e-06,
      "loss": 1.02547228,
      "num_input_tokens_seen": 5750215,
      "step": 277,
      "time_per_iteration": 2.779381275177002
    },
    {
      "auxiliary_loss_clip": 0.01332744,
      "auxiliary_loss_mlp": 0.01059679,
      "balance_loss_clip": 1.09134758,
      "balance_loss_mlp": 1.04568958,
      "epoch": 0.03342752359766729,
      "flos": 28840290566400.0,
      "grad_norm": 3.052034395652839,
      "language_loss": 0.78805721,
      "learning_rate": 3.999881069000581e-06,
      "loss": 0.81198144,
      "num_input_tokens_seen": 5769945,
      "step": 278,
      "time_per_iteration": 4.577516555786133
    },
    {
      "auxiliary_loss_clip": 0.01337845,
      "auxiliary_loss_mlp": 0.01049394,
      "balance_loss_clip": 1.09674704,
      "balance_loss_mlp": 1.03582764,
      "epoch": 0.03354776648830638,
      "flos": 19384544090880.0,
      "grad_norm": 2.835161479941073,
      "language_loss": 0.86861134,
      "learning_rate": 3.99987242232342e-06,
      "loss": 0.89248371,
      "num_input_tokens_seen": 5784950,
      "step": 279,
      "time_per_iteration": 3.576709032058716
    },
    {
      "auxiliary_loss_clip": 0.01336151,
      "auxiliary_loss_mlp": 0.01049955,
      "balance_loss_clip": 1.09769392,
      "balance_loss_mlp": 1.03590035,
      "epoch": 0.03366800937894547,
      "flos": 17858628472320.0,
      "grad_norm": 3.0008860052515454,
      "language_loss": 0.79732925,
      "learning_rate": 3.9998634722672026e-06,
      "loss": 0.82119036,
      "num_input_tokens_seen": 5805005,
      "step": 280,
      "time_per_iteration": 2.694018602371216
    },
    {
      "auxiliary_loss_clip": 0.01337212,
      "auxiliary_loss_mlp": 0.01050959,
      "balance_loss_clip": 1.09715867,
      "balance_loss_mlp": 1.03674936,
      "epoch": 0.03378825226958456,
      "flos": 35951033635200.0,
      "grad_norm": 1.9833914608646728,
      "language_loss": 0.77976882,
      "learning_rate": 3.999854218833286e-06,
      "loss": 0.80365056,
      "num_input_tokens_seen": 5825825,
      "step": 281,
      "time_per_iteration": 2.8835127353668213
    },
    {
      "auxiliary_loss_clip": 0.01336006,
      "auxiliary_loss_mlp": 0.01048855,
      "balance_loss_clip": 1.09717059,
      "balance_loss_mlp": 1.03543186,
      "epoch": 0.03390849516022365,
      "flos": 25702488126720.0,
      "grad_norm": 7.416062628628208,
      "language_loss": 0.8189227,
      "learning_rate": 3.999844662023075e-06,
      "loss": 0.84277135,
      "num_input_tokens_seen": 5845700,
      "step": 282,
      "time_per_iteration": 2.756861925125122
    },
    {
      "auxiliary_loss_clip": 0.01328241,
      "auxiliary_loss_mlp": 0.01047332,
      "balance_loss_clip": 1.09263444,
      "balance_loss_mlp": 1.03517294,
      "epoch": 0.03402873805086274,
      "flos": 21284505987840.0,
      "grad_norm": 1.8272934930057139,
      "language_loss": 0.92315215,
      "learning_rate": 3.999834801838018e-06,
      "loss": 0.94690788,
      "num_input_tokens_seen": 5864680,
      "step": 283,
      "time_per_iteration": 2.702091932296753
    },
    {
      "auxiliary_loss_clip": 0.01330363,
      "auxiliary_loss_mlp": 0.01043827,
      "balance_loss_clip": 1.09400415,
      "balance_loss_mlp": 1.03143549,
      "epoch": 0.03414898094150183,
      "flos": 22710913954560.0,
      "grad_norm": 2.9201385675533613,
      "language_loss": 0.74219394,
      "learning_rate": 3.9998246382796115e-06,
      "loss": 0.7659359,
      "num_input_tokens_seen": 5884260,
      "step": 284,
      "time_per_iteration": 2.7426586151123047
    },
    {
      "auxiliary_loss_clip": 0.01339152,
      "auxiliary_loss_mlp": 0.01049101,
      "balance_loss_clip": 1.09695017,
      "balance_loss_mlp": 1.03556442,
      "epoch": 0.03426922383214093,
      "flos": 18879927874560.0,
      "grad_norm": 2.69208050639597,
      "language_loss": 0.91137791,
      "learning_rate": 3.999814171349399e-06,
      "loss": 0.93526042,
      "num_input_tokens_seen": 5902120,
      "step": 285,
      "time_per_iteration": 2.718337059020996
    },
    {
      "auxiliary_loss_clip": 0.01329816,
      "auxiliary_loss_mlp": 0.01045847,
      "balance_loss_clip": 1.09148002,
      "balance_loss_mlp": 1.03306794,
      "epoch": 0.03438946672278002,
      "flos": 34752012716160.0,
      "grad_norm": 1.71229236036837,
      "language_loss": 0.7350924,
      "learning_rate": 3.9998034010489655e-06,
      "loss": 0.75884902,
      "num_input_tokens_seen": 5925810,
      "step": 286,
      "time_per_iteration": 2.8784754276275635
    },
    {
      "auxiliary_loss_clip": 0.01334825,
      "auxiliary_loss_mlp": 0.01047586,
      "balance_loss_clip": 1.09780109,
      "balance_loss_mlp": 1.03493786,
      "epoch": 0.03450970961341911,
      "flos": 22164102236160.0,
      "grad_norm": 2.18126020922581,
      "language_loss": 0.76005703,
      "learning_rate": 3.999792327379946e-06,
      "loss": 0.78388119,
      "num_input_tokens_seen": 5945185,
      "step": 287,
      "time_per_iteration": 2.7193641662597656
    },
    {
      "auxiliary_loss_clip": 0.01333397,
      "auxiliary_loss_mlp": 0.01048303,
      "balance_loss_clip": 1.09866226,
      "balance_loss_mlp": 1.03551829,
      "epoch": 0.034629952504058197,
      "flos": 21725740656000.0,
      "grad_norm": 2.3163462828698504,
      "language_loss": 0.96117485,
      "learning_rate": 3.999780950344021e-06,
      "loss": 0.98499185,
      "num_input_tokens_seen": 5963375,
      "step": 288,
      "time_per_iteration": 2.7083470821380615
    },
    {
      "auxiliary_loss_clip": 0.0133637,
      "auxiliary_loss_mlp": 0.01055054,
      "balance_loss_clip": 1.09743774,
      "balance_loss_mlp": 1.04204226,
      "epoch": 0.034750195394697286,
      "flos": 20047994248320.0,
      "grad_norm": 1.8503928032587733,
      "language_loss": 0.82561493,
      "learning_rate": 3.999769269942916e-06,
      "loss": 0.84952915,
      "num_input_tokens_seen": 5983415,
      "step": 289,
      "time_per_iteration": 2.7275896072387695
    },
    {
      "auxiliary_loss_clip": 0.0132747,
      "auxiliary_loss_mlp": 0.01055209,
      "balance_loss_clip": 1.09342456,
      "balance_loss_mlp": 1.04283464,
      "epoch": 0.034870438285336376,
      "flos": 27965865876480.0,
      "grad_norm": 1.9341498231337546,
      "language_loss": 0.81119931,
      "learning_rate": 3.999757286178402e-06,
      "loss": 0.83502614,
      "num_input_tokens_seen": 6005850,
      "step": 290,
      "time_per_iteration": 2.799401044845581
    },
    {
      "auxiliary_loss_clip": 0.01327405,
      "auxiliary_loss_mlp": 0.01047209,
      "balance_loss_clip": 1.09213984,
      "balance_loss_mlp": 1.03295732,
      "epoch": 0.03499068117597547,
      "flos": 22017514832640.0,
      "grad_norm": 5.173096817397172,
      "language_loss": 0.90947974,
      "learning_rate": 3.999744999052299e-06,
      "loss": 0.93322587,
      "num_input_tokens_seen": 6027240,
      "step": 291,
      "time_per_iteration": 2.7106142044067383
    },
    {
      "auxiliary_loss_clip": 0.01265485,
      "auxiliary_loss_mlp": 0.01013141,
      "balance_loss_clip": 1.09887481,
      "balance_loss_mlp": 1.00584495,
      "epoch": 0.03511092406661456,
      "flos": 57242147725440.0,
      "grad_norm": 0.9580747397684949,
      "language_loss": 0.61135793,
      "learning_rate": 3.9997324085664675e-06,
      "loss": 0.63414419,
      "num_input_tokens_seen": 6087470,
      "step": 292,
      "time_per_iteration": 3.19050669670105
    },
    {
      "auxiliary_loss_clip": 0.01326484,
      "auxiliary_loss_mlp": 0.01048959,
      "balance_loss_clip": 1.09220505,
      "balance_loss_mlp": 1.03653145,
      "epoch": 0.03523116695725365,
      "flos": 22928065626240.0,
      "grad_norm": 2.296082579829729,
      "language_loss": 0.92035115,
      "learning_rate": 3.999719514722821e-06,
      "loss": 0.94410563,
      "num_input_tokens_seen": 6107600,
      "step": 293,
      "time_per_iteration": 2.7352406978607178
    },
    {
      "auxiliary_loss_clip": 0.01323834,
      "auxiliary_loss_mlp": 0.01046206,
      "balance_loss_clip": 1.09316397,
      "balance_loss_mlp": 1.03332579,
      "epoch": 0.03535140984789274,
      "flos": 36903241226880.0,
      "grad_norm": 2.9613055609848042,
      "language_loss": 0.74819589,
      "learning_rate": 3.999706317523314e-06,
      "loss": 0.77189624,
      "num_input_tokens_seen": 6126160,
      "step": 294,
      "time_per_iteration": 2.849459648132324
    },
    {
      "auxiliary_loss_clip": 0.0132674,
      "auxiliary_loss_mlp": 0.01057503,
      "balance_loss_clip": 1.09248543,
      "balance_loss_mlp": 1.04563606,
      "epoch": 0.03547165273853183,
      "flos": 20449152316800.0,
      "grad_norm": 2.5700569200861163,
      "language_loss": 0.86365366,
      "learning_rate": 3.999692816969948e-06,
      "loss": 0.88749611,
      "num_input_tokens_seen": 6145695,
      "step": 295,
      "time_per_iteration": 2.799098491668701
    },
    {
      "auxiliary_loss_clip": 0.01260599,
      "auxiliary_loss_mlp": 0.01009384,
      "balance_loss_clip": 1.09567714,
      "balance_loss_mlp": 1.00218379,
      "epoch": 0.03559189562917092,
      "flos": 69850564871040.0,
      "grad_norm": 1.0041374230104665,
      "language_loss": 0.69402945,
      "learning_rate": 3.999679013064772e-06,
      "loss": 0.71672928,
      "num_input_tokens_seen": 6212440,
      "step": 296,
      "time_per_iteration": 3.275622606277466
    },
    {
      "auxiliary_loss_clip": 0.01322486,
      "auxiliary_loss_mlp": 0.01052437,
      "balance_loss_clip": 1.08904421,
      "balance_loss_mlp": 1.03928864,
      "epoch": 0.03571213851981002,
      "flos": 21651944163840.0,
      "grad_norm": 9.542947139609506,
      "language_loss": 0.85500067,
      "learning_rate": 3.99966490580988e-06,
      "loss": 0.87874991,
      "num_input_tokens_seen": 6229800,
      "step": 297,
      "time_per_iteration": 2.7616453170776367
    },
    {
      "auxiliary_loss_clip": 0.01328753,
      "auxiliary_loss_mlp": 0.01059982,
      "balance_loss_clip": 1.09283423,
      "balance_loss_mlp": 1.04747081,
      "epoch": 0.03583238141044911,
      "flos": 43945610757120.0,
      "grad_norm": 2.7774151684842283,
      "language_loss": 0.65936404,
      "learning_rate": 3.999650495207411e-06,
      "loss": 0.68325138,
      "num_input_tokens_seen": 6255825,
      "step": 298,
      "time_per_iteration": 2.894382953643799
    },
    {
      "auxiliary_loss_clip": 0.01325217,
      "auxiliary_loss_mlp": 0.01050845,
      "balance_loss_clip": 1.09470499,
      "balance_loss_mlp": 1.03760076,
      "epoch": 0.0359526243010882,
      "flos": 18910810592640.0,
      "grad_norm": 2.74616115415114,
      "language_loss": 0.90536159,
      "learning_rate": 3.999635781259553e-06,
      "loss": 0.92912221,
      "num_input_tokens_seen": 6271090,
      "step": 299,
      "time_per_iteration": 2.7299745082855225
    },
    {
      "auxiliary_loss_clip": 0.01251956,
      "auxiliary_loss_mlp": 0.01009736,
      "balance_loss_clip": 1.09032929,
      "balance_loss_mlp": 1.00291681,
      "epoch": 0.03607286719172729,
      "flos": 61668892782720.0,
      "grad_norm": 0.918459688057845,
      "language_loss": 0.52247119,
      "learning_rate": 3.999620763968535e-06,
      "loss": 0.54508817,
      "num_input_tokens_seen": 6329965,
      "step": 300,
      "time_per_iteration": 3.0763652324676514
    },
    {
      "auxiliary_loss_clip": 0.01321418,
      "auxiliary_loss_mlp": 0.01051347,
      "balance_loss_clip": 1.09049988,
      "balance_loss_mlp": 1.03913426,
      "epoch": 0.03619311008236638,
      "flos": 27819062991360.0,
      "grad_norm": 2.0023628024299764,
      "language_loss": 0.86660218,
      "learning_rate": 3.999605443336638e-06,
      "loss": 0.89032984,
      "num_input_tokens_seen": 6352095,
      "step": 301,
      "time_per_iteration": 2.7721498012542725
    },
    {
      "auxiliary_loss_clip": 0.01329302,
      "auxiliary_loss_mlp": 0.01050331,
      "balance_loss_clip": 1.09402323,
      "balance_loss_mlp": 1.03703284,
      "epoch": 0.03631335297300547,
      "flos": 13621133197440.0,
      "grad_norm": 3.349462954381713,
      "language_loss": 0.89777809,
      "learning_rate": 3.999589819366185e-06,
      "loss": 0.92157441,
      "num_input_tokens_seen": 6365885,
      "step": 302,
      "time_per_iteration": 2.6580698490142822
    },
    {
      "auxiliary_loss_clip": 0.01325041,
      "auxiliary_loss_mlp": 0.01055517,
      "balance_loss_clip": 1.09165907,
      "balance_loss_mlp": 1.04260087,
      "epoch": 0.036433595863644565,
      "flos": 27631788456960.0,
      "grad_norm": 2.591831385443148,
      "language_loss": 0.84824634,
      "learning_rate": 3.999573892059547e-06,
      "loss": 0.87205184,
      "num_input_tokens_seen": 6385015,
      "step": 303,
      "time_per_iteration": 3.7360095977783203
    },
    {
      "auxiliary_loss_clip": 0.01327344,
      "auxiliary_loss_mlp": 0.01052715,
      "balance_loss_clip": 1.09166026,
      "balance_loss_mlp": 1.03919077,
      "epoch": 0.036553838754283655,
      "flos": 24572020314240.0,
      "grad_norm": 2.1236231168962902,
      "language_loss": 0.81070441,
      "learning_rate": 3.999557661419138e-06,
      "loss": 0.83450496,
      "num_input_tokens_seen": 6405165,
      "step": 304,
      "time_per_iteration": 3.549142599105835
    },
    {
      "auxiliary_loss_clip": 0.01328457,
      "auxiliary_loss_mlp": 0.01055584,
      "balance_loss_clip": 1.0944314,
      "balance_loss_mlp": 1.04292989,
      "epoch": 0.036674081644922744,
      "flos": 23404313076480.0,
      "grad_norm": 2.079245562488903,
      "language_loss": 0.81450164,
      "learning_rate": 3.9995411274474225e-06,
      "loss": 0.83834201,
      "num_input_tokens_seen": 6424445,
      "step": 305,
      "time_per_iteration": 4.4558680057525635
    },
    {
      "auxiliary_loss_clip": 0.01322952,
      "auxiliary_loss_mlp": 0.01054476,
      "balance_loss_clip": 1.09091389,
      "balance_loss_mlp": 1.0422039,
      "epoch": 0.036794324535561834,
      "flos": 27489690253440.0,
      "grad_norm": 2.9377998583891745,
      "language_loss": 0.81516302,
      "learning_rate": 3.999524290146908e-06,
      "loss": 0.83893728,
      "num_input_tokens_seen": 6444650,
      "step": 306,
      "time_per_iteration": 2.78656005859375
    },
    {
      "auxiliary_loss_clip": 0.01327186,
      "auxiliary_loss_mlp": 0.01051168,
      "balance_loss_clip": 1.09433341,
      "balance_loss_mlp": 1.03826356,
      "epoch": 0.036914567426200924,
      "flos": 19463476227840.0,
      "grad_norm": 3.1902606135717475,
      "language_loss": 0.92629868,
      "learning_rate": 3.9995071495201485e-06,
      "loss": 0.95008218,
      "num_input_tokens_seen": 6461755,
      "step": 307,
      "time_per_iteration": 2.6811363697052
    },
    {
      "auxiliary_loss_clip": 0.01323956,
      "auxiliary_loss_mlp": 0.0104652,
      "balance_loss_clip": 1.09033835,
      "balance_loss_mlp": 1.03448009,
      "epoch": 0.037034810316840014,
      "flos": 22309324922880.0,
      "grad_norm": 2.818004209338718,
      "language_loss": 0.97866166,
      "learning_rate": 3.999489705569744e-06,
      "loss": 1.0023663,
      "num_input_tokens_seen": 6479455,
      "step": 308,
      "time_per_iteration": 2.7377045154571533
    },
    {
      "auxiliary_loss_clip": 0.01322331,
      "auxiliary_loss_mlp": 0.01049714,
      "balance_loss_clip": 1.09023857,
      "balance_loss_mlp": 1.03597534,
      "epoch": 0.03715505320747911,
      "flos": 18588333265920.0,
      "grad_norm": 2.683957019916267,
      "language_loss": 0.86502182,
      "learning_rate": 3.999471958298341e-06,
      "loss": 0.88874227,
      "num_input_tokens_seen": 6498365,
      "step": 309,
      "time_per_iteration": 2.654801607131958
    },
    {
      "auxiliary_loss_clip": 0.01326591,
      "auxiliary_loss_mlp": 0.01052988,
      "balance_loss_clip": 1.0928179,
      "balance_loss_mlp": 1.04085255,
      "epoch": 0.0372752960981182,
      "flos": 35955343267200.0,
      "grad_norm": 1.9380618188348777,
      "language_loss": 0.76169431,
      "learning_rate": 3.999453907708631e-06,
      "loss": 0.7854901,
      "num_input_tokens_seen": 6520770,
      "step": 310,
      "time_per_iteration": 2.8453903198242188
    },
    {
      "auxiliary_loss_clip": 0.01319266,
      "auxiliary_loss_mlp": 0.01048852,
      "balance_loss_clip": 1.09047198,
      "balance_loss_mlp": 1.03659689,
      "epoch": 0.03739553898875729,
      "flos": 20814040627200.0,
      "grad_norm": 2.16920827890716,
      "language_loss": 0.81369871,
      "learning_rate": 3.999435553803353e-06,
      "loss": 0.83737987,
      "num_input_tokens_seen": 6540170,
      "step": 311,
      "time_per_iteration": 2.6997928619384766
    },
    {
      "auxiliary_loss_clip": 0.01318903,
      "auxiliary_loss_mlp": 0.0104906,
      "balance_loss_clip": 1.08667326,
      "balance_loss_mlp": 1.03653085,
      "epoch": 0.03751578187939638,
      "flos": 20264140339200.0,
      "grad_norm": 2.9911587449391392,
      "language_loss": 0.83243591,
      "learning_rate": 3.999416896585292e-06,
      "loss": 0.85611558,
      "num_input_tokens_seen": 6557200,
      "step": 312,
      "time_per_iteration": 2.731782913208008
    },
    {
      "auxiliary_loss_clip": 0.01317331,
      "auxiliary_loss_mlp": 0.01054271,
      "balance_loss_clip": 1.0867604,
      "balance_loss_mlp": 1.04227901,
      "epoch": 0.03763602477003547,
      "flos": 20668063754880.0,
      "grad_norm": 3.86198837380606,
      "language_loss": 0.86530519,
      "learning_rate": 3.9993979360572775e-06,
      "loss": 0.88902116,
      "num_input_tokens_seen": 6577340,
      "step": 313,
      "time_per_iteration": 2.702183246612549
    },
    {
      "auxiliary_loss_clip": 0.01326459,
      "auxiliary_loss_mlp": 0.01052061,
      "balance_loss_clip": 1.09223163,
      "balance_loss_mlp": 1.03891826,
      "epoch": 0.03775626766067456,
      "flos": 16691352197760.0,
      "grad_norm": 2.5439029542667635,
      "language_loss": 0.82894719,
      "learning_rate": 3.999378672222185e-06,
      "loss": 0.85273242,
      "num_input_tokens_seen": 6595125,
      "step": 314,
      "time_per_iteration": 2.6662468910217285
    },
    {
      "auxiliary_loss_clip": 0.01317341,
      "auxiliary_loss_mlp": 0.01045863,
      "balance_loss_clip": 1.08905101,
      "balance_loss_mlp": 1.03337622,
      "epoch": 0.03787651055131366,
      "flos": 21141797253120.0,
      "grad_norm": 10.491733417348213,
      "language_loss": 0.82849598,
      "learning_rate": 3.9993591050829385e-06,
      "loss": 0.85212803,
      "num_input_tokens_seen": 6612990,
      "step": 315,
      "time_per_iteration": 2.7052464485168457
    },
    {
      "auxiliary_loss_clip": 0.01319947,
      "auxiliary_loss_mlp": 0.0104828,
      "balance_loss_clip": 1.09030318,
      "balance_loss_mlp": 1.03651404,
      "epoch": 0.037996753441952746,
      "flos": 22018089450240.0,
      "grad_norm": 2.0942061985009492,
      "language_loss": 0.7941404,
      "learning_rate": 3.999339234642506e-06,
      "loss": 0.81782264,
      "num_input_tokens_seen": 6632740,
      "step": 316,
      "time_per_iteration": 2.722428798675537
    },
    {
      "auxiliary_loss_clip": 0.01317439,
      "auxiliary_loss_mlp": 0.0104026,
      "balance_loss_clip": 1.08655167,
      "balance_loss_mlp": 1.0270164,
      "epoch": 0.038116996332591836,
      "flos": 27709391790720.0,
      "grad_norm": 2.0174633497747285,
      "language_loss": 0.8383162,
      "learning_rate": 3.9993190609038994e-06,
      "loss": 0.86189318,
      "num_input_tokens_seen": 6651505,
      "step": 317,
      "time_per_iteration": 2.7539632320404053
    },
    {
      "auxiliary_loss_clip": 0.01316851,
      "auxiliary_loss_mlp": 0.01051625,
      "balance_loss_clip": 1.08711958,
      "balance_loss_mlp": 1.03955507,
      "epoch": 0.038237239223230926,
      "flos": 21178067011200.0,
      "grad_norm": 2.093004160096706,
      "language_loss": 0.8335638,
      "learning_rate": 3.999298583870182e-06,
      "loss": 0.85724854,
      "num_input_tokens_seen": 6671090,
      "step": 318,
      "time_per_iteration": 2.667309522628784
    },
    {
      "auxiliary_loss_clip": 0.01318958,
      "auxiliary_loss_mlp": 0.01041538,
      "balance_loss_clip": 1.08885217,
      "balance_loss_mlp": 1.02857971,
      "epoch": 0.038357482113870016,
      "flos": 25556618995200.0,
      "grad_norm": 3.399519216696956,
      "language_loss": 0.77806628,
      "learning_rate": 3.999277803544458e-06,
      "loss": 0.80167127,
      "num_input_tokens_seen": 6691245,
      "step": 319,
      "time_per_iteration": 2.7995965480804443
    },
    {
      "auxiliary_loss_clip": 0.01239688,
      "auxiliary_loss_mlp": 0.0101914,
      "balance_loss_clip": 1.0828054,
      "balance_loss_mlp": 1.01265514,
      "epoch": 0.038477725004509106,
      "flos": 59227578034560.0,
      "grad_norm": 0.9629175216749136,
      "language_loss": 0.62326616,
      "learning_rate": 3.999256719929882e-06,
      "loss": 0.64585441,
      "num_input_tokens_seen": 6752520,
      "step": 320,
      "time_per_iteration": 3.231663942337036
    },
    {
      "auxiliary_loss_clip": 0.01237717,
      "auxiliary_loss_mlp": 0.01012588,
      "balance_loss_clip": 1.08157015,
      "balance_loss_mlp": 1.0061512,
      "epoch": 0.0385979678951482,
      "flos": 67317676398720.0,
      "grad_norm": 1.2155119411555246,
      "language_loss": 0.67150438,
      "learning_rate": 3.999235333029651e-06,
      "loss": 0.6940074,
      "num_input_tokens_seen": 6806460,
      "step": 321,
      "time_per_iteration": 3.197641611099243
    },
    {
      "auxiliary_loss_clip": 0.01318547,
      "auxiliary_loss_mlp": 0.01056175,
      "balance_loss_clip": 1.09142828,
      "balance_loss_mlp": 1.0431335,
      "epoch": 0.03871821078578729,
      "flos": 22746752749440.0,
      "grad_norm": 2.516307521713106,
      "language_loss": 0.82285827,
      "learning_rate": 3.999213642847009e-06,
      "loss": 0.84660548,
      "num_input_tokens_seen": 6827045,
      "step": 322,
      "time_per_iteration": 2.681889295578003
    },
    {
      "auxiliary_loss_clip": 0.01315167,
      "auxiliary_loss_mlp": 0.0105171,
      "balance_loss_clip": 1.08657694,
      "balance_loss_mlp": 1.03822148,
      "epoch": 0.03883845367642638,
      "flos": 26280613526400.0,
      "grad_norm": 2.03796713796626,
      "language_loss": 0.91236633,
      "learning_rate": 3.999191649385247e-06,
      "loss": 0.9360351,
      "num_input_tokens_seen": 6848220,
      "step": 323,
      "time_per_iteration": 2.835099220275879
    },
    {
      "auxiliary_loss_clip": 0.01234182,
      "auxiliary_loss_mlp": 0.01006151,
      "balance_loss_clip": 1.08021522,
      "balance_loss_mlp": 0.99995178,
      "epoch": 0.03895869656706547,
      "flos": 56962835568000.0,
      "grad_norm": 0.9230242679723929,
      "language_loss": 0.59729338,
      "learning_rate": 3.999169352647702e-06,
      "loss": 0.61969668,
      "num_input_tokens_seen": 6909400,
      "step": 324,
      "time_per_iteration": 3.1734840869903564
    },
    {
      "auxiliary_loss_clip": 0.01321663,
      "auxiliary_loss_mlp": 0.01058713,
      "balance_loss_clip": 1.08928835,
      "balance_loss_mlp": 1.04471779,
      "epoch": 0.03907893945770456,
      "flos": 24863363527680.0,
      "grad_norm": 2.0091982220977234,
      "language_loss": 0.83005184,
      "learning_rate": 3.999146752637755e-06,
      "loss": 0.85385561,
      "num_input_tokens_seen": 6930445,
      "step": 325,
      "time_per_iteration": 2.7144134044647217
    },
    {
      "auxiliary_loss_clip": 0.01319545,
      "auxiliary_loss_mlp": 0.0104359,
      "balance_loss_clip": 1.08923697,
      "balance_loss_mlp": 1.03090048,
      "epoch": 0.03919918234834365,
      "flos": 18368595815040.0,
      "grad_norm": 2.768991023118778,
      "language_loss": 0.89615154,
      "learning_rate": 3.999123849358836e-06,
      "loss": 0.91978294,
      "num_input_tokens_seen": 6948110,
      "step": 326,
      "time_per_iteration": 2.7900824546813965
    },
    {
      "auxiliary_loss_clip": 0.01313291,
      "auxiliary_loss_mlp": 0.01048588,
      "balance_loss_clip": 1.086285,
      "balance_loss_mlp": 1.03554666,
      "epoch": 0.03931942523898275,
      "flos": 25225414663680.0,
      "grad_norm": 2.4164799601521034,
      "language_loss": 0.75037265,
      "learning_rate": 3.999100642814418e-06,
      "loss": 0.77399147,
      "num_input_tokens_seen": 6968550,
      "step": 327,
      "time_per_iteration": 2.704773187637329
    },
    {
      "auxiliary_loss_clip": 0.013154,
      "auxiliary_loss_mlp": 0.01044787,
      "balance_loss_clip": 1.08862591,
      "balance_loss_mlp": 1.03237081,
      "epoch": 0.03943966812962184,
      "flos": 23257905240960.0,
      "grad_norm": 2.707176651369592,
      "language_loss": 0.88552624,
      "learning_rate": 3.999077133008022e-06,
      "loss": 0.90912813,
      "num_input_tokens_seen": 6987135,
      "step": 328,
      "time_per_iteration": 2.7529351711273193
    },
    {
      "auxiliary_loss_clip": 0.01319845,
      "auxiliary_loss_mlp": 0.01055489,
      "balance_loss_clip": 1.08997273,
      "balance_loss_mlp": 1.0429306,
      "epoch": 0.03955991102026093,
      "flos": 29168837291520.0,
      "grad_norm": 2.3616862390359246,
      "language_loss": 0.90758789,
      "learning_rate": 3.9990533199432145e-06,
      "loss": 0.93134123,
      "num_input_tokens_seen": 7008630,
      "step": 329,
      "time_per_iteration": 2.76733136177063
    },
    {
      "auxiliary_loss_clip": 0.01316448,
      "auxiliary_loss_mlp": 0.01057272,
      "balance_loss_clip": 1.08991146,
      "balance_loss_mlp": 1.04439723,
      "epoch": 0.03968015391090002,
      "flos": 17602441695360.0,
      "grad_norm": 3.9174584921312485,
      "language_loss": 0.75717127,
      "learning_rate": 3.999029203623608e-06,
      "loss": 0.78090847,
      "num_input_tokens_seen": 7026350,
      "step": 330,
      "time_per_iteration": 4.15687108039856
    },
    {
      "auxiliary_loss_clip": 0.01314787,
      "auxiliary_loss_mlp": 0.01055971,
      "balance_loss_clip": 1.0871383,
      "balance_loss_mlp": 1.04332924,
      "epoch": 0.03980039680153911,
      "flos": 21799285752960.0,
      "grad_norm": 2.3284742608390996,
      "language_loss": 0.87017417,
      "learning_rate": 3.99900478405286e-06,
      "loss": 0.8938818,
      "num_input_tokens_seen": 7045660,
      "step": 331,
      "time_per_iteration": 3.5975711345672607
    },
    {
      "auxiliary_loss_clip": 0.01318414,
      "auxiliary_loss_mlp": 0.01054761,
      "balance_loss_clip": 1.09129167,
      "balance_loss_mlp": 1.04277444,
      "epoch": 0.0399206396921782,
      "flos": 15195134148480.0,
      "grad_norm": 2.2779199972567574,
      "language_loss": 0.82399404,
      "learning_rate": 3.998980061234676e-06,
      "loss": 0.84772575,
      "num_input_tokens_seen": 7063575,
      "step": 332,
      "time_per_iteration": 2.6634323596954346
    },
    {
      "auxiliary_loss_clip": 0.01316697,
      "auxiliary_loss_mlp": 0.0104273,
      "balance_loss_clip": 1.08870077,
      "balance_loss_mlp": 1.03024232,
      "epoch": 0.040040882582817294,
      "flos": 14422910630400.0,
      "grad_norm": 2.4718793366970124,
      "language_loss": 0.75683868,
      "learning_rate": 3.9989550351728055e-06,
      "loss": 0.78043294,
      "num_input_tokens_seen": 7080505,
      "step": 333,
      "time_per_iteration": 3.695343017578125
    },
    {
      "auxiliary_loss_clip": 0.01313542,
      "auxiliary_loss_mlp": 0.01048339,
      "balance_loss_clip": 1.08797932,
      "balance_loss_mlp": 1.03585207,
      "epoch": 0.040161125473456384,
      "flos": 19280906375040.0,
      "grad_norm": 2.156655917441207,
      "language_loss": 0.84232414,
      "learning_rate": 3.998929705871046e-06,
      "loss": 0.86594296,
      "num_input_tokens_seen": 7097860,
      "step": 334,
      "time_per_iteration": 2.7350282669067383
    },
    {
      "auxiliary_loss_clip": 0.01311791,
      "auxiliary_loss_mlp": 0.01052154,
      "balance_loss_clip": 1.08555484,
      "balance_loss_mlp": 1.03998852,
      "epoch": 0.040281368364095474,
      "flos": 17821101738240.0,
      "grad_norm": 2.4995406128621256,
      "language_loss": 0.88937795,
      "learning_rate": 3.99890407333324e-06,
      "loss": 0.91301739,
      "num_input_tokens_seen": 7116390,
      "step": 335,
      "time_per_iteration": 2.6828529834747314
    },
    {
      "auxiliary_loss_clip": 0.01308052,
      "auxiliary_loss_mlp": 0.01045008,
      "balance_loss_clip": 1.08182216,
      "balance_loss_mlp": 1.03277719,
      "epoch": 0.040401611254734564,
      "flos": 19573757959680.0,
      "grad_norm": 1.8875575310225934,
      "language_loss": 0.87338573,
      "learning_rate": 3.998878137563275e-06,
      "loss": 0.89691627,
      "num_input_tokens_seen": 7135940,
      "step": 336,
      "time_per_iteration": 2.6861720085144043
    },
    {
      "auxiliary_loss_clip": 0.01311391,
      "auxiliary_loss_mlp": 0.01053512,
      "balance_loss_clip": 1.08543491,
      "balance_loss_mlp": 1.04143643,
      "epoch": 0.040521854145373654,
      "flos": 22054466949120.0,
      "grad_norm": 1.9172279696672405,
      "language_loss": 0.8518666,
      "learning_rate": 3.998851898565085e-06,
      "loss": 0.87551558,
      "num_input_tokens_seen": 7155745,
      "step": 337,
      "time_per_iteration": 2.71268367767334
    },
    {
      "auxiliary_loss_clip": 0.01312659,
      "auxiliary_loss_mlp": 0.01045296,
      "balance_loss_clip": 1.08573139,
      "balance_loss_mlp": 1.03326178,
      "epoch": 0.04064209703601274,
      "flos": 22674644196480.0,
      "grad_norm": 1.8772838517159163,
      "language_loss": 0.83173716,
      "learning_rate": 3.998825356342653e-06,
      "loss": 0.8553167,
      "num_input_tokens_seen": 7175920,
      "step": 338,
      "time_per_iteration": 2.7218480110168457
    },
    {
      "auxiliary_loss_clip": 0.01313424,
      "auxiliary_loss_mlp": 0.01046644,
      "balance_loss_clip": 1.08610535,
      "balance_loss_mlp": 1.03403807,
      "epoch": 0.04076233992665183,
      "flos": 38582172783360.0,
      "grad_norm": 3.2143832238226655,
      "language_loss": 0.73212987,
      "learning_rate": 3.998798510900003e-06,
      "loss": 0.75573051,
      "num_input_tokens_seen": 7198720,
      "step": 339,
      "time_per_iteration": 2.8556623458862305
    },
    {
      "auxiliary_loss_clip": 0.0131282,
      "auxiliary_loss_mlp": 0.01039789,
      "balance_loss_clip": 1.08606982,
      "balance_loss_mlp": 1.02761221,
      "epoch": 0.04088258281729093,
      "flos": 25885309374720.0,
      "grad_norm": 2.672803409064474,
      "language_loss": 0.8386302,
      "learning_rate": 3.998771362241207e-06,
      "loss": 0.86215627,
      "num_input_tokens_seen": 7219125,
      "step": 340,
      "time_per_iteration": 2.7353038787841797
    },
    {
      "auxiliary_loss_clip": 0.01309816,
      "auxiliary_loss_mlp": 0.01044002,
      "balance_loss_clip": 1.08534312,
      "balance_loss_mlp": 1.03214657,
      "epoch": 0.04100282570793002,
      "flos": 19789832223360.0,
      "grad_norm": 2.5329771268499077,
      "language_loss": 0.87972599,
      "learning_rate": 3.998743910370385e-06,
      "loss": 0.90326416,
      "num_input_tokens_seen": 7237985,
      "step": 341,
      "time_per_iteration": 2.7374048233032227
    },
    {
      "auxiliary_loss_clip": 0.01318598,
      "auxiliary_loss_mlp": 0.01057412,
      "balance_loss_clip": 1.09320378,
      "balance_loss_mlp": 1.04509795,
      "epoch": 0.04112306859856911,
      "flos": 22565152563840.0,
      "grad_norm": 2.925400086986793,
      "language_loss": 0.73227584,
      "learning_rate": 3.998716155291702e-06,
      "loss": 0.75603592,
      "num_input_tokens_seen": 7255825,
      "step": 342,
      "time_per_iteration": 2.730888605117798
    },
    {
      "auxiliary_loss_clip": 0.01314743,
      "auxiliary_loss_mlp": 0.01057503,
      "balance_loss_clip": 1.08908558,
      "balance_loss_mlp": 1.04496837,
      "epoch": 0.0412433114892082,
      "flos": 25040654081280.0,
      "grad_norm": 2.1299131893729264,
      "language_loss": 0.90687835,
      "learning_rate": 3.998688097009366e-06,
      "loss": 0.93060082,
      "num_input_tokens_seen": 7276590,
      "step": 343,
      "time_per_iteration": 2.763404130935669
    },
    {
      "auxiliary_loss_clip": 0.01316659,
      "auxiliary_loss_mlp": 0.01036381,
      "balance_loss_clip": 1.0895226,
      "balance_loss_mlp": 1.02436447,
      "epoch": 0.04136355437984729,
      "flos": 25191371548800.0,
      "grad_norm": 2.6055207203575557,
      "language_loss": 0.80188334,
      "learning_rate": 3.998659735527636e-06,
      "loss": 0.8254137,
      "num_input_tokens_seen": 7295680,
      "step": 344,
      "time_per_iteration": 2.7035250663757324
    },
    {
      "auxiliary_loss_clip": 0.01315583,
      "auxiliary_loss_mlp": 0.01060957,
      "balance_loss_clip": 1.08927393,
      "balance_loss_mlp": 1.04801083,
      "epoch": 0.04148379727048638,
      "flos": 22966777509120.0,
      "grad_norm": 2.652028115190664,
      "language_loss": 0.77530813,
      "learning_rate": 3.998631070850813e-06,
      "loss": 0.79907358,
      "num_input_tokens_seen": 7316300,
      "step": 345,
      "time_per_iteration": 2.774883508682251
    },
    {
      "auxiliary_loss_clip": 0.01315483,
      "auxiliary_loss_mlp": 0.01054911,
      "balance_loss_clip": 1.09038138,
      "balance_loss_mlp": 1.04254889,
      "epoch": 0.041604040161125476,
      "flos": 14063481187200.0,
      "grad_norm": 2.2797215700280002,
      "language_loss": 0.83475471,
      "learning_rate": 3.9986021029832455e-06,
      "loss": 0.85845864,
      "num_input_tokens_seen": 7333615,
      "step": 346,
      "time_per_iteration": 2.703885793685913
    },
    {
      "auxiliary_loss_clip": 0.0131579,
      "auxiliary_loss_mlp": 0.01045213,
      "balance_loss_clip": 1.08770394,
      "balance_loss_mlp": 1.03117597,
      "epoch": 0.041724283051764566,
      "flos": 12091877614080.0,
      "grad_norm": 3.030796941227288,
      "language_loss": 0.91599917,
      "learning_rate": 3.9985728319293285e-06,
      "loss": 0.93960923,
      "num_input_tokens_seen": 7347590,
      "step": 347,
      "time_per_iteration": 2.679015636444092
    },
    {
      "auxiliary_loss_clip": 0.01325816,
      "auxiliary_loss_mlp": 0.01039915,
      "balance_loss_clip": 1.09153771,
      "balance_loss_mlp": 1.02602744,
      "epoch": 0.041844525942403656,
      "flos": 12385303816320.0,
      "grad_norm": 3.8945768000662886,
      "language_loss": 0.85800624,
      "learning_rate": 3.998543257693501e-06,
      "loss": 0.88166356,
      "num_input_tokens_seen": 7364345,
      "step": 348,
      "time_per_iteration": 2.7280335426330566
    },
    {
      "auxiliary_loss_clip": 0.01315202,
      "auxiliary_loss_mlp": 0.01051962,
      "balance_loss_clip": 1.08983445,
      "balance_loss_mlp": 1.03850293,
      "epoch": 0.041964768833042745,
      "flos": 23769345041280.0,
      "grad_norm": 1.7167075645657504,
      "language_loss": 0.87996554,
      "learning_rate": 3.998513380280251e-06,
      "loss": 0.90363717,
      "num_input_tokens_seen": 7384625,
      "step": 349,
      "time_per_iteration": 2.739645481109619
    },
    {
      "auxiliary_loss_clip": 0.01322566,
      "auxiliary_loss_mlp": 0.01051439,
      "balance_loss_clip": 1.09183395,
      "balance_loss_mlp": 1.03821325,
      "epoch": 0.042085011723681835,
      "flos": 11875336473600.0,
      "grad_norm": 2.4076618909709055,
      "language_loss": 0.94625205,
      "learning_rate": 3.99848319969411e-06,
      "loss": 0.9699921,
      "num_input_tokens_seen": 7402225,
      "step": 350,
      "time_per_iteration": 2.692091703414917
    },
    {
      "auxiliary_loss_clip": 0.01324401,
      "auxiliary_loss_mlp": 0.01051765,
      "balance_loss_clip": 1.09357107,
      "balance_loss_mlp": 1.03833628,
      "epoch": 0.042205254614320925,
      "flos": 16873957964160.0,
      "grad_norm": 2.187281855772309,
      "language_loss": 0.79326916,
      "learning_rate": 3.9984527159396564e-06,
      "loss": 0.81703079,
      "num_input_tokens_seen": 7420865,
      "step": 351,
      "time_per_iteration": 2.7053182125091553
    },
    {
      "auxiliary_loss_clip": 0.01314505,
      "auxiliary_loss_mlp": 0.01047069,
      "balance_loss_clip": 1.08579278,
      "balance_loss_mlp": 1.03338408,
      "epoch": 0.04232549750496002,
      "flos": 25118508810240.0,
      "grad_norm": 11.98314049695068,
      "language_loss": 0.84565604,
      "learning_rate": 3.9984219290215154e-06,
      "loss": 0.86927176,
      "num_input_tokens_seen": 7441040,
      "step": 352,
      "time_per_iteration": 2.74955677986145
    },
    {
      "auxiliary_loss_clip": 0.01321811,
      "auxiliary_loss_mlp": 0.01045512,
      "balance_loss_clip": 1.09287691,
      "balance_loss_mlp": 1.03276253,
      "epoch": 0.04244574039559911,
      "flos": 26724541714560.0,
      "grad_norm": 1.7045985968687503,
      "language_loss": 0.89157057,
      "learning_rate": 3.998390838944356e-06,
      "loss": 0.9152438,
      "num_input_tokens_seen": 7462545,
      "step": 353,
      "time_per_iteration": 2.7461862564086914
    },
    {
      "auxiliary_loss_clip": 0.01318884,
      "auxiliary_loss_mlp": 0.01049363,
      "balance_loss_clip": 1.09108543,
      "balance_loss_mlp": 1.03678071,
      "epoch": 0.0425659832862382,
      "flos": 20923244951040.0,
      "grad_norm": 3.105722814466636,
      "language_loss": 0.90750313,
      "learning_rate": 3.998359445712895e-06,
      "loss": 0.9311856,
      "num_input_tokens_seen": 7481650,
      "step": 354,
      "time_per_iteration": 2.7680752277374268
    },
    {
      "auxiliary_loss_clip": 0.01317225,
      "auxiliary_loss_mlp": 0.01049417,
      "balance_loss_clip": 1.08790219,
      "balance_loss_mlp": 1.03445673,
      "epoch": 0.04268622617687729,
      "flos": 23331127115520.0,
      "grad_norm": 2.358702756349081,
      "language_loss": 0.81009972,
      "learning_rate": 3.9983277493318955e-06,
      "loss": 0.83376616,
      "num_input_tokens_seen": 7500945,
      "step": 355,
      "time_per_iteration": 2.726205587387085
    },
    {
      "auxiliary_loss_clip": 0.01319406,
      "auxiliary_loss_mlp": 0.01045485,
      "balance_loss_clip": 1.0886234,
      "balance_loss_mlp": 1.03158522,
      "epoch": 0.04280646906751638,
      "flos": 25994010908160.0,
      "grad_norm": 1.549749745689828,
      "language_loss": 0.81196773,
      "learning_rate": 3.998295749806165e-06,
      "loss": 0.83561671,
      "num_input_tokens_seen": 7522170,
      "step": 356,
      "time_per_iteration": 2.774038314819336
    },
    {
      "auxiliary_loss_clip": 0.0131737,
      "auxiliary_loss_mlp": 0.01051607,
      "balance_loss_clip": 1.09024775,
      "balance_loss_mlp": 1.03875613,
      "epoch": 0.04292671195815547,
      "flos": 26906824258560.0,
      "grad_norm": 1.7611119392803216,
      "language_loss": 0.83507997,
      "learning_rate": 3.998263447140558e-06,
      "loss": 0.85876971,
      "num_input_tokens_seen": 7542370,
      "step": 357,
      "time_per_iteration": 5.104188919067383
    },
    {
      "auxiliary_loss_clip": 0.01316185,
      "auxiliary_loss_mlp": 0.01046862,
      "balance_loss_clip": 1.08828366,
      "balance_loss_mlp": 1.03303373,
      "epoch": 0.04304695484879457,
      "flos": 39457315745280.0,
      "grad_norm": 1.811930726221024,
      "language_loss": 0.82047862,
      "learning_rate": 3.998230841339976e-06,
      "loss": 0.84410906,
      "num_input_tokens_seen": 7564380,
      "step": 358,
      "time_per_iteration": 3.7549777030944824
    },
    {
      "auxiliary_loss_clip": 0.01314138,
      "auxiliary_loss_mlp": 0.01049021,
      "balance_loss_clip": 1.08986652,
      "balance_loss_mlp": 1.03569937,
      "epoch": 0.04316719773943366,
      "flos": 19646297475840.0,
      "grad_norm": 2.2146817175965277,
      "language_loss": 0.85158539,
      "learning_rate": 3.998197932409363e-06,
      "loss": 0.87521684,
      "num_input_tokens_seen": 7582390,
      "step": 359,
      "time_per_iteration": 3.5658648014068604
    },
    {
      "auxiliary_loss_clip": 0.01313915,
      "auxiliary_loss_mlp": 0.01054772,
      "balance_loss_clip": 1.08690929,
      "balance_loss_mlp": 1.04215956,
      "epoch": 0.04328744063007275,
      "flos": 22452320966400.0,
      "grad_norm": 2.4827541839126783,
      "language_loss": 0.86081922,
      "learning_rate": 3.9981647203537125e-06,
      "loss": 0.88450605,
      "num_input_tokens_seen": 7599890,
      "step": 360,
      "time_per_iteration": 2.726011037826538
    },
    {
      "auxiliary_loss_clip": 0.01315585,
      "auxiliary_loss_mlp": 0.01046633,
      "balance_loss_clip": 1.08849406,
      "balance_loss_mlp": 1.03254795,
      "epoch": 0.04340768352071184,
      "flos": 21283033530240.0,
      "grad_norm": 2.0798969684212083,
      "language_loss": 0.96152538,
      "learning_rate": 3.998131205178063e-06,
      "loss": 0.98514754,
      "num_input_tokens_seen": 7618360,
      "step": 361,
      "time_per_iteration": 2.6479341983795166
    },
    {
      "auxiliary_loss_clip": 0.01320297,
      "auxiliary_loss_mlp": 0.01050561,
      "balance_loss_clip": 1.08982396,
      "balance_loss_mlp": 1.03768086,
      "epoch": 0.04352792641135093,
      "flos": 11583705951360.0,
      "grad_norm": 4.1648593971594945,
      "language_loss": 0.76298338,
      "learning_rate": 3.998097386887498e-06,
      "loss": 0.7866919,
      "num_input_tokens_seen": 7635435,
      "step": 362,
      "time_per_iteration": 2.6718921661376953
    },
    {
      "auxiliary_loss_clip": 0.01310542,
      "auxiliary_loss_mlp": 0.01059419,
      "balance_loss_clip": 1.08899951,
      "balance_loss_mlp": 1.04643083,
      "epoch": 0.04364816930199002,
      "flos": 23623547736960.0,
      "grad_norm": 1.843544634833622,
      "language_loss": 0.84847283,
      "learning_rate": 3.998063265487148e-06,
      "loss": 0.87217242,
      "num_input_tokens_seen": 7656485,
      "step": 363,
      "time_per_iteration": 2.730661630630493
    },
    {
      "auxiliary_loss_clip": 0.01314098,
      "auxiliary_loss_mlp": 0.01046706,
      "balance_loss_clip": 1.08850348,
      "balance_loss_mlp": 1.03391504,
      "epoch": 0.043768412192629114,
      "flos": 14429734214400.0,
      "grad_norm": 1.81722282330652,
      "language_loss": 0.8125121,
      "learning_rate": 3.99802884098219e-06,
      "loss": 0.83612013,
      "num_input_tokens_seen": 7674595,
      "step": 364,
      "time_per_iteration": 2.7239816188812256
    },
    {
      "auxiliary_loss_clip": 0.01308305,
      "auxiliary_loss_mlp": 0.01046016,
      "balance_loss_clip": 1.08349156,
      "balance_loss_mlp": 1.03339767,
      "epoch": 0.043888655083268203,
      "flos": 26468893641600.0,
      "grad_norm": 2.869313748247089,
      "language_loss": 0.82464725,
      "learning_rate": 3.997994113377845e-06,
      "loss": 0.84819049,
      "num_input_tokens_seen": 7693495,
      "step": 365,
      "time_per_iteration": 2.7568349838256836
    },
    {
      "auxiliary_loss_clip": 0.01318082,
      "auxiliary_loss_mlp": 0.01051789,
      "balance_loss_clip": 1.0910399,
      "balance_loss_mlp": 1.03711426,
      "epoch": 0.04400889797390729,
      "flos": 27235263242880.0,
      "grad_norm": 2.164494109200736,
      "language_loss": 0.83438289,
      "learning_rate": 3.9979590826793815e-06,
      "loss": 0.85808158,
      "num_input_tokens_seen": 7714685,
      "step": 366,
      "time_per_iteration": 2.746283769607544
    },
    {
      "auxiliary_loss_clip": 0.01313205,
      "auxiliary_loss_mlp": 0.01045387,
      "balance_loss_clip": 1.08795547,
      "balance_loss_mlp": 1.0310998,
      "epoch": 0.04412914086454638,
      "flos": 20119528183680.0,
      "grad_norm": 2.0672675736433423,
      "language_loss": 0.81181115,
      "learning_rate": 3.997923748892113e-06,
      "loss": 0.83539712,
      "num_input_tokens_seen": 7734005,
      "step": 367,
      "time_per_iteration": 2.7242374420166016
    },
    {
      "auxiliary_loss_clip": 0.01309647,
      "auxiliary_loss_mlp": 0.01054502,
      "balance_loss_clip": 1.0877862,
      "balance_loss_mlp": 1.04180598,
      "epoch": 0.04424938375518547,
      "flos": 22604618632320.0,
      "grad_norm": 1.6696271771260949,
      "language_loss": 0.88723159,
      "learning_rate": 3.9978881120214015e-06,
      "loss": 0.91087306,
      "num_input_tokens_seen": 7755525,
      "step": 368,
      "time_per_iteration": 2.7513535022735596
    },
    {
      "auxiliary_loss_clip": 0.01311458,
      "auxiliary_loss_mlp": 0.01056032,
      "balance_loss_clip": 1.08541811,
      "balance_loss_mlp": 1.04297245,
      "epoch": 0.04436962664582456,
      "flos": 24132365844480.0,
      "grad_norm": 2.1180917240140507,
      "language_loss": 0.79331195,
      "learning_rate": 3.997852172072652e-06,
      "loss": 0.81698686,
      "num_input_tokens_seen": 7776740,
      "step": 369,
      "time_per_iteration": 2.745258331298828
    },
    {
      "auxiliary_loss_clip": 0.01311061,
      "auxiliary_loss_mlp": 0.01059367,
      "balance_loss_clip": 1.08643055,
      "balance_loss_mlp": 1.04609931,
      "epoch": 0.04448986953646366,
      "flos": 18222906251520.0,
      "grad_norm": 2.5592079533481034,
      "language_loss": 0.89242375,
      "learning_rate": 3.9978159290513155e-06,
      "loss": 0.91612804,
      "num_input_tokens_seen": 7794820,
      "step": 370,
      "time_per_iteration": 2.6687264442443848
    },
    {
      "auxiliary_loss_clip": 0.01314967,
      "auxiliary_loss_mlp": 0.01047292,
      "balance_loss_clip": 1.08724999,
      "balance_loss_mlp": 1.03233123,
      "epoch": 0.04461011242710275,
      "flos": 30117920400000.0,
      "grad_norm": 1.780968279053194,
      "language_loss": 0.80012822,
      "learning_rate": 3.997779382962892e-06,
      "loss": 0.82375073,
      "num_input_tokens_seen": 7817705,
      "step": 371,
      "time_per_iteration": 2.8205490112304688
    },
    {
      "auxiliary_loss_clip": 0.01311867,
      "auxiliary_loss_mlp": 0.01056001,
      "balance_loss_clip": 1.08669138,
      "balance_loss_mlp": 1.04361486,
      "epoch": 0.04473035531774184,
      "flos": 29752529299200.0,
      "grad_norm": 1.7918301841376456,
      "language_loss": 0.73952609,
      "learning_rate": 3.997742533812924e-06,
      "loss": 0.76320481,
      "num_input_tokens_seen": 7840970,
      "step": 372,
      "time_per_iteration": 2.805116653442383
    },
    {
      "auxiliary_loss_clip": 0.01309313,
      "auxiliary_loss_mlp": 0.01054828,
      "balance_loss_clip": 1.08537269,
      "balance_loss_mlp": 1.04313397,
      "epoch": 0.04485059820838093,
      "flos": 13151565676800.0,
      "grad_norm": 3.3976960533425284,
      "language_loss": 0.92758977,
      "learning_rate": 3.997705381607001e-06,
      "loss": 0.95123124,
      "num_input_tokens_seen": 7857785,
      "step": 373,
      "time_per_iteration": 2.683513879776001
    },
    {
      "auxiliary_loss_clip": 0.01232293,
      "auxiliary_loss_mlp": 0.01045035,
      "balance_loss_clip": 1.07810855,
      "balance_loss_mlp": 1.03924131,
      "epoch": 0.04497084109902002,
      "flos": 68094209548800.0,
      "grad_norm": 0.9845909494433404,
      "language_loss": 0.60295713,
      "learning_rate": 3.997667926350761e-06,
      "loss": 0.6257304,
      "num_input_tokens_seen": 7916115,
      "step": 374,
      "time_per_iteration": 3.1811447143554688
    },
    {
      "auxiliary_loss_clip": 0.01231981,
      "auxiliary_loss_mlp": 0.01031396,
      "balance_loss_clip": 1.07860839,
      "balance_loss_mlp": 1.02603173,
      "epoch": 0.04509108398965911,
      "flos": 64342263346560.0,
      "grad_norm": 0.9005062009531848,
      "language_loss": 0.57775062,
      "learning_rate": 3.997630168049886e-06,
      "loss": 0.60038441,
      "num_input_tokens_seen": 7974480,
      "step": 375,
      "time_per_iteration": 3.276291847229004
    },
    {
      "auxiliary_loss_clip": 0.01315283,
      "auxiliary_loss_mlp": 0.01047712,
      "balance_loss_clip": 1.08896816,
      "balance_loss_mlp": 1.03400278,
      "epoch": 0.045211326880298205,
      "flos": 22271115830400.0,
      "grad_norm": 2.2470568470153527,
      "language_loss": 0.7748946,
      "learning_rate": 3.997592106710101e-06,
      "loss": 0.7985245,
      "num_input_tokens_seen": 7993940,
      "step": 376,
      "time_per_iteration": 2.7636470794677734
    },
    {
      "auxiliary_loss_clip": 0.01307117,
      "auxiliary_loss_mlp": 0.01049803,
      "balance_loss_clip": 1.08362293,
      "balance_loss_mlp": 1.0372268,
      "epoch": 0.045331569770937295,
      "flos": 32159441796480.0,
      "grad_norm": 2.460747961095386,
      "language_loss": 0.65821815,
      "learning_rate": 3.997553742337182e-06,
      "loss": 0.68178731,
      "num_input_tokens_seen": 8013365,
      "step": 377,
      "time_per_iteration": 2.7774670124053955
    },
    {
      "auxiliary_loss_clip": 0.01310685,
      "auxiliary_loss_mlp": 0.01052955,
      "balance_loss_clip": 1.08575654,
      "balance_loss_mlp": 1.03884637,
      "epoch": 0.045451812661576385,
      "flos": 22163455791360.0,
      "grad_norm": 3.1313931567018285,
      "language_loss": 0.91708398,
      "learning_rate": 3.997515074936949e-06,
      "loss": 0.94072044,
      "num_input_tokens_seen": 8034240,
      "step": 378,
      "time_per_iteration": 2.7017672061920166
    },
    {
      "auxiliary_loss_clip": 0.01308385,
      "auxiliary_loss_mlp": 0.01054446,
      "balance_loss_clip": 1.08645129,
      "balance_loss_mlp": 1.04257238,
      "epoch": 0.045572055552215475,
      "flos": 16581968305920.0,
      "grad_norm": 2.4086636288769863,
      "language_loss": 0.86998355,
      "learning_rate": 3.997476104515268e-06,
      "loss": 0.89361179,
      "num_input_tokens_seen": 8052430,
      "step": 379,
      "time_per_iteration": 2.669292449951172
    },
    {
      "auxiliary_loss_clip": 0.01304145,
      "auxiliary_loss_mlp": 0.01051311,
      "balance_loss_clip": 1.08414328,
      "balance_loss_mlp": 1.03856707,
      "epoch": 0.045692298442854565,
      "flos": 17603375448960.0,
      "grad_norm": 2.112666623422883,
      "language_loss": 0.77572364,
      "learning_rate": 3.9974368310780485e-06,
      "loss": 0.79927814,
      "num_input_tokens_seen": 8069605,
      "step": 380,
      "time_per_iteration": 2.7089169025421143
    },
    {
      "auxiliary_loss_clip": 0.0131717,
      "auxiliary_loss_mlp": 0.01063275,
      "balance_loss_clip": 1.08883786,
      "balance_loss_mlp": 1.04881513,
      "epoch": 0.045812541333493655,
      "flos": 26761098781440.0,
      "grad_norm": 3.1615173002974144,
      "language_loss": 0.74588084,
      "learning_rate": 3.997397254631251e-06,
      "loss": 0.76968527,
      "num_input_tokens_seen": 8090225,
      "step": 381,
      "time_per_iteration": 2.8167784214019775
    },
    {
      "auxiliary_loss_clip": 0.01221271,
      "auxiliary_loss_mlp": 0.01019251,
      "balance_loss_clip": 1.06980622,
      "balance_loss_mlp": 1.01422048,
      "epoch": 0.04593278422413275,
      "flos": 60250349894400.0,
      "grad_norm": 0.8324032917547493,
      "language_loss": 0.60016459,
      "learning_rate": 3.997357375180878e-06,
      "loss": 0.6225698,
      "num_input_tokens_seen": 8154505,
      "step": 382,
      "time_per_iteration": 4.652473211288452
    },
    {
      "auxiliary_loss_clip": 0.01306739,
      "auxiliary_loss_mlp": 0.01062215,
      "balance_loss_clip": 1.08298218,
      "balance_loss_mlp": 1.04863143,
      "epoch": 0.04605302711477184,
      "flos": 21799249839360.0,
      "grad_norm": 1.8856050327097973,
      "language_loss": 0.75321019,
      "learning_rate": 3.997317192732979e-06,
      "loss": 0.7768997,
      "num_input_tokens_seen": 8173285,
      "step": 383,
      "time_per_iteration": 2.7058777809143066
    },
    {
      "auxiliary_loss_clip": 0.01306282,
      "auxiliary_loss_mlp": 0.01053323,
      "balance_loss_clip": 1.08405638,
      "balance_loss_mlp": 1.04041219,
      "epoch": 0.04617327000541093,
      "flos": 19459705299840.0,
      "grad_norm": 1.8380701796371148,
      "language_loss": 0.82709926,
      "learning_rate": 3.99727670729365e-06,
      "loss": 0.85069525,
      "num_input_tokens_seen": 8191845,
      "step": 384,
      "time_per_iteration": 3.626425266265869
    },
    {
      "auxiliary_loss_clip": 0.01308823,
      "auxiliary_loss_mlp": 0.01046556,
      "balance_loss_clip": 1.08738935,
      "balance_loss_mlp": 1.03350842,
      "epoch": 0.04629351289605002,
      "flos": 25411468135680.0,
      "grad_norm": 1.8606068228668016,
      "language_loss": 0.78081447,
      "learning_rate": 3.997235918869033e-06,
      "loss": 0.80436826,
      "num_input_tokens_seen": 8212880,
      "step": 385,
      "time_per_iteration": 3.6179442405700684
    },
    {
      "auxiliary_loss_clip": 0.01306093,
      "auxiliary_loss_mlp": 0.01048206,
      "balance_loss_clip": 1.08484399,
      "balance_loss_mlp": 1.0359807,
      "epoch": 0.04641375578668911,
      "flos": 20558284813440.0,
      "grad_norm": 2.0906936425561584,
      "language_loss": 0.8266747,
      "learning_rate": 3.997194827465315e-06,
      "loss": 0.8502177,
      "num_input_tokens_seen": 8231475,
      "step": 386,
      "time_per_iteration": 2.7154734134674072
    },
    {
      "auxiliary_loss_clip": 0.01303595,
      "auxiliary_loss_mlp": 0.01053789,
      "balance_loss_clip": 1.08213067,
      "balance_loss_mlp": 1.04074192,
      "epoch": 0.0465339986773282,
      "flos": 13188661447680.0,
      "grad_norm": 3.124466949676271,
      "language_loss": 0.91347188,
      "learning_rate": 3.997153433088728e-06,
      "loss": 0.93704569,
      "num_input_tokens_seen": 8248600,
      "step": 387,
      "time_per_iteration": 2.8448987007141113
    },
    {
      "auxiliary_loss_clip": 0.0130874,
      "auxiliary_loss_mlp": 0.01043814,
      "balance_loss_clip": 1.08617735,
      "balance_loss_mlp": 1.031106,
      "epoch": 0.0466542415679673,
      "flos": 25556547168000.0,
      "grad_norm": 7.405105323086718,
      "language_loss": 0.81352603,
      "learning_rate": 3.997111735745554e-06,
      "loss": 0.83705151,
      "num_input_tokens_seen": 8271570,
      "step": 388,
      "time_per_iteration": 2.878868341445923
    },
    {
      "auxiliary_loss_clip": 0.01309227,
      "auxiliary_loss_mlp": 0.01052235,
      "balance_loss_clip": 1.08752251,
      "balance_loss_mlp": 1.03966391,
      "epoch": 0.04677448445860639,
      "flos": 22236749493120.0,
      "grad_norm": 2.162521827201954,
      "language_loss": 0.82643282,
      "learning_rate": 3.997069735442118e-06,
      "loss": 0.85004747,
      "num_input_tokens_seen": 8291265,
      "step": 389,
      "time_per_iteration": 2.826540231704712
    },
    {
      "auxiliary_loss_clip": 0.0130494,
      "auxiliary_loss_mlp": 0.0104528,
      "balance_loss_clip": 1.08380568,
      "balance_loss_mlp": 1.03313303,
      "epoch": 0.04689472734924548,
      "flos": 28147825198080.0,
      "grad_norm": 1.633044366981485,
      "language_loss": 0.80496359,
      "learning_rate": 3.997027432184792e-06,
      "loss": 0.82846582,
      "num_input_tokens_seen": 8315925,
      "step": 390,
      "time_per_iteration": 2.751988649368286
    },
    {
      "auxiliary_loss_clip": 0.01303805,
      "auxiliary_loss_mlp": 0.01065247,
      "balance_loss_clip": 1.08372092,
      "balance_loss_mlp": 1.05121648,
      "epoch": 0.04701497023988457,
      "flos": 23148952312320.0,
      "grad_norm": 1.8857219300919719,
      "language_loss": 0.89669561,
      "learning_rate": 3.99698482597999e-06,
      "loss": 0.9203862,
      "num_input_tokens_seen": 8333605,
      "step": 391,
      "time_per_iteration": 2.7682433128356934
    },
    {
      "auxiliary_loss_clip": 0.01211253,
      "auxiliary_loss_mlp": 0.01014835,
      "balance_loss_clip": 1.06170225,
      "balance_loss_mlp": 1.01020968,
      "epoch": 0.04713521313052366,
      "flos": 64827668764800.0,
      "grad_norm": 0.8914541389134945,
      "language_loss": 0.63909477,
      "learning_rate": 3.99694191683418e-06,
      "loss": 0.66135561,
      "num_input_tokens_seen": 8394405,
      "step": 392,
      "time_per_iteration": 3.2336103916168213
    },
    {
      "auxiliary_loss_clip": 0.01310399,
      "auxiliary_loss_mlp": 0.01053827,
      "balance_loss_clip": 1.08866251,
      "balance_loss_mlp": 1.04120898,
      "epoch": 0.047255456021162746,
      "flos": 18771585477120.0,
      "grad_norm": 2.326238302315817,
      "language_loss": 0.81915522,
      "learning_rate": 3.996898704753867e-06,
      "loss": 0.84279752,
      "num_input_tokens_seen": 8412355,
      "step": 393,
      "time_per_iteration": 2.7069883346557617
    },
    {
      "auxiliary_loss_clip": 0.01306452,
      "auxiliary_loss_mlp": 0.01052727,
      "balance_loss_clip": 1.0845263,
      "balance_loss_mlp": 1.03974509,
      "epoch": 0.04737569891180184,
      "flos": 22053820504320.0,
      "grad_norm": 2.2502587395445355,
      "language_loss": 0.87290937,
      "learning_rate": 3.996855189745609e-06,
      "loss": 0.89650112,
      "num_input_tokens_seen": 8431620,
      "step": 394,
      "time_per_iteration": 2.705063581466675
    },
    {
      "auxiliary_loss_clip": 0.0130144,
      "auxiliary_loss_mlp": 0.0105094,
      "balance_loss_clip": 1.08146942,
      "balance_loss_mlp": 1.03866744,
      "epoch": 0.04749594180244093,
      "flos": 29057370410880.0,
      "grad_norm": 2.0431820819673723,
      "language_loss": 0.92716944,
      "learning_rate": 3.996811371816007e-06,
      "loss": 0.95069331,
      "num_input_tokens_seen": 8454045,
      "step": 395,
      "time_per_iteration": 2.7023375034332275
    },
    {
      "auxiliary_loss_clip": 0.01303237,
      "auxiliary_loss_mlp": 0.01051454,
      "balance_loss_clip": 1.08438778,
      "balance_loss_mlp": 1.03881752,
      "epoch": 0.04761618469308002,
      "flos": 35112268172160.0,
      "grad_norm": 1.946860163461794,
      "language_loss": 0.78279829,
      "learning_rate": 3.996767250971707e-06,
      "loss": 0.80634522,
      "num_input_tokens_seen": 8476785,
      "step": 396,
      "time_per_iteration": 2.8344976902008057
    },
    {
      "auxiliary_loss_clip": 0.01301604,
      "auxiliary_loss_mlp": 0.01041184,
      "balance_loss_clip": 1.08351159,
      "balance_loss_mlp": 1.02909052,
      "epoch": 0.04773642758371911,
      "flos": 25630702796160.0,
      "grad_norm": 2.0423103192180565,
      "language_loss": 0.86945927,
      "learning_rate": 3.996722827219403e-06,
      "loss": 0.89288712,
      "num_input_tokens_seen": 8498400,
      "step": 397,
      "time_per_iteration": 2.7192254066467285
    },
    {
      "auxiliary_loss_clip": 0.01310648,
      "auxiliary_loss_mlp": 0.01053124,
      "balance_loss_clip": 1.08771288,
      "balance_loss_mlp": 1.04044056,
      "epoch": 0.0478566704743582,
      "flos": 20631506688000.0,
      "grad_norm": 4.045947074419687,
      "language_loss": 0.82541114,
      "learning_rate": 3.996678100565833e-06,
      "loss": 0.84904885,
      "num_input_tokens_seen": 8517455,
      "step": 398,
      "time_per_iteration": 2.669356107711792
    },
    {
      "auxiliary_loss_clip": 0.01301837,
      "auxiliary_loss_mlp": 0.01049948,
      "balance_loss_clip": 1.08380139,
      "balance_loss_mlp": 1.03848028,
      "epoch": 0.04797691336499729,
      "flos": 18835721210880.0,
      "grad_norm": 2.1652997572709913,
      "language_loss": 0.88584268,
      "learning_rate": 3.996633071017783e-06,
      "loss": 0.90936053,
      "num_input_tokens_seen": 8534085,
      "step": 399,
      "time_per_iteration": 2.671660900115967
    },
    {
      "auxiliary_loss_clip": 0.01302829,
      "auxiliary_loss_mlp": 0.01046403,
      "balance_loss_clip": 1.08495438,
      "balance_loss_mlp": 1.03416657,
      "epoch": 0.04809715625563638,
      "flos": 21099673578240.0,
      "grad_norm": 2.4506082369021764,
      "language_loss": 0.81972653,
      "learning_rate": 3.996587738582084e-06,
      "loss": 0.8432188,
      "num_input_tokens_seen": 8550885,
      "step": 400,
      "time_per_iteration": 2.7137138843536377
    },
    {
      "auxiliary_loss_clip": 0.01297889,
      "auxiliary_loss_mlp": 0.01049318,
      "balance_loss_clip": 1.08087182,
      "balance_loss_mlp": 1.03826094,
      "epoch": 0.04821739914627548,
      "flos": 23805650712960.0,
      "grad_norm": 2.47300786539538,
      "language_loss": 0.86476421,
      "learning_rate": 3.9965421032656115e-06,
      "loss": 0.88823628,
      "num_input_tokens_seen": 8570815,
      "step": 401,
      "time_per_iteration": 2.6535096168518066
    },
    {
      "auxiliary_loss_clip": 0.01302258,
      "auxiliary_loss_mlp": 0.01050644,
      "balance_loss_clip": 1.08380568,
      "balance_loss_mlp": 1.03796649,
      "epoch": 0.04833764203691457,
      "flos": 22200587475840.0,
      "grad_norm": 4.7235229006104005,
      "language_loss": 0.9414205,
      "learning_rate": 3.99649616507529e-06,
      "loss": 0.96494949,
      "num_input_tokens_seen": 8589910,
      "step": 402,
      "time_per_iteration": 2.6241989135742188
    },
    {
      "auxiliary_loss_clip": 0.0120153,
      "auxiliary_loss_mlp": 0.01007226,
      "balance_loss_clip": 1.05565655,
      "balance_loss_mlp": 1.00262499,
      "epoch": 0.04845788492755366,
      "flos": 65904376896000.0,
      "grad_norm": 0.8915303817992773,
      "language_loss": 0.63136744,
      "learning_rate": 3.996449924018088e-06,
      "loss": 0.65345502,
      "num_input_tokens_seen": 8650370,
      "step": 403,
      "time_per_iteration": 3.1479663848876953
    },
    {
      "auxiliary_loss_clip": 0.01301998,
      "auxiliary_loss_mlp": 0.01056076,
      "balance_loss_clip": 1.08253109,
      "balance_loss_mlp": 1.04414892,
      "epoch": 0.04857812781819275,
      "flos": 19281301424640.0,
      "grad_norm": 1.898238300085869,
      "language_loss": 0.797598,
      "learning_rate": 3.99640338010102e-06,
      "loss": 0.82117873,
      "num_input_tokens_seen": 8669475,
      "step": 404,
      "time_per_iteration": 2.6144745349884033
    },
    {
      "auxiliary_loss_clip": 0.01297295,
      "auxiliary_loss_mlp": 0.01048506,
      "balance_loss_clip": 1.07949197,
      "balance_loss_mlp": 1.03670478,
      "epoch": 0.04869837070883184,
      "flos": 24062376193920.0,
      "grad_norm": 3.852340539743921,
      "language_loss": 0.7872867,
      "learning_rate": 3.996356533331146e-06,
      "loss": 0.8107447,
      "num_input_tokens_seen": 8691345,
      "step": 405,
      "time_per_iteration": 2.659836769104004
    },
    {
      "auxiliary_loss_clip": 0.01307093,
      "auxiliary_loss_mlp": 0.01051027,
      "balance_loss_clip": 1.08239794,
      "balance_loss_mlp": 1.03844428,
      "epoch": 0.04881861359947093,
      "flos": 25187169657600.0,
      "grad_norm": 2.391331719155586,
      "language_loss": 0.61526769,
      "learning_rate": 3.996309383715573e-06,
      "loss": 0.6388489,
      "num_input_tokens_seen": 8710125,
      "step": 406,
      "time_per_iteration": 2.6600568294525146
    },
    {
      "auxiliary_loss_clip": 0.01301223,
      "auxiliary_loss_mlp": 0.0104623,
      "balance_loss_clip": 1.08176827,
      "balance_loss_mlp": 1.0333792,
      "epoch": 0.048938856490110025,
      "flos": 16362913213440.0,
      "grad_norm": 2.0598395705412864,
      "language_loss": 0.73861945,
      "learning_rate": 3.996261931261454e-06,
      "loss": 0.76209402,
      "num_input_tokens_seen": 8728705,
      "step": 407,
      "time_per_iteration": 2.648393392562866
    },
    {
      "auxiliary_loss_clip": 0.01305805,
      "auxiliary_loss_mlp": 0.01039101,
      "balance_loss_clip": 1.08628082,
      "balance_loss_mlp": 1.02610755,
      "epoch": 0.049059099380749115,
      "flos": 29895094379520.0,
      "grad_norm": 2.889779043895425,
      "language_loss": 0.8691566,
      "learning_rate": 3.996214175975987e-06,
      "loss": 0.89260566,
      "num_input_tokens_seen": 8749225,
      "step": 408,
      "time_per_iteration": 3.714207649230957
    },
    {
      "auxiliary_loss_clip": 0.01305654,
      "auxiliary_loss_mlp": 0.01053654,
      "balance_loss_clip": 1.08632755,
      "balance_loss_mlp": 1.04067779,
      "epoch": 0.049179342271388204,
      "flos": 35918858027520.0,
      "grad_norm": 2.0991015664106967,
      "language_loss": 0.79124856,
      "learning_rate": 3.996166117866417e-06,
      "loss": 0.81484163,
      "num_input_tokens_seen": 8771160,
      "step": 409,
      "time_per_iteration": 2.750786066055298
    },
    {
      "auxiliary_loss_clip": 0.01292757,
      "auxiliary_loss_mlp": 0.01058086,
      "balance_loss_clip": 1.0776751,
      "balance_loss_mlp": 1.04617131,
      "epoch": 0.049299585162027294,
      "flos": 14611226659200.0,
      "grad_norm": 1.96144513077882,
      "language_loss": 0.86918318,
      "learning_rate": 3.996117756940035e-06,
      "loss": 0.89269161,
      "num_input_tokens_seen": 8787845,
      "step": 410,
      "time_per_iteration": 2.665847063064575
    },
    {
      "auxiliary_loss_clip": 0.01297615,
      "auxiliary_loss_mlp": 0.01047372,
      "balance_loss_clip": 1.08052766,
      "balance_loss_mlp": 1.03495693,
      "epoch": 0.049419828052666384,
      "flos": 19567939956480.0,
      "grad_norm": 2.733931872251258,
      "language_loss": 0.97773069,
      "learning_rate": 3.996069093204175e-06,
      "loss": 1.00118065,
      "num_input_tokens_seen": 8803805,
      "step": 411,
      "time_per_iteration": 4.369104862213135
    },
    {
      "auxiliary_loss_clip": 0.01306674,
      "auxiliary_loss_mlp": 0.01047387,
      "balance_loss_clip": 1.08559394,
      "balance_loss_mlp": 1.03552532,
      "epoch": 0.049540070943305474,
      "flos": 13659916907520.0,
      "grad_norm": 2.5774324687069883,
      "language_loss": 0.88571924,
      "learning_rate": 3.996020126666221e-06,
      "loss": 0.9092598,
      "num_input_tokens_seen": 8820785,
      "step": 412,
      "time_per_iteration": 2.689587116241455
    },
    {
      "auxiliary_loss_clip": 0.01302033,
      "auxiliary_loss_mlp": 0.01061536,
      "balance_loss_clip": 1.08200908,
      "balance_loss_mlp": 1.04900694,
      "epoch": 0.04966031383394457,
      "flos": 21832035978240.0,
      "grad_norm": 1.9776795277651764,
      "language_loss": 0.82185179,
      "learning_rate": 3.995970857333601e-06,
      "loss": 0.84548748,
      "num_input_tokens_seen": 8841195,
      "step": 413,
      "time_per_iteration": 3.5829367637634277
    },
    {
      "auxiliary_loss_clip": 0.01303043,
      "auxiliary_loss_mlp": 0.01051007,
      "balance_loss_clip": 1.08392942,
      "balance_loss_mlp": 1.03863931,
      "epoch": 0.04978055672458366,
      "flos": 28618793349120.0,
      "grad_norm": 2.775821736277416,
      "language_loss": 0.79643464,
      "learning_rate": 3.995921285213789e-06,
      "loss": 0.81997514,
      "num_input_tokens_seen": 8861455,
      "step": 414,
      "time_per_iteration": 2.691967725753784
    },
    {
      "auxiliary_loss_clip": 0.01296874,
      "auxiliary_loss_mlp": 0.01048621,
      "balance_loss_clip": 1.0816195,
      "balance_loss_mlp": 1.03675365,
      "epoch": 0.04990079961522275,
      "flos": 19828220883840.0,
      "grad_norm": 2.5643306727287474,
      "language_loss": 0.8101064,
      "learning_rate": 3.995871410314305e-06,
      "loss": 0.83356136,
      "num_input_tokens_seen": 8880015,
      "step": 415,
      "time_per_iteration": 2.6686508655548096
    },
    {
      "auxiliary_loss_clip": 0.0124711,
      "auxiliary_loss_mlp": 0.01003843,
      "balance_loss_clip": 1.05279446,
      "balance_loss_mlp": 0.99933642,
      "epoch": 0.05002104250586184,
      "flos": 62735045293440.0,
      "grad_norm": 0.9044784405595301,
      "language_loss": 0.59629679,
      "learning_rate": 3.995821232642714e-06,
      "loss": 0.61880636,
      "num_input_tokens_seen": 8938420,
      "step": 416,
      "time_per_iteration": 3.3196654319763184
    },
    {
      "auxiliary_loss_clip": 0.0134773,
      "auxiliary_loss_mlp": 0.0104705,
      "balance_loss_clip": 1.08431292,
      "balance_loss_mlp": 1.03458691,
      "epoch": 0.05014128539650093,
      "flos": 27928518710400.0,
      "grad_norm": 6.105268390417708,
      "language_loss": 0.82708848,
      "learning_rate": 3.995770752206629e-06,
      "loss": 0.85103631,
      "num_input_tokens_seen": 8959495,
      "step": 417,
      "time_per_iteration": 2.7947630882263184
    },
    {
      "auxiliary_loss_clip": 0.01298887,
      "auxiliary_loss_mlp": 0.01047068,
      "balance_loss_clip": 1.08207691,
      "balance_loss_mlp": 1.03473592,
      "epoch": 0.05026152828714002,
      "flos": 17705576620800.0,
      "grad_norm": 2.1086486365012584,
      "language_loss": 0.97189355,
      "learning_rate": 3.995719969013709e-06,
      "loss": 0.9953531,
      "num_input_tokens_seen": 8976675,
      "step": 418,
      "time_per_iteration": 2.62555193901062
    },
    {
      "auxiliary_loss_clip": 0.0139337,
      "auxiliary_loss_mlp": 0.01038511,
      "balance_loss_clip": 1.07808256,
      "balance_loss_mlp": 1.0254221,
      "epoch": 0.05038177117777912,
      "flos": 19133277477120.0,
      "grad_norm": 3.3586376107421123,
      "language_loss": 0.86010313,
      "learning_rate": 3.995668883071655e-06,
      "loss": 0.88442194,
      "num_input_tokens_seen": 8992900,
      "step": 419,
      "time_per_iteration": 2.7219204902648926
    },
    {
      "auxiliary_loss_clip": 0.01302879,
      "auxiliary_loss_mlp": 0.01046034,
      "balance_loss_clip": 1.08350658,
      "balance_loss_mlp": 1.03411317,
      "epoch": 0.050502014068418206,
      "flos": 20667704618880.0,
      "grad_norm": 2.7030881434622533,
      "language_loss": 0.91208112,
      "learning_rate": 3.995617494388219e-06,
      "loss": 0.93557024,
      "num_input_tokens_seen": 9011020,
      "step": 420,
      "time_per_iteration": 2.6995599269866943
    },
    {
      "auxiliary_loss_clip": 0.01394773,
      "auxiliary_loss_mlp": 0.01051334,
      "balance_loss_clip": 1.07597494,
      "balance_loss_mlp": 1.03862,
      "epoch": 0.050622256959057296,
      "flos": 21361103740800.0,
      "grad_norm": 1.9127472907637553,
      "language_loss": 0.80652636,
      "learning_rate": 3.995565802971196e-06,
      "loss": 0.83098745,
      "num_input_tokens_seen": 9030995,
      "step": 421,
      "time_per_iteration": 2.7999706268310547
    },
    {
      "auxiliary_loss_clip": 0.01396202,
      "auxiliary_loss_mlp": 0.01042345,
      "balance_loss_clip": 1.07747638,
      "balance_loss_mlp": 1.03039432,
      "epoch": 0.050742499849696386,
      "flos": 27673588909440.0,
      "grad_norm": 1.8412309890190819,
      "language_loss": 0.67615855,
      "learning_rate": 3.995513808828427e-06,
      "loss": 0.700544,
      "num_input_tokens_seen": 9053790,
      "step": 422,
      "time_per_iteration": 2.8105592727661133
    },
    {
      "auxiliary_loss_clip": 0.01392158,
      "auxiliary_loss_mlp": 0.01050121,
      "balance_loss_clip": 1.07526004,
      "balance_loss_mlp": 1.03870106,
      "epoch": 0.050862742740335476,
      "flos": 19865999013120.0,
      "grad_norm": 1.8630515470147893,
      "language_loss": 0.76652235,
      "learning_rate": 3.9954615119678e-06,
      "loss": 0.79094517,
      "num_input_tokens_seen": 9072345,
      "step": 423,
      "time_per_iteration": 2.7337486743927
    },
    {
      "auxiliary_loss_clip": 0.01336708,
      "auxiliary_loss_mlp": 0.01045544,
      "balance_loss_clip": 1.07565689,
      "balance_loss_mlp": 1.03350353,
      "epoch": 0.050982985630974566,
      "flos": 22085098272000.0,
      "grad_norm": 1.9652330731101257,
      "language_loss": 0.80783916,
      "learning_rate": 3.995408912397248e-06,
      "loss": 0.8316617,
      "num_input_tokens_seen": 9090240,
      "step": 424,
      "time_per_iteration": 2.736022472381592
    },
    {
      "auxiliary_loss_clip": 0.01399785,
      "auxiliary_loss_mlp": 0.01049029,
      "balance_loss_clip": 1.08094311,
      "balance_loss_mlp": 1.03670907,
      "epoch": 0.05110322852161366,
      "flos": 20740962407040.0,
      "grad_norm": 2.2983619402252344,
      "language_loss": 0.93526286,
      "learning_rate": 3.99535601012475e-06,
      "loss": 0.95975101,
      "num_input_tokens_seen": 9105570,
      "step": 425,
      "time_per_iteration": 2.7143404483795166
    },
    {
      "auxiliary_loss_clip": 0.01436953,
      "auxiliary_loss_mlp": 0.02597362,
      "balance_loss_clip": 1.07382202,
      "balance_loss_mlp": 1.0003773,
      "epoch": 0.05122347141225275,
      "flos": 28547295327360.0,
      "grad_norm": 1.6750916803133802,
      "language_loss": 0.75875306,
      "learning_rate": 3.995302805158333e-06,
      "loss": 0.79909623,
      "num_input_tokens_seen": 9128225,
      "step": 426,
      "time_per_iteration": 2.8402016162872314
    },
    {
      "auxiliary_loss_clip": 0.01390382,
      "auxiliary_loss_mlp": 0.01054293,
      "balance_loss_clip": 1.07460713,
      "balance_loss_mlp": 1.04048872,
      "epoch": 0.05134371430289184,
      "flos": 19722679747200.0,
      "grad_norm": 1.7337709849598613,
      "language_loss": 0.83759689,
      "learning_rate": 3.9952492975060665e-06,
      "loss": 0.86204368,
      "num_input_tokens_seen": 9148295,
      "step": 427,
      "time_per_iteration": 2.780186176300049
    },
    {
      "auxiliary_loss_clip": 0.01341761,
      "auxiliary_loss_mlp": 0.010447,
      "balance_loss_clip": 1.07744753,
      "balance_loss_mlp": 1.0318495,
      "epoch": 0.05146395719353093,
      "flos": 34458945649920.0,
      "grad_norm": 6.024163243568287,
      "language_loss": 0.85691655,
      "learning_rate": 3.995195487176067e-06,
      "loss": 0.88078111,
      "num_input_tokens_seen": 9168525,
      "step": 428,
      "time_per_iteration": 2.801499843597412
    },
    {
      "auxiliary_loss_clip": 0.0129923,
      "auxiliary_loss_mlp": 0.01038534,
      "balance_loss_clip": 1.08121443,
      "balance_loss_mlp": 1.02676749,
      "epoch": 0.05158420008417002,
      "flos": 21760286561280.0,
      "grad_norm": 2.1027453238342937,
      "language_loss": 0.85790116,
      "learning_rate": 3.995141374176499e-06,
      "loss": 0.88127881,
      "num_input_tokens_seen": 9186920,
      "step": 429,
      "time_per_iteration": 2.669226884841919
    },
    {
      "auxiliary_loss_clip": 0.01342972,
      "auxiliary_loss_mlp": 0.0255204,
      "balance_loss_clip": 1.04525113,
      "balance_loss_mlp": 1.0001421,
      "epoch": 0.05170444297480911,
      "flos": 72553956226560.0,
      "grad_norm": 0.8714722011856999,
      "language_loss": 0.63073862,
      "learning_rate": 3.995086958515572e-06,
      "loss": 0.6696887,
      "num_input_tokens_seen": 9244940,
      "step": 430,
      "time_per_iteration": 3.4014806747436523
    },
    {
      "auxiliary_loss_clip": 0.01196332,
      "auxiliary_loss_mlp": 0.02551689,
      "balance_loss_clip": 1.0551213,
      "balance_loss_mlp": 1.00011826,
      "epoch": 0.05182468586544821,
      "flos": 62416159326720.0,
      "grad_norm": 0.8564554793007768,
      "language_loss": 0.59923142,
      "learning_rate": 3.995032240201538e-06,
      "loss": 0.63671154,
      "num_input_tokens_seen": 9307335,
      "step": 431,
      "time_per_iteration": 3.2329769134521484
    },
    {
      "auxiliary_loss_clip": 0.01302182,
      "auxiliary_loss_mlp": 0.01008717,
      "balance_loss_clip": 1.05245745,
      "balance_loss_mlp": 1.00435352,
      "epoch": 0.0519449287560873,
      "flos": 41225989432320.0,
      "grad_norm": 0.9414971054946931,
      "language_loss": 0.63128418,
      "learning_rate": 3.9949772192427e-06,
      "loss": 0.6543932,
      "num_input_tokens_seen": 9353960,
      "step": 432,
      "time_per_iteration": 3.008549451828003
    },
    {
      "auxiliary_loss_clip": 0.01394536,
      "auxiliary_loss_mlp": 0.01058504,
      "balance_loss_clip": 1.0753963,
      "balance_loss_mlp": 1.046386,
      "epoch": 0.05206517164672639,
      "flos": 17494530261120.0,
      "grad_norm": 1.8747065905286657,
      "language_loss": 0.79839253,
      "learning_rate": 3.994921895647405e-06,
      "loss": 0.82292295,
      "num_input_tokens_seen": 9372130,
      "step": 433,
      "time_per_iteration": 2.7775728702545166
    },
    {
      "auxiliary_loss_clip": 0.0119145,
      "auxiliary_loss_mlp": 0.01008196,
      "balance_loss_clip": 1.05189729,
      "balance_loss_mlp": 1.00397635,
      "epoch": 0.05218541453736548,
      "flos": 64002762973440.0,
      "grad_norm": 0.8407922208504447,
      "language_loss": 0.55314511,
      "learning_rate": 3.994866269424043e-06,
      "loss": 0.57514155,
      "num_input_tokens_seen": 9428500,
      "step": 434,
      "time_per_iteration": 4.159822463989258
    },
    {
      "auxiliary_loss_clip": 0.01537566,
      "auxiliary_loss_mlp": 0.01049466,
      "balance_loss_clip": 1.06101596,
      "balance_loss_mlp": 1.0377593,
      "epoch": 0.05230565742800457,
      "flos": 19317319787520.0,
      "grad_norm": 2.4313177801304517,
      "language_loss": 0.78066647,
      "learning_rate": 3.9948103405810545e-06,
      "loss": 0.80653679,
      "num_input_tokens_seen": 9447450,
      "step": 435,
      "time_per_iteration": 2.8475990295410156
    },
    {
      "auxiliary_loss_clip": 0.01419809,
      "auxiliary_loss_mlp": 0.01041245,
      "balance_loss_clip": 1.06692147,
      "balance_loss_mlp": 1.02863836,
      "epoch": 0.05242590031864366,
      "flos": 25298636538240.0,
      "grad_norm": 2.015964201738096,
      "language_loss": 0.86227846,
      "learning_rate": 3.994754109126923e-06,
      "loss": 0.88688898,
      "num_input_tokens_seen": 9468945,
      "step": 436,
      "time_per_iteration": 2.873560667037964
    },
    {
      "auxiliary_loss_clip": 0.01531019,
      "auxiliary_loss_mlp": 0.01040433,
      "balance_loss_clip": 1.06792402,
      "balance_loss_mlp": 1.02915621,
      "epoch": 0.052546143209282754,
      "flos": 26211629456640.0,
      "grad_norm": 1.82253190827884,
      "language_loss": 0.93660522,
      "learning_rate": 3.994697575070181e-06,
      "loss": 0.96231973,
      "num_input_tokens_seen": 9488405,
      "step": 437,
      "time_per_iteration": 4.026853561401367
    },
    {
      "auxiliary_loss_clip": 0.01389368,
      "auxiliary_loss_mlp": 0.01057408,
      "balance_loss_clip": 1.07791817,
      "balance_loss_mlp": 1.04564834,
      "epoch": 0.052666386099921844,
      "flos": 22158140578560.0,
      "grad_norm": 2.0829977259153076,
      "language_loss": 0.91440421,
      "learning_rate": 3.994640738419402e-06,
      "loss": 0.93887198,
      "num_input_tokens_seen": 9507780,
      "step": 438,
      "time_per_iteration": 3.8544328212738037
    },
    {
      "auxiliary_loss_clip": 0.01340916,
      "auxiliary_loss_mlp": 0.01046391,
      "balance_loss_clip": 1.0772779,
      "balance_loss_mlp": 1.03474414,
      "epoch": 0.052786628990560934,
      "flos": 23881817502720.0,
      "grad_norm": 1.9444326144038642,
      "language_loss": 0.80792636,
      "learning_rate": 3.9945835991832075e-06,
      "loss": 0.83179939,
      "num_input_tokens_seen": 9529665,
      "step": 439,
      "time_per_iteration": 3.6504065990448
    },
    {
      "auxiliary_loss_clip": 0.01294665,
      "auxiliary_loss_mlp": 0.01040761,
      "balance_loss_clip": 1.08163095,
      "balance_loss_mlp": 1.02906668,
      "epoch": 0.052906871881200024,
      "flos": 24605021934720.0,
      "grad_norm": 3.983418448540554,
      "language_loss": 0.92991138,
      "learning_rate": 3.994526157370268e-06,
      "loss": 0.95326567,
      "num_input_tokens_seen": 9548280,
      "step": 440,
      "time_per_iteration": 2.757430076599121
    },
    {
      "auxiliary_loss_clip": 0.01298935,
      "auxiliary_loss_mlp": 0.01016342,
      "balance_loss_clip": 1.04920578,
      "balance_loss_mlp": 1.01205051,
      "epoch": 0.053027114771839114,
      "flos": 56461631143680.0,
      "grad_norm": 0.8932555456730389,
      "language_loss": 0.59291697,
      "learning_rate": 3.994468412989296e-06,
      "loss": 0.61606967,
      "num_input_tokens_seen": 9609690,
      "step": 441,
      "time_per_iteration": 3.386768102645874
    },
    {
      "auxiliary_loss_clip": 0.01366553,
      "auxiliary_loss_mlp": 0.01042994,
      "balance_loss_clip": 1.06777358,
      "balance_loss_mlp": 1.03172231,
      "epoch": 0.053147357662478203,
      "flos": 17311098481920.0,
      "grad_norm": 2.158043807001391,
      "language_loss": 0.92804426,
      "learning_rate": 3.994410366049052e-06,
      "loss": 0.95213974,
      "num_input_tokens_seen": 9627550,
      "step": 442,
      "time_per_iteration": 2.7907450199127197
    },
    {
      "auxiliary_loss_clip": 0.0133812,
      "auxiliary_loss_mlp": 0.01042536,
      "balance_loss_clip": 1.07762647,
      "balance_loss_mlp": 1.03061485,
      "epoch": 0.0532676005531173,
      "flos": 17164977955200.0,
      "grad_norm": 2.9533715110317895,
      "language_loss": 0.82980627,
      "learning_rate": 3.994352016558341e-06,
      "loss": 0.85361284,
      "num_input_tokens_seen": 9644855,
      "step": 443,
      "time_per_iteration": 2.822378635406494
    },
    {
      "auxiliary_loss_clip": 0.01334747,
      "auxiliary_loss_mlp": 0.01048665,
      "balance_loss_clip": 1.07721233,
      "balance_loss_mlp": 1.03736424,
      "epoch": 0.05338784344375639,
      "flos": 27819960831360.0,
      "grad_norm": 1.8758465701518077,
      "language_loss": 0.73982,
      "learning_rate": 3.994293364526014e-06,
      "loss": 0.76365411,
      "num_input_tokens_seen": 9665740,
      "step": 444,
      "time_per_iteration": 2.7393693923950195
    },
    {
      "auxiliary_loss_clip": 0.01384783,
      "auxiliary_loss_mlp": 0.01046374,
      "balance_loss_clip": 1.07551718,
      "balance_loss_mlp": 1.03366649,
      "epoch": 0.05350808633439548,
      "flos": 21507691144320.0,
      "grad_norm": 2.471644391263642,
      "language_loss": 0.84749568,
      "learning_rate": 3.99423440996097e-06,
      "loss": 0.87180722,
      "num_input_tokens_seen": 9685280,
      "step": 445,
      "time_per_iteration": 2.788076162338257
    },
    {
      "auxiliary_loss_clip": 0.01398057,
      "auxiliary_loss_mlp": 0.01049958,
      "balance_loss_clip": 1.08285117,
      "balance_loss_mlp": 1.03889513,
      "epoch": 0.05362832922503457,
      "flos": 20084299920000.0,
      "grad_norm": 2.9121968788971593,
      "language_loss": 0.81373096,
      "learning_rate": 3.994175152872152e-06,
      "loss": 0.83821106,
      "num_input_tokens_seen": 9704365,
      "step": 446,
      "time_per_iteration": 2.7593419551849365
    },
    {
      "auxiliary_loss_clip": 0.01341608,
      "auxiliary_loss_mlp": 0.01049559,
      "balance_loss_clip": 1.07508349,
      "balance_loss_mlp": 1.03784704,
      "epoch": 0.05374857211567366,
      "flos": 26137222433280.0,
      "grad_norm": 2.39896849108379,
      "language_loss": 0.78849542,
      "learning_rate": 3.994115593268548e-06,
      "loss": 0.81240708,
      "num_input_tokens_seen": 9724145,
      "step": 447,
      "time_per_iteration": 2.788229465484619
    },
    {
      "auxiliary_loss_clip": 0.01296644,
      "auxiliary_loss_mlp": 0.01046651,
      "balance_loss_clip": 1.08274269,
      "balance_loss_mlp": 1.03444433,
      "epoch": 0.05386881500631275,
      "flos": 27486817165440.0,
      "grad_norm": 2.026108188713631,
      "language_loss": 0.82509351,
      "learning_rate": 3.994055731159195e-06,
      "loss": 0.84852648,
      "num_input_tokens_seen": 9741615,
      "step": 448,
      "time_per_iteration": 2.7141990661621094
    },
    {
      "auxiliary_loss_clip": 0.01342537,
      "auxiliary_loss_mlp": 0.01056244,
      "balance_loss_clip": 1.08059764,
      "balance_loss_mlp": 1.04547346,
      "epoch": 0.053989057896951846,
      "flos": 23585087249280.0,
      "grad_norm": 1.949897700982153,
      "language_loss": 0.86970377,
      "learning_rate": 3.993995566553172e-06,
      "loss": 0.89369166,
      "num_input_tokens_seen": 9760580,
      "step": 449,
      "time_per_iteration": 2.78975248336792
    },
    {
      "auxiliary_loss_clip": 0.01377909,
      "auxiliary_loss_mlp": 0.0104596,
      "balance_loss_clip": 1.06948781,
      "balance_loss_mlp": 1.03417647,
      "epoch": 0.054109300787590936,
      "flos": 25228862369280.0,
      "grad_norm": 1.6208445232214201,
      "language_loss": 0.77037418,
      "learning_rate": 3.993935099459607e-06,
      "loss": 0.79461288,
      "num_input_tokens_seen": 9782195,
      "step": 450,
      "time_per_iteration": 2.7688324451446533
    },
    {
      "auxiliary_loss_clip": 0.01285195,
      "auxiliary_loss_mlp": 0.01041679,
      "balance_loss_clip": 1.07643306,
      "balance_loss_mlp": 1.03062201,
      "epoch": 0.054229543678230026,
      "flos": 23841525421440.0,
      "grad_norm": 2.006765565759573,
      "language_loss": 0.74087507,
      "learning_rate": 3.993874329887673e-06,
      "loss": 0.76414382,
      "num_input_tokens_seen": 9800850,
      "step": 451,
      "time_per_iteration": 2.743246555328369
    },
    {
      "auxiliary_loss_clip": 0.01340068,
      "auxiliary_loss_mlp": 0.01042565,
      "balance_loss_clip": 1.07964218,
      "balance_loss_mlp": 1.03103185,
      "epoch": 0.054349786568869116,
      "flos": 16320933192960.0,
      "grad_norm": 2.4071449574114636,
      "language_loss": 0.86846983,
      "learning_rate": 3.993813257846589e-06,
      "loss": 0.8922962,
      "num_input_tokens_seen": 9817605,
      "step": 452,
      "time_per_iteration": 2.733025074005127
    },
    {
      "auxiliary_loss_clip": 0.0134501,
      "auxiliary_loss_mlp": 0.01042654,
      "balance_loss_clip": 1.08063376,
      "balance_loss_mlp": 1.03079247,
      "epoch": 0.054470029459508205,
      "flos": 18660729127680.0,
      "grad_norm": 2.2943209594250926,
      "language_loss": 0.92818052,
      "learning_rate": 3.993751883345619e-06,
      "loss": 0.95205712,
      "num_input_tokens_seen": 9835965,
      "step": 453,
      "time_per_iteration": 2.735626459121704
    },
    {
      "auxiliary_loss_clip": 0.01384718,
      "auxiliary_loss_mlp": 0.01041515,
      "balance_loss_clip": 1.07490015,
      "balance_loss_mlp": 1.02978504,
      "epoch": 0.054590272350147295,
      "flos": 17785298856960.0,
      "grad_norm": 2.4483402481110863,
      "language_loss": 0.87466812,
      "learning_rate": 3.993690206394073e-06,
      "loss": 0.89893043,
      "num_input_tokens_seen": 9852265,
      "step": 454,
      "time_per_iteration": 2.780975103378296
    },
    {
      "auxiliary_loss_clip": 0.01393323,
      "auxiliary_loss_mlp": 0.01061858,
      "balance_loss_clip": 1.07783532,
      "balance_loss_mlp": 1.05010366,
      "epoch": 0.054710515240786385,
      "flos": 17785945301760.0,
      "grad_norm": 3.149653630663544,
      "language_loss": 0.87755013,
      "learning_rate": 3.993628227001307e-06,
      "loss": 0.90210193,
      "num_input_tokens_seen": 9870465,
      "step": 455,
      "time_per_iteration": 2.7956645488739014
    },
    {
      "auxiliary_loss_clip": 0.01389681,
      "auxiliary_loss_mlp": 0.01046148,
      "balance_loss_clip": 1.07415664,
      "balance_loss_mlp": 1.03465593,
      "epoch": 0.05483075813142548,
      "flos": 48210900180480.0,
      "grad_norm": 2.065949763334928,
      "language_loss": 0.71102798,
      "learning_rate": 3.993565945176726e-06,
      "loss": 0.73538625,
      "num_input_tokens_seen": 9891490,
      "step": 456,
      "time_per_iteration": 3.014958381652832
    },
    {
      "auxiliary_loss_clip": 0.01390092,
      "auxiliary_loss_mlp": 0.01043874,
      "balance_loss_clip": 1.07963181,
      "balance_loss_mlp": 1.03284085,
      "epoch": 0.05495100102206457,
      "flos": 19682244011520.0,
      "grad_norm": 3.712574980724963,
      "language_loss": 0.83908689,
      "learning_rate": 3.993503360929776e-06,
      "loss": 0.86342657,
      "num_input_tokens_seen": 9910375,
      "step": 457,
      "time_per_iteration": 2.722482204437256
    },
    {
      "auxiliary_loss_clip": 0.01579178,
      "auxiliary_loss_mlp": 0.01046989,
      "balance_loss_clip": 1.0671066,
      "balance_loss_mlp": 1.0338757,
      "epoch": 0.05507124391270366,
      "flos": 26360048453760.0,
      "grad_norm": 1.8679573968181775,
      "language_loss": 0.81036043,
      "learning_rate": 3.99344047426995e-06,
      "loss": 0.83662212,
      "num_input_tokens_seen": 9931635,
      "step": 458,
      "time_per_iteration": 3.264531135559082
    },
    {
      "auxiliary_loss_clip": 0.01482147,
      "auxiliary_loss_mlp": 0.01050426,
      "balance_loss_clip": 1.06989145,
      "balance_loss_mlp": 1.03733075,
      "epoch": 0.05519148680334275,
      "flos": 22601314581120.0,
      "grad_norm": 2.2804849320614453,
      "language_loss": 0.93377417,
      "learning_rate": 3.993377285206789e-06,
      "loss": 0.95909989,
      "num_input_tokens_seen": 9951420,
      "step": 459,
      "time_per_iteration": 4.1829447746276855
    },
    {
      "auxiliary_loss_clip": 0.01410458,
      "auxiliary_loss_mlp": 0.0104385,
      "balance_loss_clip": 1.06527376,
      "balance_loss_mlp": 1.03159487,
      "epoch": 0.05531172969398184,
      "flos": 40552519380480.0,
      "grad_norm": 2.0233703894585577,
      "language_loss": 0.86709929,
      "learning_rate": 3.99331379374988e-06,
      "loss": 0.89164239,
      "num_input_tokens_seen": 9975025,
      "step": 460,
      "time_per_iteration": 3.1145362854003906
    },
    {
      "auxiliary_loss_clip": 0.01383109,
      "auxiliary_loss_mlp": 0.01045914,
      "balance_loss_clip": 1.0669744,
      "balance_loss_mlp": 1.03407037,
      "epoch": 0.05543197258462093,
      "flos": 23477894087040.0,
      "grad_norm": 1.9690926836623683,
      "language_loss": 0.80230081,
      "learning_rate": 3.993249999908852e-06,
      "loss": 0.82659101,
      "num_input_tokens_seen": 9995175,
      "step": 461,
      "time_per_iteration": 2.9279301166534424
    },
    {
      "auxiliary_loss_clip": 0.01290579,
      "auxiliary_loss_mlp": 0.0104518,
      "balance_loss_clip": 1.07807219,
      "balance_loss_mlp": 1.03398657,
      "epoch": 0.05555221547526003,
      "flos": 18624603024000.0,
      "grad_norm": 6.29333754228556,
      "language_loss": 0.87430573,
      "learning_rate": 3.993185903693384e-06,
      "loss": 0.8976633,
      "num_input_tokens_seen": 10011975,
      "step": 462,
      "time_per_iteration": 2.8366644382476807
    },
    {
      "auxiliary_loss_clip": 0.01385726,
      "auxiliary_loss_mlp": 0.01040281,
      "balance_loss_clip": 1.07429326,
      "balance_loss_mlp": 1.02858675,
      "epoch": 0.05567245836589912,
      "flos": 23587098410880.0,
      "grad_norm": 2.73463022450099,
      "language_loss": 0.82378399,
      "learning_rate": 3.9931215051131995e-06,
      "loss": 0.8480441,
      "num_input_tokens_seen": 10032620,
      "step": 463,
      "time_per_iteration": 4.6471028327941895
    },
    {
      "auxiliary_loss_clip": 0.01395373,
      "auxiliary_loss_mlp": 0.0104778,
      "balance_loss_clip": 1.07382774,
      "balance_loss_mlp": 1.03647256,
      "epoch": 0.05579270125653821,
      "flos": 27746667129600.0,
      "grad_norm": 2.0088925792448338,
      "language_loss": 0.80269474,
      "learning_rate": 3.993056804178068e-06,
      "loss": 0.82712626,
      "num_input_tokens_seen": 10054165,
      "step": 464,
      "time_per_iteration": 2.782848834991455
    },
    {
      "auxiliary_loss_clip": 0.014867,
      "auxiliary_loss_mlp": 0.01048236,
      "balance_loss_clip": 1.07388723,
      "balance_loss_mlp": 1.03719091,
      "epoch": 0.0559129441471773,
      "flos": 27014161075200.0,
      "grad_norm": 2.123317855320974,
      "language_loss": 0.84404427,
      "learning_rate": 3.992991800897803e-06,
      "loss": 0.86939359,
      "num_input_tokens_seen": 10073970,
      "step": 465,
      "time_per_iteration": 3.7424652576446533
    },
    {
      "auxiliary_loss_clip": 0.01291301,
      "auxiliary_loss_mlp": 0.01046513,
      "balance_loss_clip": 1.07906985,
      "balance_loss_mlp": 1.03517032,
      "epoch": 0.05603318703781639,
      "flos": 15229787794560.0,
      "grad_norm": 2.5701531458329567,
      "language_loss": 0.89960182,
      "learning_rate": 3.9929264952822665e-06,
      "loss": 0.92297995,
      "num_input_tokens_seen": 10091505,
      "step": 466,
      "time_per_iteration": 2.628903388977051
    },
    {
      "auxiliary_loss_clip": 0.01341056,
      "auxiliary_loss_mlp": 0.01044613,
      "balance_loss_clip": 1.07647586,
      "balance_loss_mlp": 1.03302598,
      "epoch": 0.05615342992845548,
      "flos": 22266482976000.0,
      "grad_norm": 3.948731737540374,
      "language_loss": 0.88446224,
      "learning_rate": 3.992860887341366e-06,
      "loss": 0.908319,
      "num_input_tokens_seen": 10109675,
      "step": 467,
      "time_per_iteration": 2.730656147003174
    },
    {
      "auxiliary_loss_clip": 0.01421514,
      "auxiliary_loss_mlp": 0.01041206,
      "balance_loss_clip": 1.06796205,
      "balance_loss_mlp": 1.03010714,
      "epoch": 0.056273672819094574,
      "flos": 23584979508480.0,
      "grad_norm": 2.5359429892669207,
      "language_loss": 0.81432533,
      "learning_rate": 3.992794977085052e-06,
      "loss": 0.83895254,
      "num_input_tokens_seen": 10127675,
      "step": 468,
      "time_per_iteration": 2.821216344833374
    },
    {
      "auxiliary_loss_clip": 0.01437074,
      "auxiliary_loss_mlp": 0.01041455,
      "balance_loss_clip": 1.07284641,
      "balance_loss_mlp": 1.02937365,
      "epoch": 0.056393915709733664,
      "flos": 19858708552320.0,
      "grad_norm": 2.141113394266978,
      "language_loss": 0.84916866,
      "learning_rate": 3.992728764523326e-06,
      "loss": 0.873954,
      "num_input_tokens_seen": 10146620,
      "step": 469,
      "time_per_iteration": 2.899751663208008
    },
    {
      "auxiliary_loss_clip": 0.0138808,
      "auxiliary_loss_mlp": 0.01043418,
      "balance_loss_clip": 1.07883132,
      "balance_loss_mlp": 1.0317595,
      "epoch": 0.05651415860037275,
      "flos": 22163779013760.0,
      "grad_norm": 1.7982611965769124,
      "language_loss": 0.80770737,
      "learning_rate": 3.99266224966623e-06,
      "loss": 0.83202237,
      "num_input_tokens_seen": 10167535,
      "step": 470,
      "time_per_iteration": 2.778958320617676
    },
    {
      "auxiliary_loss_clip": 0.01373724,
      "auxiliary_loss_mlp": 0.01046484,
      "balance_loss_clip": 1.07433283,
      "balance_loss_mlp": 1.03508735,
      "epoch": 0.05663440149101184,
      "flos": 19463548055040.0,
      "grad_norm": 2.172366940929174,
      "language_loss": 0.88026309,
      "learning_rate": 3.992595432523855e-06,
      "loss": 0.90446514,
      "num_input_tokens_seen": 10184825,
      "step": 471,
      "time_per_iteration": 2.794980049133301
    },
    {
      "auxiliary_loss_clip": 0.01423649,
      "auxiliary_loss_mlp": 0.01044927,
      "balance_loss_clip": 1.07352173,
      "balance_loss_mlp": 1.03448462,
      "epoch": 0.05675464438165093,
      "flos": 22670226823680.0,
      "grad_norm": 2.1498525691929626,
      "language_loss": 0.85855269,
      "learning_rate": 3.992528313106338e-06,
      "loss": 0.88323843,
      "num_input_tokens_seen": 10203025,
      "step": 472,
      "time_per_iteration": 2.8148341178894043
    },
    {
      "auxiliary_loss_clip": 0.01289423,
      "auxiliary_loss_mlp": 0.02589687,
      "balance_loss_clip": 1.08206439,
      "balance_loss_mlp": 1.00020552,
      "epoch": 0.05687488727229002,
      "flos": 16901177495040.0,
      "grad_norm": 2.468596090875915,
      "language_loss": 0.82414341,
      "learning_rate": 3.9924608914238595e-06,
      "loss": 0.86293447,
      "num_input_tokens_seen": 10218020,
      "step": 473,
      "time_per_iteration": 2.6749465465545654
    },
    {
      "auxiliary_loss_clip": 0.01338408,
      "auxiliary_loss_mlp": 0.01045723,
      "balance_loss_clip": 1.07976532,
      "balance_loss_mlp": 1.03454757,
      "epoch": 0.05699513016292912,
      "flos": 29168980945920.0,
      "grad_norm": 2.181273164539161,
      "language_loss": 0.83925503,
      "learning_rate": 3.992393167486648e-06,
      "loss": 0.86309636,
      "num_input_tokens_seen": 10237170,
      "step": 474,
      "time_per_iteration": 2.8562870025634766
    },
    {
      "auxiliary_loss_clip": 0.012906,
      "auxiliary_loss_mlp": 0.01044744,
      "balance_loss_clip": 1.08134127,
      "balance_loss_mlp": 1.03306186,
      "epoch": 0.05711537305356821,
      "flos": 18916197632640.0,
      "grad_norm": 3.0803289129230014,
      "language_loss": 0.80859125,
      "learning_rate": 3.992325141304977e-06,
      "loss": 0.8319447,
      "num_input_tokens_seen": 10255125,
      "step": 475,
      "time_per_iteration": 2.679352045059204
    },
    {
      "auxiliary_loss_clip": 0.0141644,
      "auxiliary_loss_mlp": 0.01039482,
      "balance_loss_clip": 1.06840444,
      "balance_loss_mlp": 1.02889669,
      "epoch": 0.0572356159442073,
      "flos": 26758979879040.0,
      "grad_norm": 2.236982242254772,
      "language_loss": 0.86659515,
      "learning_rate": 3.992256812889166e-06,
      "loss": 0.89115435,
      "num_input_tokens_seen": 10271230,
      "step": 476,
      "time_per_iteration": 2.824808120727539
    },
    {
      "auxiliary_loss_clip": 0.01291539,
      "auxiliary_loss_mlp": 0.01053718,
      "balance_loss_clip": 1.0834769,
      "balance_loss_mlp": 1.04252386,
      "epoch": 0.05735585883484639,
      "flos": 35116146840960.0,
      "grad_norm": 2.3015394024671534,
      "language_loss": 0.76655966,
      "learning_rate": 3.992188182249582e-06,
      "loss": 0.79001224,
      "num_input_tokens_seen": 10293125,
      "step": 477,
      "time_per_iteration": 2.855139970779419
    },
    {
      "auxiliary_loss_clip": 0.01389415,
      "auxiliary_loss_mlp": 0.01046667,
      "balance_loss_clip": 1.08053946,
      "balance_loss_mlp": 1.03440642,
      "epoch": 0.05747610172548548,
      "flos": 18734381965440.0,
      "grad_norm": 2.9884243216618267,
      "language_loss": 0.90674686,
      "learning_rate": 3.992119249396633e-06,
      "loss": 0.93110776,
      "num_input_tokens_seen": 10311810,
      "step": 478,
      "time_per_iteration": 2.724443197250366
    },
    {
      "auxiliary_loss_clip": 0.0137713,
      "auxiliary_loss_mlp": 0.02592626,
      "balance_loss_clip": 1.07214403,
      "balance_loss_mlp": 1.00026309,
      "epoch": 0.05759634461612457,
      "flos": 27964752554880.0,
      "grad_norm": 1.9109725623678235,
      "language_loss": 0.82141632,
      "learning_rate": 3.992050014340778e-06,
      "loss": 0.86111391,
      "num_input_tokens_seen": 10332165,
      "step": 479,
      "time_per_iteration": 2.8482606410980225
    },
    {
      "auxiliary_loss_clip": 0.01234104,
      "auxiliary_loss_mlp": 0.01012394,
      "balance_loss_clip": 1.05112767,
      "balance_loss_mlp": 1.0084362,
      "epoch": 0.057716587506763666,
      "flos": 69292009405440.0,
      "grad_norm": 0.831231748248587,
      "language_loss": 0.54982752,
      "learning_rate": 3.99198047709252e-06,
      "loss": 0.57229251,
      "num_input_tokens_seen": 10393685,
      "step": 480,
      "time_per_iteration": 3.3505139350891113
    },
    {
      "auxiliary_loss_clip": 0.01422743,
      "auxiliary_loss_mlp": 0.01051564,
      "balance_loss_clip": 1.06864119,
      "balance_loss_mlp": 1.03974462,
      "epoch": 0.057836830397402755,
      "flos": 25009196745600.0,
      "grad_norm": 1.785208231763675,
      "language_loss": 0.78725946,
      "learning_rate": 3.991910637662408e-06,
      "loss": 0.81200254,
      "num_input_tokens_seen": 10413975,
      "step": 481,
      "time_per_iteration": 2.808192253112793
    },
    {
      "auxiliary_loss_clip": 0.0128449,
      "auxiliary_loss_mlp": 0.01047578,
      "balance_loss_clip": 1.0779866,
      "balance_loss_mlp": 1.03674781,
      "epoch": 0.057957073288041845,
      "flos": 25593894334080.0,
      "grad_norm": 1.821964399784007,
      "language_loss": 0.80953604,
      "learning_rate": 3.9918404960610355e-06,
      "loss": 0.83285677,
      "num_input_tokens_seen": 10433005,
      "step": 482,
      "time_per_iteration": 2.778062105178833
    },
    {
      "auxiliary_loss_clip": 0.01340844,
      "auxiliary_loss_mlp": 0.01039604,
      "balance_loss_clip": 1.07803798,
      "balance_loss_mlp": 1.02808785,
      "epoch": 0.058077316178680935,
      "flos": 20777411733120.0,
      "grad_norm": 2.4630790653793975,
      "language_loss": 0.77438307,
      "learning_rate": 3.991770052299043e-06,
      "loss": 0.79818749,
      "num_input_tokens_seen": 10451235,
      "step": 483,
      "time_per_iteration": 2.742734432220459
    },
    {
      "auxiliary_loss_clip": 0.01388433,
      "auxiliary_loss_mlp": 0.01049127,
      "balance_loss_clip": 1.07291698,
      "balance_loss_mlp": 1.03792763,
      "epoch": 0.058197559069320025,
      "flos": 18916484941440.0,
      "grad_norm": 2.4682943087630167,
      "language_loss": 0.87775803,
      "learning_rate": 3.991699306387118e-06,
      "loss": 0.9021337,
      "num_input_tokens_seen": 10469705,
      "step": 484,
      "time_per_iteration": 2.773005485534668
    },
    {
      "auxiliary_loss_clip": 0.01338138,
      "auxiliary_loss_mlp": 0.0104326,
      "balance_loss_clip": 1.07489038,
      "balance_loss_mlp": 1.03263259,
      "epoch": 0.058317801959959115,
      "flos": 24863327614080.0,
      "grad_norm": 1.927129271754468,
      "language_loss": 0.78184831,
      "learning_rate": 3.991628258335991e-06,
      "loss": 0.80566227,
      "num_input_tokens_seen": 10491910,
      "step": 485,
      "time_per_iteration": 3.9529383182525635
    },
    {
      "auxiliary_loss_clip": 0.01421832,
      "auxiliary_loss_mlp": 0.01033799,
      "balance_loss_clip": 1.06895614,
      "balance_loss_mlp": 1.02245069,
      "epoch": 0.05843804485059821,
      "flos": 23257977068160.0,
      "grad_norm": 2.9437093109542296,
      "language_loss": 0.8764087,
      "learning_rate": 3.991556908156442e-06,
      "loss": 0.90096503,
      "num_input_tokens_seen": 10508435,
      "step": 486,
      "time_per_iteration": 2.85764479637146
    },
    {
      "auxiliary_loss_clip": 0.01386267,
      "auxiliary_loss_mlp": 0.01042802,
      "balance_loss_clip": 1.07248592,
      "balance_loss_mlp": 1.03169811,
      "epoch": 0.0585582877412373,
      "flos": 23150532510720.0,
      "grad_norm": 4.502778052358142,
      "language_loss": 0.88107145,
      "learning_rate": 3.9914852558592914e-06,
      "loss": 0.90536219,
      "num_input_tokens_seen": 10529485,
      "step": 487,
      "time_per_iteration": 2.782132863998413
    },
    {
      "auxiliary_loss_clip": 0.01336898,
      "auxiliary_loss_mlp": 0.01045366,
      "balance_loss_clip": 1.07810688,
      "balance_loss_mlp": 1.03389239,
      "epoch": 0.05867853063187639,
      "flos": 23506406507520.0,
      "grad_norm": 4.525075121576098,
      "language_loss": 0.8113057,
      "learning_rate": 3.991413301455413e-06,
      "loss": 0.83512831,
      "num_input_tokens_seen": 10545935,
      "step": 488,
      "time_per_iteration": 2.7843244075775146
    },
    {
      "auxiliary_loss_clip": 0.01372749,
      "auxiliary_loss_mlp": 0.01040144,
      "balance_loss_clip": 1.07131362,
      "balance_loss_mlp": 1.02951002,
      "epoch": 0.05879877352251548,
      "flos": 29495803818240.0,
      "grad_norm": 2.3229675753938794,
      "language_loss": 0.77898836,
      "learning_rate": 3.991341044955719e-06,
      "loss": 0.80311728,
      "num_input_tokens_seen": 10565690,
      "step": 489,
      "time_per_iteration": 4.61077356338501
    },
    {
      "auxiliary_loss_clip": 0.01333344,
      "auxiliary_loss_mlp": 0.02594116,
      "balance_loss_clip": 1.07560611,
      "balance_loss_mlp": 1.00033402,
      "epoch": 0.05891901641315457,
      "flos": 20157485880960.0,
      "grad_norm": 1.9999151645334488,
      "language_loss": 0.81527877,
      "learning_rate": 3.991268486371172e-06,
      "loss": 0.85455334,
      "num_input_tokens_seen": 10584245,
      "step": 490,
      "time_per_iteration": 3.6616294384002686
    },
    {
      "auxiliary_loss_clip": 0.01377256,
      "auxiliary_loss_mlp": 0.01043719,
      "balance_loss_clip": 1.07121789,
      "balance_loss_mlp": 1.03199482,
      "epoch": 0.05903925930379366,
      "flos": 24644200694400.0,
      "grad_norm": 3.334826920784108,
      "language_loss": 0.87661719,
      "learning_rate": 3.991195625712779e-06,
      "loss": 0.90082693,
      "num_input_tokens_seen": 10601210,
      "step": 491,
      "time_per_iteration": 2.7935218811035156
    },
    {
      "auxiliary_loss_clip": 0.0128739,
      "auxiliary_loss_mlp": 0.01053026,
      "balance_loss_clip": 1.08054471,
      "balance_loss_mlp": 1.04252362,
      "epoch": 0.05915950219443276,
      "flos": 21250391045760.0,
      "grad_norm": 2.461401774917469,
      "language_loss": 0.8168065,
      "learning_rate": 3.991122462991592e-06,
      "loss": 0.84021062,
      "num_input_tokens_seen": 10620730,
      "step": 492,
      "time_per_iteration": 2.7021284103393555
    },
    {
      "auxiliary_loss_clip": 0.0128876,
      "auxiliary_loss_mlp": 0.01055389,
      "balance_loss_clip": 1.08047962,
      "balance_loss_mlp": 1.04312801,
      "epoch": 0.05927974508507185,
      "flos": 9902727319680.0,
      "grad_norm": 3.19515233785663,
      "language_loss": 0.81496066,
      "learning_rate": 3.991048998218712e-06,
      "loss": 0.83840215,
      "num_input_tokens_seen": 10634035,
      "step": 493,
      "time_per_iteration": 2.613429307937622
    },
    {
      "auxiliary_loss_clip": 0.01337661,
      "auxiliary_loss_mlp": 0.01051205,
      "balance_loss_clip": 1.07701743,
      "balance_loss_mlp": 1.03971326,
      "epoch": 0.05939998797571094,
      "flos": 18259499232000.0,
      "grad_norm": 2.1911182378063323,
      "language_loss": 0.7665711,
      "learning_rate": 3.990975231405281e-06,
      "loss": 0.79045975,
      "num_input_tokens_seen": 10652485,
      "step": 494,
      "time_per_iteration": 2.7370147705078125
    },
    {
      "auxiliary_loss_clip": 0.01332136,
      "auxiliary_loss_mlp": 0.01051308,
      "balance_loss_clip": 1.07468891,
      "balance_loss_mlp": 1.03963721,
      "epoch": 0.05952023086635003,
      "flos": 28256598558720.0,
      "grad_norm": 2.135056577901825,
      "language_loss": 0.78709203,
      "learning_rate": 3.990901162562491e-06,
      "loss": 0.8109265,
      "num_input_tokens_seen": 10673175,
      "step": 495,
      "time_per_iteration": 2.7803773880004883
    },
    {
      "auxiliary_loss_clip": 0.01430342,
      "auxiliary_loss_mlp": 0.02592104,
      "balance_loss_clip": 1.06937909,
      "balance_loss_mlp": 1.00029397,
      "epoch": 0.05964047375698912,
      "flos": 14902498045440.0,
      "grad_norm": 2.4378212281429685,
      "language_loss": 0.90592283,
      "learning_rate": 3.9908267917015765e-06,
      "loss": 0.94614732,
      "num_input_tokens_seen": 10691235,
      "step": 496,
      "time_per_iteration": 2.7880215644836426
    },
    {
      "auxiliary_loss_clip": 0.01323754,
      "auxiliary_loss_mlp": 0.01049831,
      "balance_loss_clip": 1.07179236,
      "balance_loss_mlp": 1.037642,
      "epoch": 0.059760716647628206,
      "flos": 23185581206400.0,
      "grad_norm": 2.4235067996303847,
      "language_loss": 0.9322477,
      "learning_rate": 3.990752118833821e-06,
      "loss": 0.95598358,
      "num_input_tokens_seen": 10708675,
      "step": 497,
      "time_per_iteration": 2.7675631046295166
    },
    {
      "auxiliary_loss_clip": 0.01285998,
      "auxiliary_loss_mlp": 0.01052016,
      "balance_loss_clip": 1.0789845,
      "balance_loss_mlp": 1.04069102,
      "epoch": 0.0598809595382673,
      "flos": 22746968231040.0,
      "grad_norm": 2.2308288588271448,
      "language_loss": 0.77986097,
      "learning_rate": 3.990677143970553e-06,
      "loss": 0.80324113,
      "num_input_tokens_seen": 10729485,
      "step": 498,
      "time_per_iteration": 2.667393684387207
    },
    {
      "auxiliary_loss_clip": 0.01430072,
      "auxiliary_loss_mlp": 0.01043177,
      "balance_loss_clip": 1.07685113,
      "balance_loss_mlp": 1.03130984,
      "epoch": 0.06000120242890639,
      "flos": 22127221946880.0,
      "grad_norm": 2.4246089819123253,
      "language_loss": 0.81347179,
      "learning_rate": 3.990601867123144e-06,
      "loss": 0.83820426,
      "num_input_tokens_seen": 10749210,
      "step": 499,
      "time_per_iteration": 2.8088667392730713
    },
    {
      "auxiliary_loss_clip": 0.01485477,
      "auxiliary_loss_mlp": 0.0104259,
      "balance_loss_clip": 1.07154405,
      "balance_loss_mlp": 1.03090715,
      "epoch": 0.06012144531954548,
      "flos": 19171773878400.0,
      "grad_norm": 2.7322123837250634,
      "language_loss": 0.85156941,
      "learning_rate": 3.990526288303014e-06,
      "loss": 0.87685007,
      "num_input_tokens_seen": 10768000,
      "step": 500,
      "time_per_iteration": 2.7381744384765625
    },
    {
      "auxiliary_loss_clip": 0.01384167,
      "auxiliary_loss_mlp": 0.02587236,
      "balance_loss_clip": 1.07754529,
      "balance_loss_mlp": 1.00030696,
      "epoch": 0.06024168821018457,
      "flos": 22783345729920.0,
      "grad_norm": 1.7152141506577199,
      "language_loss": 0.90749979,
      "learning_rate": 3.9904504075216295e-06,
      "loss": 0.94721377,
      "num_input_tokens_seen": 10788760,
      "step": 501,
      "time_per_iteration": 2.785727024078369
    },
    {
      "auxiliary_loss_clip": 0.01442569,
      "auxiliary_loss_mlp": 0.01046587,
      "balance_loss_clip": 1.07199991,
      "balance_loss_mlp": 1.03532732,
      "epoch": 0.06036193110082366,
      "flos": 18770687637120.0,
      "grad_norm": 2.479196328665803,
      "language_loss": 0.93807387,
      "learning_rate": 3.990374224790501e-06,
      "loss": 0.96296537,
      "num_input_tokens_seen": 10806965,
      "step": 502,
      "time_per_iteration": 2.7510111331939697
    },
    {
      "auxiliary_loss_clip": 0.01388776,
      "auxiliary_loss_mlp": 0.01049471,
      "balance_loss_clip": 1.07992327,
      "balance_loss_mlp": 1.03713894,
      "epoch": 0.06048217399146275,
      "flos": 17201570935680.0,
      "grad_norm": 2.1543029842506236,
      "language_loss": 0.7078101,
      "learning_rate": 3.990297740121185e-06,
      "loss": 0.73219258,
      "num_input_tokens_seen": 10824900,
      "step": 503,
      "time_per_iteration": 2.8235971927642822
    },
    {
      "auxiliary_loss_clip": 0.01336011,
      "auxiliary_loss_mlp": 0.02588922,
      "balance_loss_clip": 1.0780803,
      "balance_loss_mlp": 1.00034034,
      "epoch": 0.06060241688210185,
      "flos": 24024131187840.0,
      "grad_norm": 1.8713905346154884,
      "language_loss": 0.78052092,
      "learning_rate": 3.990220953525284e-06,
      "loss": 0.81977022,
      "num_input_tokens_seen": 10842010,
      "step": 504,
      "time_per_iteration": 2.714449405670166
    },
    {
      "auxiliary_loss_clip": 0.01371065,
      "auxiliary_loss_mlp": 0.01044412,
      "balance_loss_clip": 1.07048845,
      "balance_loss_mlp": 1.03305721,
      "epoch": 0.06072265977274094,
      "flos": 14611190745600.0,
      "grad_norm": 2.436693492620822,
      "language_loss": 0.74442101,
      "learning_rate": 3.9901438650144465e-06,
      "loss": 0.76857579,
      "num_input_tokens_seen": 10858260,
      "step": 505,
      "time_per_iteration": 2.8620800971984863
    },
    {
      "auxiliary_loss_clip": 0.01322448,
      "auxiliary_loss_mlp": 0.01045818,
      "balance_loss_clip": 1.07269657,
      "balance_loss_mlp": 1.03424287,
      "epoch": 0.06084290266338003,
      "flos": 20558284813440.0,
      "grad_norm": 3.1939837615412157,
      "language_loss": 0.91763836,
      "learning_rate": 3.990066474600367e-06,
      "loss": 0.94132102,
      "num_input_tokens_seen": 10876230,
      "step": 506,
      "time_per_iteration": 2.7474286556243896
    },
    {
      "auxiliary_loss_clip": 0.01320437,
      "auxiliary_loss_mlp": 0.0103795,
      "balance_loss_clip": 1.06985033,
      "balance_loss_mlp": 1.02667284,
      "epoch": 0.06096314555401912,
      "flos": 22309217182080.0,
      "grad_norm": 1.9259156873359928,
      "language_loss": 0.67850578,
      "learning_rate": 3.989988782294786e-06,
      "loss": 0.70208967,
      "num_input_tokens_seen": 10896320,
      "step": 507,
      "time_per_iteration": 2.7614176273345947
    },
    {
      "auxiliary_loss_clip": 0.01418214,
      "auxiliary_loss_mlp": 0.01042148,
      "balance_loss_clip": 1.06969333,
      "balance_loss_mlp": 1.03081703,
      "epoch": 0.06108338844465821,
      "flos": 19131374056320.0,
      "grad_norm": 2.0349497899977242,
      "language_loss": 0.94901383,
      "learning_rate": 3.989910788109489e-06,
      "loss": 0.97361743,
      "num_input_tokens_seen": 10912970,
      "step": 508,
      "time_per_iteration": 2.794086456298828
    },
    {
      "auxiliary_loss_clip": 0.01428153,
      "auxiliary_loss_mlp": 0.01048314,
      "balance_loss_clip": 1.07287478,
      "balance_loss_mlp": 1.03617287,
      "epoch": 0.0612036313352973,
      "flos": 33584018169600.0,
      "grad_norm": 2.1477411826776485,
      "language_loss": 0.74773109,
      "learning_rate": 3.989832492056307e-06,
      "loss": 0.77249575,
      "num_input_tokens_seen": 10933995,
      "step": 509,
      "time_per_iteration": 2.873129367828369
    },
    {
      "auxiliary_loss_clip": 0.0133014,
      "auxiliary_loss_mlp": 0.01051488,
      "balance_loss_clip": 1.07802272,
      "balance_loss_mlp": 1.03936481,
      "epoch": 0.06132387422593639,
      "flos": 27490552179840.0,
      "grad_norm": 2.3494585717691403,
      "language_loss": 0.80999517,
      "learning_rate": 3.989753894147119e-06,
      "loss": 0.83381152,
      "num_input_tokens_seen": 10954120,
      "step": 510,
      "time_per_iteration": 2.802497386932373
    },
    {
      "auxiliary_loss_clip": 0.01331325,
      "auxiliary_loss_mlp": 0.01043877,
      "balance_loss_clip": 1.08285379,
      "balance_loss_mlp": 1.03346395,
      "epoch": 0.061444117116575485,
      "flos": 25885057979520.0,
      "grad_norm": 1.8059239012009642,
      "language_loss": 0.80135679,
      "learning_rate": 3.989674994393846e-06,
      "loss": 0.82510877,
      "num_input_tokens_seen": 10973595,
      "step": 511,
      "time_per_iteration": 3.693615436553955
    },
    {
      "auxiliary_loss_clip": 0.01329435,
      "auxiliary_loss_mlp": 0.01049822,
      "balance_loss_clip": 1.07935524,
      "balance_loss_mlp": 1.03767443,
      "epoch": 0.061564360007214575,
      "flos": 28512031150080.0,
      "grad_norm": 2.4528722970029935,
      "language_loss": 0.94220328,
      "learning_rate": 3.98959579280846e-06,
      "loss": 0.96599591,
      "num_input_tokens_seen": 10991995,
      "step": 512,
      "time_per_iteration": 2.7505152225494385
    },
    {
      "auxiliary_loss_clip": 0.01466094,
      "auxiliary_loss_mlp": 0.01040726,
      "balance_loss_clip": 1.07189322,
      "balance_loss_mlp": 1.02960968,
      "epoch": 0.061684602897853665,
      "flos": 12094355652480.0,
      "grad_norm": 2.001891067819318,
      "language_loss": 0.82848465,
      "learning_rate": 3.989516289402973e-06,
      "loss": 0.85355282,
      "num_input_tokens_seen": 11007625,
      "step": 513,
      "time_per_iteration": 2.8366262912750244
    },
    {
      "auxiliary_loss_clip": 0.01504798,
      "auxiliary_loss_mlp": 0.01042325,
      "balance_loss_clip": 1.05627012,
      "balance_loss_mlp": 1.03072596,
      "epoch": 0.061804845788492754,
      "flos": 19532639865600.0,
      "grad_norm": 4.434682594424167,
      "language_loss": 0.80288541,
      "learning_rate": 3.989436484189447e-06,
      "loss": 0.82835662,
      "num_input_tokens_seen": 11025570,
      "step": 514,
      "time_per_iteration": 4.645785570144653
    },
    {
      "auxiliary_loss_clip": 0.01332659,
      "auxiliary_loss_mlp": 0.01043157,
      "balance_loss_clip": 1.07402205,
      "balance_loss_mlp": 1.03117657,
      "epoch": 0.061925088679131844,
      "flos": 15341111020800.0,
      "grad_norm": 4.065112550920332,
      "language_loss": 0.80581224,
      "learning_rate": 3.9893563771799885e-06,
      "loss": 0.82957041,
      "num_input_tokens_seen": 11042045,
      "step": 515,
      "time_per_iteration": 2.6870930194854736
    },
    {
      "auxiliary_loss_clip": 0.01285072,
      "auxiliary_loss_mlp": 0.01060984,
      "balance_loss_clip": 1.07961071,
      "balance_loss_mlp": 1.04918861,
      "epoch": 0.062045331569770934,
      "flos": 25919927107200.0,
      "grad_norm": 2.146904990666255,
      "language_loss": 0.86439621,
      "learning_rate": 3.989275968386749e-06,
      "loss": 0.88785678,
      "num_input_tokens_seen": 11059955,
      "step": 516,
      "time_per_iteration": 2.8156635761260986
    },
    {
      "auxiliary_loss_clip": 0.01375638,
      "auxiliary_loss_mlp": 0.01037372,
      "balance_loss_clip": 1.07051945,
      "balance_loss_mlp": 1.02710235,
      "epoch": 0.06216557446041003,
      "flos": 28110621686400.0,
      "grad_norm": 2.1978842227881494,
      "language_loss": 0.76982653,
      "learning_rate": 3.989195257821926e-06,
      "loss": 0.79395664,
      "num_input_tokens_seen": 11078440,
      "step": 517,
      "time_per_iteration": 3.6766786575317383
    },
    {
      "auxiliary_loss_clip": 0.01375151,
      "auxiliary_loss_mlp": 0.01054189,
      "balance_loss_clip": 1.07570076,
      "balance_loss_mlp": 1.04158807,
      "epoch": 0.06228581735104912,
      "flos": 23478181395840.0,
      "grad_norm": 2.1642201517837263,
      "language_loss": 0.84413886,
      "learning_rate": 3.989114245497765e-06,
      "loss": 0.86843228,
      "num_input_tokens_seen": 11098240,
      "step": 518,
      "time_per_iteration": 2.7432920932769775
    },
    {
      "auxiliary_loss_clip": 0.01330062,
      "auxiliary_loss_mlp": 0.01052381,
      "balance_loss_clip": 1.0717392,
      "balance_loss_mlp": 1.04171157,
      "epoch": 0.06240606024168821,
      "flos": 15195205975680.0,
      "grad_norm": 2.8716325533816605,
      "language_loss": 0.94697893,
      "learning_rate": 3.989032931426554e-06,
      "loss": 0.97080332,
      "num_input_tokens_seen": 11115395,
      "step": 519,
      "time_per_iteration": 2.6574881076812744
    },
    {
      "auxiliary_loss_clip": 0.01378085,
      "auxiliary_loss_mlp": 0.01042819,
      "balance_loss_clip": 1.07664013,
      "balance_loss_mlp": 1.03232884,
      "epoch": 0.06252630313232731,
      "flos": 20631829910400.0,
      "grad_norm": 1.935553595238754,
      "language_loss": 0.86917228,
      "learning_rate": 3.9889513156206295e-06,
      "loss": 0.8933813,
      "num_input_tokens_seen": 11134835,
      "step": 520,
      "time_per_iteration": 2.7724766731262207
    },
    {
      "auxiliary_loss_clip": 0.01430354,
      "auxiliary_loss_mlp": 0.01042185,
      "balance_loss_clip": 1.0720861,
      "balance_loss_mlp": 1.03035378,
      "epoch": 0.06264654602296639,
      "flos": 20778058177920.0,
      "grad_norm": 3.7777106651528074,
      "language_loss": 0.7383883,
      "learning_rate": 3.988869398092371e-06,
      "loss": 0.76311362,
      "num_input_tokens_seen": 11154745,
      "step": 521,
      "time_per_iteration": 2.799110174179077
    },
    {
      "auxiliary_loss_clip": 0.01377958,
      "auxiliary_loss_mlp": 0.01051011,
      "balance_loss_clip": 1.07433844,
      "balance_loss_mlp": 1.03897083,
      "epoch": 0.06276678891360549,
      "flos": 29605798241280.0,
      "grad_norm": 2.194106022234836,
      "language_loss": 0.78713441,
      "learning_rate": 3.988787178854206e-06,
      "loss": 0.81142402,
      "num_input_tokens_seen": 11174280,
      "step": 522,
      "time_per_iteration": 2.744763135910034
    },
    {
      "auxiliary_loss_clip": 0.01284873,
      "auxiliary_loss_mlp": 0.01039092,
      "balance_loss_clip": 1.0806365,
      "balance_loss_mlp": 1.02803564,
      "epoch": 0.06288703180424457,
      "flos": 22126288193280.0,
      "grad_norm": 2.1323494005652774,
      "language_loss": 0.87759608,
      "learning_rate": 3.988704657918608e-06,
      "loss": 0.90083575,
      "num_input_tokens_seen": 11193340,
      "step": 523,
      "time_per_iteration": 2.698155403137207
    },
    {
      "auxiliary_loss_clip": 0.01332438,
      "auxiliary_loss_mlp": 0.01042295,
      "balance_loss_clip": 1.07992041,
      "balance_loss_mlp": 1.03207326,
      "epoch": 0.06300727469488367,
      "flos": 14976689587200.0,
      "grad_norm": 2.7193814959760765,
      "language_loss": 0.79990411,
      "learning_rate": 3.988621835298094e-06,
      "loss": 0.82365149,
      "num_input_tokens_seen": 11210555,
      "step": 524,
      "time_per_iteration": 2.6493992805480957
    },
    {
      "auxiliary_loss_clip": 0.01278493,
      "auxiliary_loss_mlp": 0.01044815,
      "balance_loss_clip": 1.0786624,
      "balance_loss_mlp": 1.03411043,
      "epoch": 0.06312751758552275,
      "flos": 24535391420160.0,
      "grad_norm": 1.9434929368163127,
      "language_loss": 0.9168281,
      "learning_rate": 3.988538711005229e-06,
      "loss": 0.94006121,
      "num_input_tokens_seen": 11230010,
      "step": 525,
      "time_per_iteration": 2.767162322998047
    },
    {
      "auxiliary_loss_clip": 0.01326543,
      "auxiliary_loss_mlp": 0.01042389,
      "balance_loss_clip": 1.07694626,
      "balance_loss_mlp": 1.03083193,
      "epoch": 0.06324776047616185,
      "flos": 21507008785920.0,
      "grad_norm": 2.4724727137926794,
      "language_loss": 0.88501745,
      "learning_rate": 3.988455285052622e-06,
      "loss": 0.90870678,
      "num_input_tokens_seen": 11246190,
      "step": 526,
      "time_per_iteration": 2.7261240482330322
    },
    {
      "auxiliary_loss_clip": 0.01327678,
      "auxiliary_loss_mlp": 0.0104665,
      "balance_loss_clip": 1.07648516,
      "balance_loss_mlp": 1.03585017,
      "epoch": 0.06336800336680094,
      "flos": 21688034353920.0,
      "grad_norm": 2.1771662745987452,
      "language_loss": 0.83844984,
      "learning_rate": 3.98837155745293e-06,
      "loss": 0.86219311,
      "num_input_tokens_seen": 11264230,
      "step": 527,
      "time_per_iteration": 2.7480103969573975
    },
    {
      "auxiliary_loss_clip": 0.01329406,
      "auxiliary_loss_mlp": 0.01042957,
      "balance_loss_clip": 1.07641196,
      "balance_loss_mlp": 1.03230548,
      "epoch": 0.06348824625744003,
      "flos": 19500895221120.0,
      "grad_norm": 2.441567102108424,
      "language_loss": 0.76065409,
      "learning_rate": 3.988287528218854e-06,
      "loss": 0.78437769,
      "num_input_tokens_seen": 11283015,
      "step": 528,
      "time_per_iteration": 2.6685051918029785
    },
    {
      "auxiliary_loss_clip": 0.01327573,
      "auxiliary_loss_mlp": 0.01034965,
      "balance_loss_clip": 1.0787344,
      "balance_loss_mlp": 1.0246594,
      "epoch": 0.06360848914807912,
      "flos": 15481233976320.0,
      "grad_norm": 2.057891123789396,
      "language_loss": 0.90468502,
      "learning_rate": 3.98820319736314e-06,
      "loss": 0.92831039,
      "num_input_tokens_seen": 11299630,
      "step": 529,
      "time_per_iteration": 2.7294294834136963
    },
    {
      "auxiliary_loss_clip": 0.0143093,
      "auxiliary_loss_mlp": 0.010437,
      "balance_loss_clip": 1.07174361,
      "balance_loss_mlp": 1.03217864,
      "epoch": 0.0637287320387182,
      "flos": 20593369422720.0,
      "grad_norm": 1.86635856665021,
      "language_loss": 0.85371739,
      "learning_rate": 3.988118564898582e-06,
      "loss": 0.87846375,
      "num_input_tokens_seen": 11319170,
      "step": 530,
      "time_per_iteration": 2.738189220428467
    },
    {
      "auxiliary_loss_clip": 0.0142237,
      "auxiliary_loss_mlp": 0.02593444,
      "balance_loss_clip": 1.07156849,
      "balance_loss_mlp": 1.00047576,
      "epoch": 0.0638489749293573,
      "flos": 17412222245760.0,
      "grad_norm": 2.3759302907147792,
      "language_loss": 0.89304161,
      "learning_rate": 3.988033630838019e-06,
      "loss": 0.9331997,
      "num_input_tokens_seen": 11333210,
      "step": 531,
      "time_per_iteration": 2.7597556114196777
    },
    {
      "auxiliary_loss_clip": 0.01339918,
      "auxiliary_loss_mlp": 0.01051675,
      "balance_loss_clip": 1.07938445,
      "balance_loss_mlp": 1.0406003,
      "epoch": 0.0639692178199964,
      "flos": 23807661874560.0,
      "grad_norm": 1.7461418145768164,
      "language_loss": 0.88191944,
      "learning_rate": 3.987948395194334e-06,
      "loss": 0.90583539,
      "num_input_tokens_seen": 11355590,
      "step": 532,
      "time_per_iteration": 2.7791481018066406
    },
    {
      "auxiliary_loss_clip": 0.01325163,
      "auxiliary_loss_mlp": 0.01035621,
      "balance_loss_clip": 1.07355201,
      "balance_loss_mlp": 1.02403438,
      "epoch": 0.06408946071063548,
      "flos": 18477225521280.0,
      "grad_norm": 2.273154722351999,
      "language_loss": 0.76750576,
      "learning_rate": 3.987862857980458e-06,
      "loss": 0.79111362,
      "num_input_tokens_seen": 11371535,
      "step": 533,
      "time_per_iteration": 2.6336472034454346
    },
    {
      "auxiliary_loss_clip": 0.01430264,
      "auxiliary_loss_mlp": 0.01045312,
      "balance_loss_clip": 1.07289875,
      "balance_loss_mlp": 1.03416002,
      "epoch": 0.06420970360127458,
      "flos": 27162220936320.0,
      "grad_norm": 2.4488289136278563,
      "language_loss": 0.76991928,
      "learning_rate": 3.987777019209368e-06,
      "loss": 0.79467499,
      "num_input_tokens_seen": 11392050,
      "step": 534,
      "time_per_iteration": 2.832770586013794
    },
    {
      "auxiliary_loss_clip": 0.01280086,
      "auxiliary_loss_mlp": 0.01045136,
      "balance_loss_clip": 1.07737648,
      "balance_loss_mlp": 1.03350079,
      "epoch": 0.06432994649191366,
      "flos": 23659673840640.0,
      "grad_norm": 2.1449124327835545,
      "language_loss": 0.81161028,
      "learning_rate": 3.987690878894084e-06,
      "loss": 0.83486247,
      "num_input_tokens_seen": 11411765,
      "step": 535,
      "time_per_iteration": 2.656219244003296
    },
    {
      "auxiliary_loss_clip": 0.01387869,
      "auxiliary_loss_mlp": 0.01047259,
      "balance_loss_clip": 1.07502842,
      "balance_loss_mlp": 1.03486753,
      "epoch": 0.06445018938255276,
      "flos": 23403953940480.0,
      "grad_norm": 3.0061928758697025,
      "language_loss": 0.85024822,
      "learning_rate": 3.987604437047673e-06,
      "loss": 0.87459946,
      "num_input_tokens_seen": 11431565,
      "step": 536,
      "time_per_iteration": 2.7694480419158936
    },
    {
      "auxiliary_loss_clip": 0.01330637,
      "auxiliary_loss_mlp": 0.01054386,
      "balance_loss_clip": 1.07715964,
      "balance_loss_mlp": 1.04384208,
      "epoch": 0.06457043227319184,
      "flos": 19646692525440.0,
      "grad_norm": 2.2007826894850866,
      "language_loss": 0.77364302,
      "learning_rate": 3.987517693683251e-06,
      "loss": 0.79749328,
      "num_input_tokens_seen": 11450140,
      "step": 537,
      "time_per_iteration": 3.570066452026367
    },
    {
      "auxiliary_loss_clip": 0.01373933,
      "auxiliary_loss_mlp": 0.01042407,
      "balance_loss_clip": 1.07787561,
      "balance_loss_mlp": 1.03118348,
      "epoch": 0.06469067516383094,
      "flos": 16978744915200.0,
      "grad_norm": 2.791769252946017,
      "language_loss": 0.96036196,
      "learning_rate": 3.9874306488139745e-06,
      "loss": 0.98452526,
      "num_input_tokens_seen": 11465400,
      "step": 538,
      "time_per_iteration": 2.701536178588867
    },
    {
      "auxiliary_loss_clip": 0.01423505,
      "auxiliary_loss_mlp": 0.01045905,
      "balance_loss_clip": 1.07269454,
      "balance_loss_mlp": 1.03521752,
      "epoch": 0.06481091805447003,
      "flos": 23296401642240.0,
      "grad_norm": 2.050594570437705,
      "language_loss": 0.88076347,
      "learning_rate": 3.987343302453049e-06,
      "loss": 0.90545756,
      "num_input_tokens_seen": 11486675,
      "step": 539,
      "time_per_iteration": 2.840214729309082
    },
    {
      "auxiliary_loss_clip": 0.01382282,
      "auxiliary_loss_mlp": 0.01047498,
      "balance_loss_clip": 1.07679033,
      "balance_loss_mlp": 1.03601861,
      "epoch": 0.06493116094510912,
      "flos": 29172356824320.0,
      "grad_norm": 1.6440472646862636,
      "language_loss": 0.82530093,
      "learning_rate": 3.987255654613724e-06,
      "loss": 0.84959877,
      "num_input_tokens_seen": 11510440,
      "step": 540,
      "time_per_iteration": 4.506173133850098
    },
    {
      "auxiliary_loss_clip": 0.01422225,
      "auxiliary_loss_mlp": 0.01043608,
      "balance_loss_clip": 1.06882322,
      "balance_loss_mlp": 1.03243184,
      "epoch": 0.06505140383574821,
      "flos": 19865065259520.0,
      "grad_norm": 2.710527587529789,
      "language_loss": 0.70056903,
      "learning_rate": 3.987167705309296e-06,
      "loss": 0.72522736,
      "num_input_tokens_seen": 11529715,
      "step": 541,
      "time_per_iteration": 2.756824016571045
    },
    {
      "auxiliary_loss_clip": 0.0133355,
      "auxiliary_loss_mlp": 0.02586979,
      "balance_loss_clip": 1.07855594,
      "balance_loss_mlp": 1.00040293,
      "epoch": 0.0651716467263873,
      "flos": 17924703540480.0,
      "grad_norm": 1.9473578397455387,
      "language_loss": 0.95363802,
      "learning_rate": 3.987079454553108e-06,
      "loss": 0.99284327,
      "num_input_tokens_seen": 11547665,
      "step": 542,
      "time_per_iteration": 2.724221706390381
    },
    {
      "auxiliary_loss_clip": 0.01426009,
      "auxiliary_loss_mlp": 0.01044853,
      "balance_loss_clip": 1.07419801,
      "balance_loss_mlp": 1.0328846,
      "epoch": 0.0652918896170264,
      "flos": 20842840356480.0,
      "grad_norm": 2.6866994754450335,
      "language_loss": 0.91175014,
      "learning_rate": 3.986990902358546e-06,
      "loss": 0.93645871,
      "num_input_tokens_seen": 11564605,
      "step": 543,
      "time_per_iteration": 3.602112293243408
    },
    {
      "auxiliary_loss_clip": 0.0133351,
      "auxiliary_loss_mlp": 0.01045581,
      "balance_loss_clip": 1.07617486,
      "balance_loss_mlp": 1.03428602,
      "epoch": 0.06541213250766549,
      "flos": 21872507627520.0,
      "grad_norm": 2.088561049797057,
      "language_loss": 0.93313849,
      "learning_rate": 3.986902048739045e-06,
      "loss": 0.95692945,
      "num_input_tokens_seen": 11584550,
      "step": 544,
      "time_per_iteration": 2.8568389415740967
    },
    {
      "auxiliary_loss_clip": 0.01385083,
      "auxiliary_loss_mlp": 0.01053188,
      "balance_loss_clip": 1.07593298,
      "balance_loss_mlp": 1.04098141,
      "epoch": 0.06553237539830457,
      "flos": 23110743219840.0,
      "grad_norm": 2.588043030370696,
      "language_loss": 0.80191618,
      "learning_rate": 3.986812893708082e-06,
      "loss": 0.82629895,
      "num_input_tokens_seen": 11600740,
      "step": 545,
      "time_per_iteration": 2.861245632171631
    },
    {
      "auxiliary_loss_clip": 0.01383875,
      "auxiliary_loss_mlp": 0.0103367,
      "balance_loss_clip": 1.0738008,
      "balance_loss_mlp": 1.02233934,
      "epoch": 0.06565261828894367,
      "flos": 17923769786880.0,
      "grad_norm": 2.3582632391546245,
      "language_loss": 0.81240129,
      "learning_rate": 3.9867234372791826e-06,
      "loss": 0.8365767,
      "num_input_tokens_seen": 11618695,
      "step": 546,
      "time_per_iteration": 2.842209577560425
    },
    {
      "auxiliary_loss_clip": 0.01334307,
      "auxiliary_loss_mlp": 0.01041158,
      "balance_loss_clip": 1.07940674,
      "balance_loss_mlp": 1.03022027,
      "epoch": 0.06577286117958275,
      "flos": 22783058421120.0,
      "grad_norm": 3.6556345993853805,
      "language_loss": 0.87577903,
      "learning_rate": 3.986633679465918e-06,
      "loss": 0.89953369,
      "num_input_tokens_seen": 11638850,
      "step": 547,
      "time_per_iteration": 2.8336970806121826
    },
    {
      "auxiliary_loss_clip": 0.0147805,
      "auxiliary_loss_mlp": 0.01051888,
      "balance_loss_clip": 1.07172573,
      "balance_loss_mlp": 1.04076576,
      "epoch": 0.06589310407022185,
      "flos": 23696194993920.0,
      "grad_norm": 2.8779770303274246,
      "language_loss": 0.81105512,
      "learning_rate": 3.986543620281904e-06,
      "loss": 0.83635449,
      "num_input_tokens_seen": 11658500,
      "step": 548,
      "time_per_iteration": 2.8675637245178223
    },
    {
      "auxiliary_loss_clip": 0.01363289,
      "auxiliary_loss_mlp": 0.01041466,
      "balance_loss_clip": 1.07140517,
      "balance_loss_mlp": 1.03048134,
      "epoch": 0.06601334696086093,
      "flos": 26864772410880.0,
      "grad_norm": 1.7735646885983758,
      "language_loss": 0.91261196,
      "learning_rate": 3.986453259740802e-06,
      "loss": 0.93665946,
      "num_input_tokens_seen": 11676670,
      "step": 549,
      "time_per_iteration": 2.9319019317626953
    },
    {
      "auxiliary_loss_clip": 0.01381551,
      "auxiliary_loss_mlp": 0.01055833,
      "balance_loss_clip": 1.07925749,
      "balance_loss_mlp": 1.04370975,
      "epoch": 0.06613358985150003,
      "flos": 12567694101120.0,
      "grad_norm": 2.6453955041953425,
      "language_loss": 0.79335439,
      "learning_rate": 3.986362597856319e-06,
      "loss": 0.81772828,
      "num_input_tokens_seen": 11693170,
      "step": 550,
      "time_per_iteration": 2.8234612941741943
    },
    {
      "auxiliary_loss_clip": 0.01378367,
      "auxiliary_loss_mlp": 0.0259381,
      "balance_loss_clip": 1.07358885,
      "balance_loss_mlp": 1.00035644,
      "epoch": 0.06625383274213913,
      "flos": 18332505624960.0,
      "grad_norm": 2.88873710215487,
      "language_loss": 0.81916302,
      "learning_rate": 3.986271634642211e-06,
      "loss": 0.85888481,
      "num_input_tokens_seen": 11710150,
      "step": 551,
      "time_per_iteration": 2.8693578243255615
    },
    {
      "auxiliary_loss_clip": 0.01284504,
      "auxiliary_loss_mlp": 0.01057406,
      "balance_loss_clip": 1.0833869,
      "balance_loss_mlp": 1.04538369,
      "epoch": 0.06637407563277821,
      "flos": 15375585098880.0,
      "grad_norm": 2.749595750168124,
      "language_loss": 0.81660116,
      "learning_rate": 3.986180370112274e-06,
      "loss": 0.8400203,
      "num_input_tokens_seen": 11726670,
      "step": 552,
      "time_per_iteration": 2.7583374977111816
    },
    {
      "auxiliary_loss_clip": 0.01337732,
      "auxiliary_loss_mlp": 0.02593384,
      "balance_loss_clip": 1.08093166,
      "balance_loss_mlp": 1.0003171,
      "epoch": 0.0664943185234173,
      "flos": 24025244509440.0,
      "grad_norm": 1.8043063801694297,
      "language_loss": 0.74654502,
      "learning_rate": 3.986088804280354e-06,
      "loss": 0.78585619,
      "num_input_tokens_seen": 11746400,
      "step": 553,
      "time_per_iteration": 2.880911111831665
    },
    {
      "auxiliary_loss_clip": 0.01386154,
      "auxiliary_loss_mlp": 0.01047414,
      "balance_loss_clip": 1.0786674,
      "balance_loss_mlp": 1.03630972,
      "epoch": 0.06661456141405639,
      "flos": 20957503547520.0,
      "grad_norm": 2.4995736339455368,
      "language_loss": 0.94176149,
      "learning_rate": 3.985996937160342e-06,
      "loss": 0.96609712,
      "num_input_tokens_seen": 11765590,
      "step": 554,
      "time_per_iteration": 2.8973259925842285
    },
    {
      "auxiliary_loss_clip": 0.01330298,
      "auxiliary_loss_mlp": 0.01056266,
      "balance_loss_clip": 1.08067977,
      "balance_loss_mlp": 1.04418373,
      "epoch": 0.06673480430469549,
      "flos": 52223953322880.0,
      "grad_norm": 2.166176094058847,
      "language_loss": 0.689996,
      "learning_rate": 3.985904768766173e-06,
      "loss": 0.71386158,
      "num_input_tokens_seen": 11788365,
      "step": 555,
      "time_per_iteration": 3.039015054702759
    },
    {
      "auxiliary_loss_clip": 0.0142996,
      "auxiliary_loss_mlp": 0.01046628,
      "balance_loss_clip": 1.07243133,
      "balance_loss_mlp": 1.03402138,
      "epoch": 0.06685504719533458,
      "flos": 16217079995520.0,
      "grad_norm": 6.81220451517536,
      "language_loss": 0.75713813,
      "learning_rate": 3.98581229911183e-06,
      "loss": 0.78190404,
      "num_input_tokens_seen": 11807285,
      "step": 556,
      "time_per_iteration": 2.874995708465576
    },
    {
      "auxiliary_loss_clip": 0.01328016,
      "auxiliary_loss_mlp": 0.01044436,
      "balance_loss_clip": 1.07273531,
      "balance_loss_mlp": 1.03306913,
      "epoch": 0.06697529008597367,
      "flos": 22491535639680.0,
      "grad_norm": 3.113459100639243,
      "language_loss": 0.92233849,
      "learning_rate": 3.985719528211341e-06,
      "loss": 0.94606304,
      "num_input_tokens_seen": 11826655,
      "step": 557,
      "time_per_iteration": 2.6852171421051025
    },
    {
      "auxiliary_loss_clip": 0.01312094,
      "auxiliary_loss_mlp": 0.01015488,
      "balance_loss_clip": 1.06258845,
      "balance_loss_mlp": 1.00971842,
      "epoch": 0.06709553297661276,
      "flos": 62688216936960.0,
      "grad_norm": 0.8502188889705806,
      "language_loss": 0.63034678,
      "learning_rate": 3.985626456078777e-06,
      "loss": 0.65362263,
      "num_input_tokens_seen": 11891310,
      "step": 558,
      "time_per_iteration": 3.4033663272857666
    },
    {
      "auxiliary_loss_clip": 0.01429979,
      "auxiliary_loss_mlp": 0.01048635,
      "balance_loss_clip": 1.07456017,
      "balance_loss_mlp": 1.03753054,
      "epoch": 0.06721577586725185,
      "flos": 11216590997760.0,
      "grad_norm": 2.3250178521845726,
      "language_loss": 0.86175084,
      "learning_rate": 3.985533082728259e-06,
      "loss": 0.88653702,
      "num_input_tokens_seen": 11906965,
      "step": 559,
      "time_per_iteration": 2.73799204826355
    },
    {
      "auxiliary_loss_clip": 0.01282404,
      "auxiliary_loss_mlp": 0.01043173,
      "balance_loss_clip": 1.07829249,
      "balance_loss_mlp": 1.03116882,
      "epoch": 0.06733601875789094,
      "flos": 25922189664000.0,
      "grad_norm": 1.932863032138568,
      "language_loss": 0.75136709,
      "learning_rate": 3.985439408173951e-06,
      "loss": 0.77462286,
      "num_input_tokens_seen": 11927190,
      "step": 560,
      "time_per_iteration": 2.718461036682129
    },
    {
      "auxiliary_loss_clip": 0.01283836,
      "auxiliary_loss_mlp": 0.01053204,
      "balance_loss_clip": 1.07893491,
      "balance_loss_mlp": 1.04202199,
      "epoch": 0.06745626164853002,
      "flos": 20813645577600.0,
      "grad_norm": 1.9784239340395544,
      "language_loss": 0.71076858,
      "learning_rate": 3.9853454324300634e-06,
      "loss": 0.73413897,
      "num_input_tokens_seen": 11946400,
      "step": 561,
      "time_per_iteration": 2.691452741622925
    },
    {
      "auxiliary_loss_clip": 0.01531704,
      "auxiliary_loss_mlp": 0.01059851,
      "balance_loss_clip": 1.06669199,
      "balance_loss_mlp": 1.04826987,
      "epoch": 0.06757650453916912,
      "flos": 19829262378240.0,
      "grad_norm": 2.7025763554139486,
      "language_loss": 0.77929604,
      "learning_rate": 3.985251155510852e-06,
      "loss": 0.80521154,
      "num_input_tokens_seen": 11965430,
      "step": 562,
      "time_per_iteration": 2.7816076278686523
    },
    {
      "auxiliary_loss_clip": 0.01475176,
      "auxiliary_loss_mlp": 0.01050699,
      "balance_loss_clip": 1.07255602,
      "balance_loss_mlp": 1.03933799,
      "epoch": 0.06769674742980822,
      "flos": 25739224761600.0,
      "grad_norm": 1.7970612625131432,
      "language_loss": 0.80439258,
      "learning_rate": 3.98515657743062e-06,
      "loss": 0.82965136,
      "num_input_tokens_seen": 11984895,
      "step": 563,
      "time_per_iteration": 3.810678720474243
    },
    {
      "auxiliary_loss_clip": 0.01375468,
      "auxiliary_loss_mlp": 0.01046475,
      "balance_loss_clip": 1.07121444,
      "balance_loss_mlp": 1.03489947,
      "epoch": 0.0678169903204473,
      "flos": 13074788355840.0,
      "grad_norm": 2.175017138614066,
      "language_loss": 0.77940625,
      "learning_rate": 3.985061698203711e-06,
      "loss": 0.8036257,
      "num_input_tokens_seen": 12002010,
      "step": 564,
      "time_per_iteration": 2.651338815689087
    },
    {
      "auxiliary_loss_clip": 0.01198954,
      "auxiliary_loss_mlp": 0.01005594,
      "balance_loss_clip": 1.06411529,
      "balance_loss_mlp": 0.9999668,
      "epoch": 0.0679372332110864,
      "flos": 70865830788480.0,
      "grad_norm": 0.8831298396705786,
      "language_loss": 0.63821709,
      "learning_rate": 3.984966517844523e-06,
      "loss": 0.66026258,
      "num_input_tokens_seen": 12057255,
      "step": 565,
      "time_per_iteration": 3.186741590499878
    },
    {
      "auxiliary_loss_clip": 0.01282993,
      "auxiliary_loss_mlp": 0.01043832,
      "balance_loss_clip": 1.07873702,
      "balance_loss_mlp": 1.03239346,
      "epoch": 0.06805747610172548,
      "flos": 28256418990720.0,
      "grad_norm": 2.811485283995285,
      "language_loss": 0.808442,
      "learning_rate": 3.984871036367492e-06,
      "loss": 0.83171022,
      "num_input_tokens_seen": 12077280,
      "step": 566,
      "time_per_iteration": 3.574991226196289
    },
    {
      "auxiliary_loss_clip": 0.01331603,
      "auxiliary_loss_mlp": 0.02592403,
      "balance_loss_clip": 1.07974792,
      "balance_loss_mlp": 1.00019908,
      "epoch": 0.06817771899236458,
      "flos": 20120533764480.0,
      "grad_norm": 2.132544219830055,
      "language_loss": 0.83489972,
      "learning_rate": 3.984775253787102e-06,
      "loss": 0.87413979,
      "num_input_tokens_seen": 12095570,
      "step": 567,
      "time_per_iteration": 3.6243817806243896
    },
    {
      "auxiliary_loss_clip": 0.01334866,
      "auxiliary_loss_mlp": 0.01048819,
      "balance_loss_clip": 1.07502401,
      "balance_loss_mlp": 1.03769064,
      "epoch": 0.06829796188300366,
      "flos": 17930629284480.0,
      "grad_norm": 3.624395823004769,
      "language_loss": 0.88061869,
      "learning_rate": 3.984679170117885e-06,
      "loss": 0.90445548,
      "num_input_tokens_seen": 12111775,
      "step": 568,
      "time_per_iteration": 2.646669387817383
    },
    {
      "auxiliary_loss_clip": 0.01323984,
      "auxiliary_loss_mlp": 0.010413,
      "balance_loss_clip": 1.07420087,
      "balance_loss_mlp": 1.02984393,
      "epoch": 0.06841820477364276,
      "flos": 14501627285760.0,
      "grad_norm": 4.905166664749562,
      "language_loss": 0.78321946,
      "learning_rate": 3.984582785374415e-06,
      "loss": 0.80687225,
      "num_input_tokens_seen": 12129215,
      "step": 569,
      "time_per_iteration": 3.559953212738037
    },
    {
      "auxiliary_loss_clip": 0.01379948,
      "auxiliary_loss_mlp": 0.02591523,
      "balance_loss_clip": 1.07702088,
      "balance_loss_mlp": 1.00020242,
      "epoch": 0.06853844766428185,
      "flos": 21938474954880.0,
      "grad_norm": 2.2353455054588016,
      "language_loss": 0.80898416,
      "learning_rate": 3.9844860995713155e-06,
      "loss": 0.84869879,
      "num_input_tokens_seen": 12148755,
      "step": 570,
      "time_per_iteration": 2.75514554977417
    },
    {
      "auxiliary_loss_clip": 0.01332288,
      "auxiliary_loss_mlp": 0.01051538,
      "balance_loss_clip": 1.081002,
      "balance_loss_mlp": 1.03925359,
      "epoch": 0.06865869055492094,
      "flos": 16800628348800.0,
      "grad_norm": 2.380604135870617,
      "language_loss": 0.82608557,
      "learning_rate": 3.9843891127232524e-06,
      "loss": 0.84992379,
      "num_input_tokens_seen": 12166290,
      "step": 571,
      "time_per_iteration": 2.7105190753936768
    },
    {
      "auxiliary_loss_clip": 0.01469889,
      "auxiliary_loss_mlp": 0.01044517,
      "balance_loss_clip": 1.06804264,
      "balance_loss_mlp": 1.03313804,
      "epoch": 0.06877893344556003,
      "flos": 19937281553280.0,
      "grad_norm": 2.323454812733296,
      "language_loss": 0.67126977,
      "learning_rate": 3.984291824844938e-06,
      "loss": 0.69641387,
      "num_input_tokens_seen": 12181385,
      "step": 572,
      "time_per_iteration": 2.7614810466766357
    },
    {
      "auxiliary_loss_clip": 0.01278101,
      "auxiliary_loss_mlp": 0.01046328,
      "balance_loss_clip": 1.07646096,
      "balance_loss_mlp": 1.03456211,
      "epoch": 0.06889917633619912,
      "flos": 23039388852480.0,
      "grad_norm": 2.3684043832581123,
      "language_loss": 0.85084629,
      "learning_rate": 3.984194235951132e-06,
      "loss": 0.87409055,
      "num_input_tokens_seen": 12197530,
      "step": 573,
      "time_per_iteration": 2.6283726692199707
    },
    {
      "auxiliary_loss_clip": 0.01284537,
      "auxiliary_loss_mlp": 0.01044973,
      "balance_loss_clip": 1.08174014,
      "balance_loss_mlp": 1.03303409,
      "epoch": 0.06901941922683821,
      "flos": 20960556203520.0,
      "grad_norm": 2.5980287288834596,
      "language_loss": 0.84594715,
      "learning_rate": 3.9840963460566375e-06,
      "loss": 0.86924231,
      "num_input_tokens_seen": 12216310,
      "step": 574,
      "time_per_iteration": 2.619202136993408
    },
    {
      "auxiliary_loss_clip": 0.01515479,
      "auxiliary_loss_mlp": 0.0103723,
      "balance_loss_clip": 1.0642643,
      "balance_loss_mlp": 1.02582741,
      "epoch": 0.06913966211747731,
      "flos": 24821850384000.0,
      "grad_norm": 1.6440755303063885,
      "language_loss": 0.89472568,
      "learning_rate": 3.983998155176305e-06,
      "loss": 0.92025274,
      "num_input_tokens_seen": 12236670,
      "step": 575,
      "time_per_iteration": 2.8686869144439697
    },
    {
      "auxiliary_loss_clip": 0.01193103,
      "auxiliary_loss_mlp": 0.01011794,
      "balance_loss_clip": 1.06166148,
      "balance_loss_mlp": 1.00614381,
      "epoch": 0.06925990500811639,
      "flos": 58367446957440.0,
      "grad_norm": 0.8174076643515461,
      "language_loss": 0.57029331,
      "learning_rate": 3.9838996633250305e-06,
      "loss": 0.59234226,
      "num_input_tokens_seen": 12297185,
      "step": 576,
      "time_per_iteration": 3.1449880599975586
    },
    {
      "auxiliary_loss_clip": 0.01327795,
      "auxiliary_loss_mlp": 0.01045623,
      "balance_loss_clip": 1.07266545,
      "balance_loss_mlp": 1.0342505,
      "epoch": 0.06938014789875549,
      "flos": 12749940731520.0,
      "grad_norm": 2.0650688660850354,
      "language_loss": 0.88425511,
      "learning_rate": 3.983800870517753e-06,
      "loss": 0.90798926,
      "num_input_tokens_seen": 12313975,
      "step": 577,
      "time_per_iteration": 2.640011787414551
    },
    {
      "auxiliary_loss_clip": 0.01326504,
      "auxiliary_loss_mlp": 0.01049885,
      "balance_loss_clip": 1.08058357,
      "balance_loss_mlp": 1.03818464,
      "epoch": 0.06950039078939457,
      "flos": 22820226019200.0,
      "grad_norm": 2.995720277815689,
      "language_loss": 0.77964306,
      "learning_rate": 3.983701776769463e-06,
      "loss": 0.80340689,
      "num_input_tokens_seen": 12331385,
      "step": 578,
      "time_per_iteration": 2.6770009994506836
    },
    {
      "auxiliary_loss_clip": 0.0131564,
      "auxiliary_loss_mlp": 0.0105205,
      "balance_loss_clip": 1.07288527,
      "balance_loss_mlp": 1.04086816,
      "epoch": 0.06962063368003367,
      "flos": 21941348042880.0,
      "grad_norm": 2.091972156753445,
      "language_loss": 0.85534918,
      "learning_rate": 3.9836023820951885e-06,
      "loss": 0.87902611,
      "num_input_tokens_seen": 12350600,
      "step": 579,
      "time_per_iteration": 2.661500930786133
    },
    {
      "auxiliary_loss_clip": 0.01414079,
      "auxiliary_loss_mlp": 0.01046674,
      "balance_loss_clip": 1.06534457,
      "balance_loss_mlp": 1.03568256,
      "epoch": 0.06974087657067275,
      "flos": 20706021452160.0,
      "grad_norm": 1.9213090339996477,
      "language_loss": 0.68743938,
      "learning_rate": 3.983502686510011e-06,
      "loss": 0.71204686,
      "num_input_tokens_seen": 12371430,
      "step": 580,
      "time_per_iteration": 2.7717554569244385
    },
    {
      "auxiliary_loss_clip": 0.01327992,
      "auxiliary_loss_mlp": 0.02590111,
      "balance_loss_clip": 1.07304001,
      "balance_loss_mlp": 1.00012863,
      "epoch": 0.06986111946131185,
      "flos": 22638230784000.0,
      "grad_norm": 1.8138744500343353,
      "language_loss": 0.73308027,
      "learning_rate": 3.9834026900290525e-06,
      "loss": 0.77226132,
      "num_input_tokens_seen": 12390825,
      "step": 581,
      "time_per_iteration": 2.641983985900879
    },
    {
      "auxiliary_loss_clip": 0.01279935,
      "auxiliary_loss_mlp": 0.01036456,
      "balance_loss_clip": 1.07861042,
      "balance_loss_mlp": 1.02573895,
      "epoch": 0.06998136235195095,
      "flos": 26943453152640.0,
      "grad_norm": 3.0064046867762197,
      "language_loss": 1.00068271,
      "learning_rate": 3.983302392667482e-06,
      "loss": 1.02384663,
      "num_input_tokens_seen": 12411670,
      "step": 582,
      "time_per_iteration": 2.7552082538604736
    },
    {
      "auxiliary_loss_clip": 0.01325849,
      "auxiliary_loss_mlp": 0.010386,
      "balance_loss_clip": 1.07711685,
      "balance_loss_mlp": 1.02761459,
      "epoch": 0.07010160524259003,
      "flos": 22492505306880.0,
      "grad_norm": 1.7923193074041923,
      "language_loss": 0.9359417,
      "learning_rate": 3.983201794440517e-06,
      "loss": 0.9595862,
      "num_input_tokens_seen": 12431245,
      "step": 583,
      "time_per_iteration": 2.679828643798828
    },
    {
      "auxiliary_loss_clip": 0.0136172,
      "auxiliary_loss_mlp": 0.01044291,
      "balance_loss_clip": 1.06959522,
      "balance_loss_mlp": 1.03293014,
      "epoch": 0.07022184813322913,
      "flos": 18332541538560.0,
      "grad_norm": 1.9162713801064692,
      "language_loss": 0.67672759,
      "learning_rate": 3.9831008953634165e-06,
      "loss": 0.70078766,
      "num_input_tokens_seen": 12450535,
      "step": 584,
      "time_per_iteration": 2.740778684616089
    },
    {
      "auxiliary_loss_clip": 0.01458923,
      "auxiliary_loss_mlp": 0.01049088,
      "balance_loss_clip": 1.06546712,
      "balance_loss_mlp": 1.03757215,
      "epoch": 0.07034209102386821,
      "flos": 24675550289280.0,
      "grad_norm": 1.9446417163439316,
      "language_loss": 0.81267118,
      "learning_rate": 3.9829996954514864e-06,
      "loss": 0.83775127,
      "num_input_tokens_seen": 12469675,
      "step": 585,
      "time_per_iteration": 2.8337106704711914
    },
    {
      "auxiliary_loss_clip": 0.01320857,
      "auxiliary_loss_mlp": 0.01040231,
      "balance_loss_clip": 1.07396078,
      "balance_loss_mlp": 1.02917981,
      "epoch": 0.0704623339145073,
      "flos": 25995878415360.0,
      "grad_norm": 1.9407157925798153,
      "language_loss": 0.84394038,
      "learning_rate": 3.982898194720079e-06,
      "loss": 0.86755127,
      "num_input_tokens_seen": 12490405,
      "step": 586,
      "time_per_iteration": 2.7459158897399902
    },
    {
      "auxiliary_loss_clip": 0.01372263,
      "auxiliary_loss_mlp": 0.02589897,
      "balance_loss_clip": 1.07277942,
      "balance_loss_mlp": 1.00015092,
      "epoch": 0.0705825768051464,
      "flos": 25338318088320.0,
      "grad_norm": 2.4170370707422655,
      "language_loss": 0.82439572,
      "learning_rate": 3.982796393184592e-06,
      "loss": 0.86401737,
      "num_input_tokens_seen": 12509485,
      "step": 587,
      "time_per_iteration": 2.792184352874756
    },
    {
      "auxiliary_loss_clip": 0.01246312,
      "auxiliary_loss_mlp": 0.01004903,
      "balance_loss_clip": 1.06016707,
      "balance_loss_mlp": 0.99956208,
      "epoch": 0.07070281969578548,
      "flos": 66047552507520.0,
      "grad_norm": 0.7929743237528213,
      "language_loss": 0.62719899,
      "learning_rate": 3.98269429086047e-06,
      "loss": 0.64971119,
      "num_input_tokens_seen": 12567325,
      "step": 588,
      "time_per_iteration": 4.102558851242065
    },
    {
      "auxiliary_loss_clip": 0.01364728,
      "auxiliary_loss_mlp": 0.01041187,
      "balance_loss_clip": 1.07058382,
      "balance_loss_mlp": 1.0285871,
      "epoch": 0.07082306258642458,
      "flos": 23653568528640.0,
      "grad_norm": 3.6028106634356956,
      "language_loss": 0.86754799,
      "learning_rate": 3.982591887763199e-06,
      "loss": 0.89160717,
      "num_input_tokens_seen": 12584785,
      "step": 589,
      "time_per_iteration": 2.740933895111084
    },
    {
      "auxiliary_loss_clip": 0.01401204,
      "auxiliary_loss_mlp": 0.01035615,
      "balance_loss_clip": 1.05884552,
      "balance_loss_mlp": 1.02393281,
      "epoch": 0.07094330547706366,
      "flos": 13880049408000.0,
      "grad_norm": 2.358007728259653,
      "language_loss": 0.82031995,
      "learning_rate": 3.982489183908316e-06,
      "loss": 0.84468818,
      "num_input_tokens_seen": 12601205,
      "step": 590,
      "time_per_iteration": 2.7460474967956543
    },
    {
      "auxiliary_loss_clip": 0.01500306,
      "auxiliary_loss_mlp": 0.010464,
      "balance_loss_clip": 1.05585265,
      "balance_loss_mlp": 1.03578436,
      "epoch": 0.07106354836770276,
      "flos": 24645098534400.0,
      "grad_norm": 2.046104845492566,
      "language_loss": 0.84522831,
      "learning_rate": 3.982386179311399e-06,
      "loss": 0.87069535,
      "num_input_tokens_seen": 12621725,
      "step": 591,
      "time_per_iteration": 2.794407367706299
    },
    {
      "auxiliary_loss_clip": 0.01323779,
      "auxiliary_loss_mlp": 0.01039341,
      "balance_loss_clip": 1.07319307,
      "balance_loss_mlp": 1.02775931,
      "epoch": 0.07118379125834184,
      "flos": 16217223649920.0,
      "grad_norm": 2.404099222864785,
      "language_loss": 0.87385124,
      "learning_rate": 3.982282873988075e-06,
      "loss": 0.89748245,
      "num_input_tokens_seen": 12639600,
      "step": 592,
      "time_per_iteration": 3.615161180496216
    },
    {
      "auxiliary_loss_clip": 0.01369181,
      "auxiliary_loss_mlp": 0.01044211,
      "balance_loss_clip": 1.07268095,
      "balance_loss_mlp": 1.03310037,
      "epoch": 0.07130403414898094,
      "flos": 19719986227200.0,
      "grad_norm": 1.6258481454373053,
      "language_loss": 0.86850631,
      "learning_rate": 3.982179267954016e-06,
      "loss": 0.89264023,
      "num_input_tokens_seen": 12660030,
      "step": 593,
      "time_per_iteration": 3.680293560028076
    },
    {
      "auxiliary_loss_clip": 0.01275702,
      "auxiliary_loss_mlp": 0.01040114,
      "balance_loss_clip": 1.07606971,
      "balance_loss_mlp": 1.02884829,
      "epoch": 0.07142427703962004,
      "flos": 21871933009920.0,
      "grad_norm": 2.6214719704868488,
      "language_loss": 0.96085095,
      "learning_rate": 3.982075361224937e-06,
      "loss": 0.98400915,
      "num_input_tokens_seen": 12678395,
      "step": 594,
      "time_per_iteration": 3.594738245010376
    },
    {
      "auxiliary_loss_clip": 0.01318117,
      "auxiliary_loss_mlp": 0.02588341,
      "balance_loss_clip": 1.07338858,
      "balance_loss_mlp": 1.0000689,
      "epoch": 0.07154451993025912,
      "flos": 18296595002880.0,
      "grad_norm": 2.705273344435775,
      "language_loss": 0.88262516,
      "learning_rate": 3.981971153816602e-06,
      "loss": 0.92168975,
      "num_input_tokens_seen": 12696000,
      "step": 595,
      "time_per_iteration": 2.696108818054199
    },
    {
      "auxiliary_loss_clip": 0.01271215,
      "auxiliary_loss_mlp": 0.01042222,
      "balance_loss_clip": 1.07816267,
      "balance_loss_mlp": 1.03194594,
      "epoch": 0.07166476282089822,
      "flos": 22160690444160.0,
      "grad_norm": 1.5766541195687978,
      "language_loss": 0.96275836,
      "learning_rate": 3.981866645744819e-06,
      "loss": 0.98589271,
      "num_input_tokens_seen": 12716715,
      "step": 596,
      "time_per_iteration": 2.6414456367492676
    },
    {
      "auxiliary_loss_clip": 0.01275667,
      "auxiliary_loss_mlp": 0.02588094,
      "balance_loss_clip": 1.07619989,
      "balance_loss_mlp": 1.00006914,
      "epoch": 0.0717850057115373,
      "flos": 14136343925760.0,
      "grad_norm": 2.634035392804029,
      "language_loss": 0.81332111,
      "learning_rate": 3.9817618370254416e-06,
      "loss": 0.85195875,
      "num_input_tokens_seen": 12733370,
      "step": 597,
      "time_per_iteration": 2.63093638420105
    },
    {
      "auxiliary_loss_clip": 0.01274481,
      "auxiliary_loss_mlp": 0.01041099,
      "balance_loss_clip": 1.07569158,
      "balance_loss_mlp": 1.03001833,
      "epoch": 0.0719052486021764,
      "flos": 30917794412160.0,
      "grad_norm": 2.034289880150531,
      "language_loss": 0.87367582,
      "learning_rate": 3.9816567276743684e-06,
      "loss": 0.89683163,
      "num_input_tokens_seen": 12753235,
      "step": 598,
      "time_per_iteration": 2.705021619796753
    },
    {
      "auxiliary_loss_clip": 0.01361201,
      "auxiliary_loss_mlp": 0.01035671,
      "balance_loss_clip": 1.07030821,
      "balance_loss_mlp": 1.0254662,
      "epoch": 0.0720254914928155,
      "flos": 21287019939840.0,
      "grad_norm": 2.169846062661895,
      "language_loss": 0.77494448,
      "learning_rate": 3.9815513177075466e-06,
      "loss": 0.79891324,
      "num_input_tokens_seen": 12772020,
      "step": 599,
      "time_per_iteration": 2.747922658920288
    },
    {
      "auxiliary_loss_clip": 0.01316175,
      "auxiliary_loss_mlp": 0.01040987,
      "balance_loss_clip": 1.07418871,
      "balance_loss_mlp": 1.0296979,
      "epoch": 0.07214573438345458,
      "flos": 27819170732160.0,
      "grad_norm": 1.8050628545500094,
      "language_loss": 0.70299006,
      "learning_rate": 3.9814456071409646e-06,
      "loss": 0.72656167,
      "num_input_tokens_seen": 12792555,
      "step": 600,
      "time_per_iteration": 2.7633063793182373
    },
    {
      "auxiliary_loss_clip": 0.01470169,
      "auxiliary_loss_mlp": 0.01046465,
      "balance_loss_clip": 1.06658196,
      "balance_loss_mlp": 1.03414512,
      "epoch": 0.07226597727409367,
      "flos": 25483576688640.0,
      "grad_norm": 2.752607975583128,
      "language_loss": 0.85236084,
      "learning_rate": 3.981339595990659e-06,
      "loss": 0.87752718,
      "num_input_tokens_seen": 12811085,
      "step": 601,
      "time_per_iteration": 2.8157124519348145
    },
    {
      "auxiliary_loss_clip": 0.01317142,
      "auxiliary_loss_mlp": 0.010377,
      "balance_loss_clip": 1.07230854,
      "balance_loss_mlp": 1.0262146,
      "epoch": 0.07238622016473276,
      "flos": 23513840622720.0,
      "grad_norm": 2.271951294932348,
      "language_loss": 0.80880862,
      "learning_rate": 3.981233284272713e-06,
      "loss": 0.83235705,
      "num_input_tokens_seen": 12830830,
      "step": 602,
      "time_per_iteration": 2.720517158508301
    },
    {
      "auxiliary_loss_clip": 0.01418958,
      "auxiliary_loss_mlp": 0.01041325,
      "balance_loss_clip": 1.06824207,
      "balance_loss_mlp": 1.03125238,
      "epoch": 0.07250646305537185,
      "flos": 25453519983360.0,
      "grad_norm": 1.866897519682805,
      "language_loss": 0.90240413,
      "learning_rate": 3.981126672003253e-06,
      "loss": 0.92700696,
      "num_input_tokens_seen": 12853505,
      "step": 603,
      "time_per_iteration": 2.8238914012908936
    },
    {
      "auxiliary_loss_clip": 0.01374475,
      "auxiliary_loss_mlp": 0.0103508,
      "balance_loss_clip": 1.06761181,
      "balance_loss_mlp": 1.02534616,
      "epoch": 0.07262670594601094,
      "flos": 27155038216320.0,
      "grad_norm": 2.4516474696374697,
      "language_loss": 0.78286362,
      "learning_rate": 3.981019759198451e-06,
      "loss": 0.80695915,
      "num_input_tokens_seen": 12872455,
      "step": 604,
      "time_per_iteration": 2.780959367752075
    },
    {
      "auxiliary_loss_clip": 0.01370477,
      "auxiliary_loss_mlp": 0.01039585,
      "balance_loss_clip": 1.06986582,
      "balance_loss_mlp": 1.02950597,
      "epoch": 0.07274694883665003,
      "flos": 26651607148800.0,
      "grad_norm": 2.2060154818369635,
      "language_loss": 0.84251404,
      "learning_rate": 3.980912545874528e-06,
      "loss": 0.8666147,
      "num_input_tokens_seen": 12892620,
      "step": 605,
      "time_per_iteration": 2.716996431350708
    },
    {
      "auxiliary_loss_clip": 0.01317729,
      "auxiliary_loss_mlp": 0.02586734,
      "balance_loss_clip": 1.0708034,
      "balance_loss_mlp": 1.00000763,
      "epoch": 0.07286719172728913,
      "flos": 29862344154240.0,
      "grad_norm": 1.991931988063435,
      "language_loss": 0.8560164,
      "learning_rate": 3.980805032047746e-06,
      "loss": 0.89506108,
      "num_input_tokens_seen": 12914090,
      "step": 606,
      "time_per_iteration": 2.7763590812683105
    },
    {
      "auxiliary_loss_clip": 0.01370352,
      "auxiliary_loss_mlp": 0.0103848,
      "balance_loss_clip": 1.07419145,
      "balance_loss_mlp": 1.02638626,
      "epoch": 0.07298743461792821,
      "flos": 17382057799680.0,
      "grad_norm": 2.5649290994390754,
      "language_loss": 0.80951202,
      "learning_rate": 3.980697217734415e-06,
      "loss": 0.83360034,
      "num_input_tokens_seen": 12931830,
      "step": 607,
      "time_per_iteration": 2.6567769050598145
    },
    {
      "auxiliary_loss_clip": 0.01466115,
      "auxiliary_loss_mlp": 0.02583623,
      "balance_loss_clip": 1.06662631,
      "balance_loss_mlp": 1.0000236,
      "epoch": 0.07310767750856731,
      "flos": 19498201701120.0,
      "grad_norm": 2.1525396954012863,
      "language_loss": 0.91592002,
      "learning_rate": 3.980589102950891e-06,
      "loss": 0.95641744,
      "num_input_tokens_seen": 12949995,
      "step": 608,
      "time_per_iteration": 2.7612292766571045
    },
    {
      "auxiliary_loss_clip": 0.01372268,
      "auxiliary_loss_mlp": 0.01040613,
      "balance_loss_clip": 1.07645941,
      "balance_loss_mlp": 1.02950263,
      "epoch": 0.07322792039920639,
      "flos": 29168693637120.0,
      "grad_norm": 2.9314771281598087,
      "language_loss": 0.76049507,
      "learning_rate": 3.9804806877135755e-06,
      "loss": 0.78462386,
      "num_input_tokens_seen": 12968040,
      "step": 609,
      "time_per_iteration": 2.786144971847534
    },
    {
      "auxiliary_loss_clip": 0.01319401,
      "auxiliary_loss_mlp": 0.02588991,
      "balance_loss_clip": 1.06915426,
      "balance_loss_mlp": 1.00003171,
      "epoch": 0.07334816328984549,
      "flos": 23477822259840.0,
      "grad_norm": 2.589581944171058,
      "language_loss": 0.86401415,
      "learning_rate": 3.980371972038915e-06,
      "loss": 0.90309811,
      "num_input_tokens_seen": 12988530,
      "step": 610,
      "time_per_iteration": 2.6964573860168457
    },
    {
      "auxiliary_loss_clip": 0.01274618,
      "auxiliary_loss_mlp": 0.01049529,
      "balance_loss_clip": 1.0769341,
      "balance_loss_mlp": 1.03879404,
      "epoch": 0.07346840618048459,
      "flos": 22962467877120.0,
      "grad_norm": 2.1131833168189282,
      "language_loss": 0.84383118,
      "learning_rate": 3.980262955943399e-06,
      "loss": 0.86707258,
      "num_input_tokens_seen": 13008195,
      "step": 611,
      "time_per_iteration": 2.6877377033233643
    },
    {
      "auxiliary_loss_clip": 0.01366884,
      "auxiliary_loss_mlp": 0.01046273,
      "balance_loss_clip": 1.07567906,
      "balance_loss_mlp": 1.0353651,
      "epoch": 0.07358864907112367,
      "flos": 17673903803520.0,
      "grad_norm": 2.5141329825158545,
      "language_loss": 0.8668468,
      "learning_rate": 3.980153639443569e-06,
      "loss": 0.8909784,
      "num_input_tokens_seen": 13024180,
      "step": 612,
      "time_per_iteration": 2.6435089111328125
    },
    {
      "auxiliary_loss_clip": 0.0137721,
      "auxiliary_loss_mlp": 0.01045731,
      "balance_loss_clip": 1.0724529,
      "balance_loss_mlp": 1.03463817,
      "epoch": 0.07370889196176277,
      "flos": 24097029840000.0,
      "grad_norm": 2.713643393749694,
      "language_loss": 0.80228472,
      "learning_rate": 3.980044022556005e-06,
      "loss": 0.82651412,
      "num_input_tokens_seen": 13043865,
      "step": 613,
      "time_per_iteration": 2.7358334064483643
    },
    {
      "auxiliary_loss_clip": 0.01322716,
      "auxiliary_loss_mlp": 0.01045138,
      "balance_loss_clip": 1.0754385,
      "balance_loss_mlp": 1.0336287,
      "epoch": 0.07382913485240185,
      "flos": 25885919905920.0,
      "grad_norm": 2.167438549205323,
      "language_loss": 0.73055214,
      "learning_rate": 3.9799341052973375e-06,
      "loss": 0.75423074,
      "num_input_tokens_seen": 13063700,
      "step": 614,
      "time_per_iteration": 3.617845058441162
    },
    {
      "auxiliary_loss_clip": 0.01369878,
      "auxiliary_loss_mlp": 0.01043272,
      "balance_loss_clip": 1.07542133,
      "balance_loss_mlp": 1.03163719,
      "epoch": 0.07394937774304094,
      "flos": 16873850223360.0,
      "grad_norm": 2.4416784665132534,
      "language_loss": 0.75106311,
      "learning_rate": 3.979823887684241e-06,
      "loss": 0.77519464,
      "num_input_tokens_seen": 13082640,
      "step": 615,
      "time_per_iteration": 2.750213861465454
    },
    {
      "auxiliary_loss_clip": 0.01270006,
      "auxiliary_loss_mlp": 0.01037998,
      "balance_loss_clip": 1.07558298,
      "balance_loss_mlp": 1.02725697,
      "epoch": 0.07406962063368003,
      "flos": 20703471586560.0,
      "grad_norm": 2.418715251030413,
      "language_loss": 0.84618211,
      "learning_rate": 3.979713369733434e-06,
      "loss": 0.8692621,
      "num_input_tokens_seen": 13100505,
      "step": 616,
      "time_per_iteration": 2.6227598190307617
    },
    {
      "auxiliary_loss_clip": 0.01315707,
      "auxiliary_loss_mlp": 0.01054328,
      "balance_loss_clip": 1.07169461,
      "balance_loss_mlp": 1.04257393,
      "epoch": 0.07418986352431912,
      "flos": 21430985650560.0,
      "grad_norm": 1.9995778997369211,
      "language_loss": 0.85182232,
      "learning_rate": 3.979602551461683e-06,
      "loss": 0.87552267,
      "num_input_tokens_seen": 13121285,
      "step": 617,
      "time_per_iteration": 2.6946120262145996
    },
    {
      "auxiliary_loss_clip": 0.01365554,
      "auxiliary_loss_mlp": 0.01039776,
      "balance_loss_clip": 1.0724659,
      "balance_loss_mlp": 1.02854037,
      "epoch": 0.07431010641495822,
      "flos": 12021133777920.0,
      "grad_norm": 4.100800929031688,
      "language_loss": 0.9167217,
      "learning_rate": 3.979491432885799e-06,
      "loss": 0.94077504,
      "num_input_tokens_seen": 13137550,
      "step": 618,
      "time_per_iteration": 4.443578004837036
    },
    {
      "auxiliary_loss_clip": 0.01403637,
      "auxiliary_loss_mlp": 0.02582791,
      "balance_loss_clip": 1.06645751,
      "balance_loss_mlp": 0.99995482,
      "epoch": 0.0744303493055973,
      "flos": 20957575374720.0,
      "grad_norm": 1.9997598613473169,
      "language_loss": 0.83144248,
      "learning_rate": 3.97938001402264e-06,
      "loss": 0.87130678,
      "num_input_tokens_seen": 13156675,
      "step": 619,
      "time_per_iteration": 2.769505262374878
    },
    {
      "auxiliary_loss_clip": 0.01417089,
      "auxiliary_loss_mlp": 0.01049562,
      "balance_loss_clip": 1.07027745,
      "balance_loss_mlp": 1.03823113,
      "epoch": 0.0745505921962364,
      "flos": 16253134272000.0,
      "grad_norm": 11.70966078542185,
      "language_loss": 0.80008483,
      "learning_rate": 3.979268294889105e-06,
      "loss": 0.82475138,
      "num_input_tokens_seen": 13172225,
      "step": 620,
      "time_per_iteration": 2.726390838623047
    },
    {
      "auxiliary_loss_clip": 0.01271643,
      "auxiliary_loss_mlp": 0.01044451,
      "balance_loss_clip": 1.07534194,
      "balance_loss_mlp": 1.03347182,
      "epoch": 0.07467083508687548,
      "flos": 50944635550080.0,
      "grad_norm": 2.715976302951772,
      "language_loss": 0.73882186,
      "learning_rate": 3.979156275502143e-06,
      "loss": 0.7619828,
      "num_input_tokens_seen": 13195885,
      "step": 621,
      "time_per_iteration": 3.7880072593688965
    },
    {
      "auxiliary_loss_clip": 0.0141842,
      "auxiliary_loss_mlp": 0.01043934,
      "balance_loss_clip": 1.0701499,
      "balance_loss_mlp": 1.03241885,
      "epoch": 0.07479107797751458,
      "flos": 17529686697600.0,
      "grad_norm": 2.410617047985421,
      "language_loss": 0.91473055,
      "learning_rate": 3.979043955878749e-06,
      "loss": 0.93935406,
      "num_input_tokens_seen": 13213730,
      "step": 622,
      "time_per_iteration": 2.7290053367614746
    },
    {
      "auxiliary_loss_clip": 0.01368003,
      "auxiliary_loss_mlp": 0.01052275,
      "balance_loss_clip": 1.07316232,
      "balance_loss_mlp": 1.04105711,
      "epoch": 0.07491132086815366,
      "flos": 23473943591040.0,
      "grad_norm": 1.9931475024015706,
      "language_loss": 0.83341336,
      "learning_rate": 3.978931336035959e-06,
      "loss": 0.85761613,
      "num_input_tokens_seen": 13232540,
      "step": 623,
      "time_per_iteration": 2.816680431365967
    },
    {
      "auxiliary_loss_clip": 0.01320053,
      "auxiliary_loss_mlp": 0.01038005,
      "balance_loss_clip": 1.07497036,
      "balance_loss_mlp": 1.02707386,
      "epoch": 0.07503156375879276,
      "flos": 20157557708160.0,
      "grad_norm": 2.3023782102516406,
      "language_loss": 0.82541013,
      "learning_rate": 3.9788184159908595e-06,
      "loss": 0.84899068,
      "num_input_tokens_seen": 13249670,
      "step": 624,
      "time_per_iteration": 2.680757999420166
    },
    {
      "auxiliary_loss_clip": 0.01362111,
      "auxiliary_loss_mlp": 0.01041706,
      "balance_loss_clip": 1.0692277,
      "balance_loss_mlp": 1.03023851,
      "epoch": 0.07515180664943186,
      "flos": 15115519653120.0,
      "grad_norm": 3.796256522796148,
      "language_loss": 0.8212083,
      "learning_rate": 3.97870519576058e-06,
      "loss": 0.84524649,
      "num_input_tokens_seen": 13266095,
      "step": 625,
      "time_per_iteration": 2.6835975646972656
    },
    {
      "auxiliary_loss_clip": 0.01410353,
      "auxiliary_loss_mlp": 0.02584336,
      "balance_loss_clip": 1.0658592,
      "balance_loss_mlp": 0.99998689,
      "epoch": 0.07527204954007094,
      "flos": 21287702298240.0,
      "grad_norm": 2.421101239204768,
      "language_loss": 0.81034607,
      "learning_rate": 3.978591675362295e-06,
      "loss": 0.85029292,
      "num_input_tokens_seen": 13284810,
      "step": 626,
      "time_per_iteration": 2.8495218753814697
    },
    {
      "auxiliary_loss_clip": 0.01464973,
      "auxiliary_loss_mlp": 0.01037388,
      "balance_loss_clip": 1.07168055,
      "balance_loss_mlp": 1.02637863,
      "epoch": 0.07539229243071004,
      "flos": 21324187537920.0,
      "grad_norm": 2.104075856930006,
      "language_loss": 0.87813652,
      "learning_rate": 3.978477854813226e-06,
      "loss": 0.9031601,
      "num_input_tokens_seen": 13304150,
      "step": 627,
      "time_per_iteration": 2.7442119121551514
    },
    {
      "auxiliary_loss_clip": 0.01312933,
      "auxiliary_loss_mlp": 0.0104323,
      "balance_loss_clip": 1.06802881,
      "balance_loss_mlp": 1.03216171,
      "epoch": 0.07551253532134912,
      "flos": 13042540920960.0,
      "grad_norm": 1.921031606221562,
      "language_loss": 0.82181978,
      "learning_rate": 3.97836373413064e-06,
      "loss": 0.84538138,
      "num_input_tokens_seen": 13322205,
      "step": 628,
      "time_per_iteration": 2.661221742630005
    },
    {
      "auxiliary_loss_clip": 0.0127106,
      "auxiliary_loss_mlp": 0.01038532,
      "balance_loss_clip": 1.07547498,
      "balance_loss_mlp": 1.02770209,
      "epoch": 0.07563277821198822,
      "flos": 19208761908480.0,
      "grad_norm": 2.0445131805385466,
      "language_loss": 0.74909699,
      "learning_rate": 3.978249313331848e-06,
      "loss": 0.77219296,
      "num_input_tokens_seen": 13340435,
      "step": 629,
      "time_per_iteration": 2.633612632751465
    },
    {
      "auxiliary_loss_clip": 0.01325148,
      "auxiliary_loss_mlp": 0.02586405,
      "balance_loss_clip": 1.07289243,
      "balance_loss_mlp": 0.99997771,
      "epoch": 0.07575302110262731,
      "flos": 19537200892800.0,
      "grad_norm": 4.988731211072895,
      "language_loss": 0.62452149,
      "learning_rate": 3.978134592434208e-06,
      "loss": 0.66363704,
      "num_input_tokens_seen": 13358185,
      "step": 630,
      "time_per_iteration": 2.747529983520508
    },
    {
      "auxiliary_loss_clip": 0.01378924,
      "auxiliary_loss_mlp": 0.01006631,
      "balance_loss_clip": 1.04696894,
      "balance_loss_mlp": 1.00253069,
      "epoch": 0.0758732639932664,
      "flos": 67961808017280.0,
      "grad_norm": 1.0190153987490642,
      "language_loss": 0.59394765,
      "learning_rate": 3.978019571455123e-06,
      "loss": 0.61780316,
      "num_input_tokens_seen": 13410130,
      "step": 631,
      "time_per_iteration": 3.343780040740967
    },
    {
      "auxiliary_loss_clip": 0.01263092,
      "auxiliary_loss_mlp": 0.01042146,
      "balance_loss_clip": 1.072896,
      "balance_loss_mlp": 1.03248382,
      "epoch": 0.07599350688390549,
      "flos": 18989204025600.0,
      "grad_norm": 2.4984586642493722,
      "language_loss": 0.84336615,
      "learning_rate": 3.977904250412042e-06,
      "loss": 0.86641848,
      "num_input_tokens_seen": 13429085,
      "step": 632,
      "time_per_iteration": 2.6478588581085205
    },
    {
      "auxiliary_loss_clip": 0.01367995,
      "auxiliary_loss_mlp": 0.01047633,
      "balance_loss_clip": 1.06855333,
      "balance_loss_mlp": 1.03718472,
      "epoch": 0.07611374977454458,
      "flos": 21069006341760.0,
      "grad_norm": 2.2757272039393626,
      "language_loss": 0.85617548,
      "learning_rate": 3.97778862932246e-06,
      "loss": 0.88033175,
      "num_input_tokens_seen": 13446250,
      "step": 633,
      "time_per_iteration": 2.678265333175659
    },
    {
      "auxiliary_loss_clip": 0.01587019,
      "auxiliary_loss_mlp": 0.01037079,
      "balance_loss_clip": 1.0459404,
      "balance_loss_mlp": 1.02720833,
      "epoch": 0.07623399266518367,
      "flos": 18514536773760.0,
      "grad_norm": 2.2287751583473328,
      "language_loss": 0.94204599,
      "learning_rate": 3.9776727082039144e-06,
      "loss": 0.96828699,
      "num_input_tokens_seen": 13463220,
      "step": 634,
      "time_per_iteration": 3.0301592350006104
    },
    {
      "auxiliary_loss_clip": 0.01189881,
      "auxiliary_loss_mlp": 0.01008643,
      "balance_loss_clip": 1.06086266,
      "balance_loss_mlp": 1.00349355,
      "epoch": 0.07635423555582276,
      "flos": 44663036077440.0,
      "grad_norm": 0.8076855183480597,
      "language_loss": 0.55457997,
      "learning_rate": 3.977556487073991e-06,
      "loss": 0.57656527,
      "num_input_tokens_seen": 13517775,
      "step": 635,
      "time_per_iteration": 3.4673964977264404
    },
    {
      "auxiliary_loss_clip": 0.01354606,
      "auxiliary_loss_mlp": 0.0104114,
      "balance_loss_clip": 1.06308961,
      "balance_loss_mlp": 1.03010142,
      "epoch": 0.07647447844646185,
      "flos": 21761148487680.0,
      "grad_norm": 1.7477138630038351,
      "language_loss": 0.81525916,
      "learning_rate": 3.97743996595032e-06,
      "loss": 0.83921659,
      "num_input_tokens_seen": 13537815,
      "step": 636,
      "time_per_iteration": 2.745070695877075
    },
    {
      "auxiliary_loss_clip": 0.01266199,
      "auxiliary_loss_mlp": 0.01044778,
      "balance_loss_clip": 1.07383657,
      "balance_loss_mlp": 1.03403091,
      "epoch": 0.07659472133710095,
      "flos": 23806799948160.0,
      "grad_norm": 1.6920031253231589,
      "language_loss": 0.81616622,
      "learning_rate": 3.9773231448505804e-06,
      "loss": 0.83927596,
      "num_input_tokens_seen": 13559605,
      "step": 637,
      "time_per_iteration": 2.736001491546631
    },
    {
      "auxiliary_loss_clip": 0.01361731,
      "auxiliary_loss_mlp": 0.0258318,
      "balance_loss_clip": 1.07162213,
      "balance_loss_mlp": 0.99994522,
      "epoch": 0.07671496422774003,
      "flos": 21469984842240.0,
      "grad_norm": 2.1112790470177845,
      "language_loss": 0.78240907,
      "learning_rate": 3.977206023792491e-06,
      "loss": 0.82185823,
      "num_input_tokens_seen": 13579495,
      "step": 638,
      "time_per_iteration": 2.7096707820892334
    },
    {
      "auxiliary_loss_clip": 0.0131472,
      "auxiliary_loss_mlp": 0.01044972,
      "balance_loss_clip": 1.07485461,
      "balance_loss_mlp": 1.03427863,
      "epoch": 0.07683520711837913,
      "flos": 16980971558400.0,
      "grad_norm": 2.655831997104829,
      "language_loss": 0.81327236,
      "learning_rate": 3.97708860279382e-06,
      "loss": 0.83686924,
      "num_input_tokens_seen": 13597605,
      "step": 639,
      "time_per_iteration": 2.697643756866455
    },
    {
      "auxiliary_loss_clip": 0.01410196,
      "auxiliary_loss_mlp": 0.01042089,
      "balance_loss_clip": 1.06550765,
      "balance_loss_mlp": 1.03143156,
      "epoch": 0.07695545000901821,
      "flos": 23476744851840.0,
      "grad_norm": 1.87958228618036,
      "language_loss": 0.77932179,
      "learning_rate": 3.97697088187238e-06,
      "loss": 0.80384469,
      "num_input_tokens_seen": 13618120,
      "step": 640,
      "time_per_iteration": 3.794719696044922
    },
    {
      "auxiliary_loss_clip": 0.01362012,
      "auxiliary_loss_mlp": 0.01042051,
      "balance_loss_clip": 1.07248259,
      "balance_loss_mlp": 1.03158426,
      "epoch": 0.07707569289965731,
      "flos": 17634258167040.0,
      "grad_norm": 2.1444766363261785,
      "language_loss": 0.91988015,
      "learning_rate": 3.976852861046029e-06,
      "loss": 0.94392079,
      "num_input_tokens_seen": 13634735,
      "step": 641,
      "time_per_iteration": 2.6646389961242676
    },
    {
      "auxiliary_loss_clip": 0.01450535,
      "auxiliary_loss_mlp": 0.01040663,
      "balance_loss_clip": 1.06318104,
      "balance_loss_mlp": 1.03047633,
      "epoch": 0.0771959357902964,
      "flos": 25775674087680.0,
      "grad_norm": 2.225725985946451,
      "language_loss": 0.80374354,
      "learning_rate": 3.97673454033267e-06,
      "loss": 0.82865554,
      "num_input_tokens_seen": 13656835,
      "step": 642,
      "time_per_iteration": 2.8306381702423096
    },
    {
      "auxiliary_loss_clip": 0.0136648,
      "auxiliary_loss_mlp": 0.01037098,
      "balance_loss_clip": 1.06988263,
      "balance_loss_mlp": 1.02686429,
      "epoch": 0.07731617868093549,
      "flos": 19828651847040.0,
      "grad_norm": 2.1134771855324064,
      "language_loss": 0.82620847,
      "learning_rate": 3.976615919750254e-06,
      "loss": 0.85024428,
      "num_input_tokens_seen": 13674535,
      "step": 643,
      "time_per_iteration": 2.7072434425354004
    },
    {
      "auxiliary_loss_clip": 0.01311302,
      "auxiliary_loss_mlp": 0.01040538,
      "balance_loss_clip": 1.07239223,
      "balance_loss_mlp": 1.0305661,
      "epoch": 0.07743642157157458,
      "flos": 21324654414720.0,
      "grad_norm": 1.8677005912745428,
      "language_loss": 0.87147593,
      "learning_rate": 3.976496999316775e-06,
      "loss": 0.89499426,
      "num_input_tokens_seen": 13693290,
      "step": 644,
      "time_per_iteration": 4.388904094696045
    },
    {
      "auxiliary_loss_clip": 0.01361242,
      "auxiliary_loss_mlp": 0.01045683,
      "balance_loss_clip": 1.0734725,
      "balance_loss_mlp": 1.034621,
      "epoch": 0.07755666446221367,
      "flos": 19969133938560.0,
      "grad_norm": 2.5602259073398894,
      "language_loss": 0.84191418,
      "learning_rate": 3.976377779050271e-06,
      "loss": 0.86598343,
      "num_input_tokens_seen": 13711420,
      "step": 645,
      "time_per_iteration": 2.718125581741333
    },
    {
      "auxiliary_loss_clip": 0.01306414,
      "auxiliary_loss_mlp": 0.0103932,
      "balance_loss_clip": 1.06855416,
      "balance_loss_mlp": 1.028198,
      "epoch": 0.07767690735285276,
      "flos": 23623224514560.0,
      "grad_norm": 2.6345616907279337,
      "language_loss": 0.84358037,
      "learning_rate": 3.976258258968831e-06,
      "loss": 0.86703771,
      "num_input_tokens_seen": 13729965,
      "step": 646,
      "time_per_iteration": 3.6911776065826416
    },
    {
      "auxiliary_loss_clip": 0.01410764,
      "auxiliary_loss_mlp": 0.01045527,
      "balance_loss_clip": 1.07140398,
      "balance_loss_mlp": 1.03406525,
      "epoch": 0.07779715024349185,
      "flos": 22236246702720.0,
      "grad_norm": 2.396067901460766,
      "language_loss": 0.74805725,
      "learning_rate": 3.976138439090583e-06,
      "loss": 0.77262014,
      "num_input_tokens_seen": 13748045,
      "step": 647,
      "time_per_iteration": 2.830609083175659
    },
    {
      "auxiliary_loss_clip": 0.01410726,
      "auxiliary_loss_mlp": 0.01039081,
      "balance_loss_clip": 1.06949258,
      "balance_loss_mlp": 1.02756548,
      "epoch": 0.07791739313413094,
      "flos": 20955097336320.0,
      "grad_norm": 2.84078444117434,
      "language_loss": 0.8527171,
      "learning_rate": 3.976018319433706e-06,
      "loss": 0.87721515,
      "num_input_tokens_seen": 13765590,
      "step": 648,
      "time_per_iteration": 2.787923812866211
    },
    {
      "auxiliary_loss_clip": 0.01310687,
      "auxiliary_loss_mlp": 0.01039637,
      "balance_loss_clip": 1.07126689,
      "balance_loss_mlp": 1.02941442,
      "epoch": 0.07803763602477004,
      "flos": 19312327797120.0,
      "grad_norm": 2.2673977423636846,
      "language_loss": 0.91244781,
      "learning_rate": 3.9758979000164205e-06,
      "loss": 0.93595099,
      "num_input_tokens_seen": 13782410,
      "step": 649,
      "time_per_iteration": 2.6814589500427246
    },
    {
      "auxiliary_loss_clip": 0.01419086,
      "auxiliary_loss_mlp": 0.01030094,
      "balance_loss_clip": 1.06971252,
      "balance_loss_mlp": 1.01899576,
      "epoch": 0.07815787891540912,
      "flos": 22710806213760.0,
      "grad_norm": 1.8115009914860862,
      "language_loss": 0.71952432,
      "learning_rate": 3.975777180856995e-06,
      "loss": 0.74401617,
      "num_input_tokens_seen": 13801530,
      "step": 650,
      "time_per_iteration": 2.810716390609741
    },
    {
      "auxiliary_loss_clip": 0.01263775,
      "auxiliary_loss_mlp": 0.01042589,
      "balance_loss_clip": 1.07106519,
      "balance_loss_mlp": 1.03206921,
      "epoch": 0.07827812180604822,
      "flos": 22711129436160.0,
      "grad_norm": 4.330494463606602,
      "language_loss": 0.86133635,
      "learning_rate": 3.975656161973742e-06,
      "loss": 0.88440001,
      "num_input_tokens_seen": 13820615,
      "step": 651,
      "time_per_iteration": 2.651812791824341
    },
    {
      "auxiliary_loss_clip": 0.01265407,
      "auxiliary_loss_mlp": 0.01046799,
      "balance_loss_clip": 1.07265532,
      "balance_loss_mlp": 1.03549242,
      "epoch": 0.0783983646966873,
      "flos": 21725597001600.0,
      "grad_norm": 4.194050394785483,
      "language_loss": 0.88856524,
      "learning_rate": 3.9755348433850194e-06,
      "loss": 0.91168725,
      "num_input_tokens_seen": 13835955,
      "step": 652,
      "time_per_iteration": 2.7462494373321533
    },
    {
      "auxiliary_loss_clip": 0.01286369,
      "auxiliary_loss_mlp": 0.01004568,
      "balance_loss_clip": 1.05582106,
      "balance_loss_mlp": 1.00058615,
      "epoch": 0.0785186075873264,
      "flos": 60640877537280.0,
      "grad_norm": 0.9666744035646373,
      "language_loss": 0.63580668,
      "learning_rate": 3.975413225109232e-06,
      "loss": 0.65871596,
      "num_input_tokens_seen": 13896505,
      "step": 653,
      "time_per_iteration": 3.2896504402160645
    },
    {
      "auxiliary_loss_clip": 0.01313808,
      "auxiliary_loss_mlp": 0.01036479,
      "balance_loss_clip": 1.07289994,
      "balance_loss_mlp": 1.02604818,
      "epoch": 0.0786388504779655,
      "flos": 23877902920320.0,
      "grad_norm": 4.452199040366271,
      "language_loss": 0.93651056,
      "learning_rate": 3.975291307164829e-06,
      "loss": 0.96001345,
      "num_input_tokens_seen": 13915150,
      "step": 654,
      "time_per_iteration": 2.8129379749298096
    },
    {
      "auxiliary_loss_clip": 0.01401579,
      "auxiliary_loss_mlp": 0.01031796,
      "balance_loss_clip": 1.06469536,
      "balance_loss_mlp": 1.02234268,
      "epoch": 0.07875909336860458,
      "flos": 15158684822400.0,
      "grad_norm": 2.475213529453682,
      "language_loss": 0.85347223,
      "learning_rate": 3.975169089570306e-06,
      "loss": 0.87780601,
      "num_input_tokens_seen": 13933525,
      "step": 655,
      "time_per_iteration": 2.8161113262176514
    },
    {
      "auxiliary_loss_clip": 0.01301611,
      "auxiliary_loss_mlp": 0.0103798,
      "balance_loss_clip": 1.06883311,
      "balance_loss_mlp": 1.02731133,
      "epoch": 0.07887933625924368,
      "flos": 22236857233920.0,
      "grad_norm": 2.22432295673805,
      "language_loss": 0.91681844,
      "learning_rate": 3.975046572344202e-06,
      "loss": 0.94021434,
      "num_input_tokens_seen": 13949985,
      "step": 656,
      "time_per_iteration": 2.73071026802063
    },
    {
      "auxiliary_loss_clip": 0.01414539,
      "auxiliary_loss_mlp": 0.01035784,
      "balance_loss_clip": 1.06401336,
      "balance_loss_mlp": 1.02551389,
      "epoch": 0.07899957914988276,
      "flos": 20777734955520.0,
      "grad_norm": 1.80850665105051,
      "language_loss": 0.71157044,
      "learning_rate": 3.974923755505103e-06,
      "loss": 0.73607361,
      "num_input_tokens_seen": 13969215,
      "step": 657,
      "time_per_iteration": 2.7933425903320312
    },
    {
      "auxiliary_loss_clip": 0.01408148,
      "auxiliary_loss_mlp": 0.01036315,
      "balance_loss_clip": 1.06573009,
      "balance_loss_mlp": 1.02629519,
      "epoch": 0.07911982204052186,
      "flos": 23003047267200.0,
      "grad_norm": 2.418848461148237,
      "language_loss": 0.91312742,
      "learning_rate": 3.974800639071641e-06,
      "loss": 0.93757206,
      "num_input_tokens_seen": 13989935,
      "step": 658,
      "time_per_iteration": 2.7976441383361816
    },
    {
      "auxiliary_loss_clip": 0.01503583,
      "auxiliary_loss_mlp": 0.02582603,
      "balance_loss_clip": 1.06052828,
      "balance_loss_mlp": 0.99993908,
      "epoch": 0.07924006493116094,
      "flos": 23111389664640.0,
      "grad_norm": 3.827365474743317,
      "language_loss": 1.00730145,
      "learning_rate": 3.974677223062492e-06,
      "loss": 1.04816329,
      "num_input_tokens_seen": 14007150,
      "step": 659,
      "time_per_iteration": 2.994826316833496
    },
    {
      "auxiliary_loss_clip": 0.01359994,
      "auxiliary_loss_mlp": 0.01029487,
      "balance_loss_clip": 1.07393765,
      "balance_loss_mlp": 1.0203855,
      "epoch": 0.07936030782180004,
      "flos": 16472153450880.0,
      "grad_norm": 2.0519239915557663,
      "language_loss": 0.74577761,
      "learning_rate": 3.974553507496378e-06,
      "loss": 0.76967239,
      "num_input_tokens_seen": 14025725,
      "step": 660,
      "time_per_iteration": 3.0851714611053467
    },
    {
      "auxiliary_loss_clip": 0.0135571,
      "auxiliary_loss_mlp": 0.01042213,
      "balance_loss_clip": 1.07052684,
      "balance_loss_mlp": 1.03188324,
      "epoch": 0.07948055071243913,
      "flos": 23733290764800.0,
      "grad_norm": 2.7590544493417344,
      "language_loss": 0.89596009,
      "learning_rate": 3.974429492392068e-06,
      "loss": 0.91993934,
      "num_input_tokens_seen": 14045750,
      "step": 661,
      "time_per_iteration": 2.8539884090423584
    },
    {
      "auxiliary_loss_clip": 0.01263218,
      "auxiliary_loss_mlp": 0.02581248,
      "balance_loss_clip": 1.07579803,
      "balance_loss_mlp": 0.99993646,
      "epoch": 0.07960079360307822,
      "flos": 19573326996480.0,
      "grad_norm": 2.160894253854102,
      "language_loss": 0.91024911,
      "learning_rate": 3.974305177768373e-06,
      "loss": 0.94869375,
      "num_input_tokens_seen": 14063960,
      "step": 662,
      "time_per_iteration": 2.6899778842926025
    },
    {
      "auxiliary_loss_clip": 0.014046,
      "auxiliary_loss_mlp": 0.01038017,
      "balance_loss_clip": 1.06904674,
      "balance_loss_mlp": 1.02746129,
      "epoch": 0.07972103649371731,
      "flos": 23513409659520.0,
      "grad_norm": 2.4196737054207365,
      "language_loss": 0.87168801,
      "learning_rate": 3.974180563644152e-06,
      "loss": 0.89611423,
      "num_input_tokens_seen": 14082525,
      "step": 663,
      "time_per_iteration": 2.735672950744629
    },
    {
      "auxiliary_loss_clip": 0.01364285,
      "auxiliary_loss_mlp": 0.0104179,
      "balance_loss_clip": 1.06969213,
      "balance_loss_mlp": 1.03160977,
      "epoch": 0.0798412793843564,
      "flos": 16726867770240.0,
      "grad_norm": 2.2317896603682508,
      "language_loss": 0.88769913,
      "learning_rate": 3.97405565003831e-06,
      "loss": 0.91175985,
      "num_input_tokens_seen": 14098610,
      "step": 664,
      "time_per_iteration": 2.7124035358428955
    },
    {
      "auxiliary_loss_clip": 0.01411397,
      "auxiliary_loss_mlp": 0.01038003,
      "balance_loss_clip": 1.06897187,
      "balance_loss_mlp": 1.02776933,
      "epoch": 0.07996152227499549,
      "flos": 18223337214720.0,
      "grad_norm": 2.2202219116534176,
      "language_loss": 0.78043801,
      "learning_rate": 3.973930436969794e-06,
      "loss": 0.804932,
      "num_input_tokens_seen": 14117065,
      "step": 665,
      "time_per_iteration": 2.7314093112945557
    },
    {
      "auxiliary_loss_clip": 0.01351569,
      "auxiliary_loss_mlp": 0.01038754,
      "balance_loss_clip": 1.06511188,
      "balance_loss_mlp": 1.02831078,
      "epoch": 0.08008176516563459,
      "flos": 20594877793920.0,
      "grad_norm": 2.2298956317520697,
      "language_loss": 0.85867828,
      "learning_rate": 3.973804924457602e-06,
      "loss": 0.88258147,
      "num_input_tokens_seen": 14135145,
      "step": 666,
      "time_per_iteration": 3.6435461044311523
    },
    {
      "auxiliary_loss_clip": 0.01356209,
      "auxiliary_loss_mlp": 0.01046172,
      "balance_loss_clip": 1.06595874,
      "balance_loss_mlp": 1.03523493,
      "epoch": 0.08020200805627367,
      "flos": 31834306863360.0,
      "grad_norm": 1.8231781356165335,
      "language_loss": 0.85696983,
      "learning_rate": 3.973679112520771e-06,
      "loss": 0.88099372,
      "num_input_tokens_seen": 14156860,
      "step": 667,
      "time_per_iteration": 2.7839086055755615
    },
    {
      "auxiliary_loss_clip": 0.01402569,
      "auxiliary_loss_mlp": 0.01034475,
      "balance_loss_clip": 1.06276011,
      "balance_loss_mlp": 1.02384806,
      "epoch": 0.08032225094691277,
      "flos": 17783503176960.0,
      "grad_norm": 1.9005585747904765,
      "language_loss": 0.98955309,
      "learning_rate": 3.973553001178389e-06,
      "loss": 1.01392365,
      "num_input_tokens_seen": 14174365,
      "step": 668,
      "time_per_iteration": 2.7513139247894287
    },
    {
      "auxiliary_loss_clip": 0.01409803,
      "auxiliary_loss_mlp": 0.01041733,
      "balance_loss_clip": 1.07002664,
      "balance_loss_mlp": 1.03115964,
      "epoch": 0.08044249383755185,
      "flos": 24061693835520.0,
      "grad_norm": 2.6202573737832022,
      "language_loss": 0.75817412,
      "learning_rate": 3.973426590449585e-06,
      "loss": 0.78268945,
      "num_input_tokens_seen": 14192320,
      "step": 669,
      "time_per_iteration": 3.7121331691741943
    },
    {
      "auxiliary_loss_clip": 0.01456722,
      "auxiliary_loss_mlp": 0.01043782,
      "balance_loss_clip": 1.06553507,
      "balance_loss_mlp": 1.03324413,
      "epoch": 0.08056273672819095,
      "flos": 18223624523520.0,
      "grad_norm": 2.9911098963833727,
      "language_loss": 0.75127769,
      "learning_rate": 3.9732998803535364e-06,
      "loss": 0.77628273,
      "num_input_tokens_seen": 14210380,
      "step": 670,
      "time_per_iteration": 3.675887107849121
    },
    {
      "auxiliary_loss_clip": 0.01262034,
      "auxiliary_loss_mlp": 0.01049825,
      "balance_loss_clip": 1.07189822,
      "balance_loss_mlp": 1.0396862,
      "epoch": 0.08068297961883003,
      "flos": 19676856971520.0,
      "grad_norm": 2.7144893555440266,
      "language_loss": 0.85259402,
      "learning_rate": 3.973172870909465e-06,
      "loss": 0.87571263,
      "num_input_tokens_seen": 14225145,
      "step": 671,
      "time_per_iteration": 2.6463539600372314
    },
    {
      "auxiliary_loss_clip": 0.01364566,
      "auxiliary_loss_mlp": 0.01037014,
      "balance_loss_clip": 1.06731057,
      "balance_loss_mlp": 1.02623761,
      "epoch": 0.08080322250946913,
      "flos": 23148736830720.0,
      "grad_norm": 3.223350592851593,
      "language_loss": 0.80817735,
      "learning_rate": 3.973045562136638e-06,
      "loss": 0.83219314,
      "num_input_tokens_seen": 14241960,
      "step": 672,
      "time_per_iteration": 2.7328684329986572
    },
    {
      "auxiliary_loss_clip": 0.0131357,
      "auxiliary_loss_mlp": 0.01038761,
      "balance_loss_clip": 1.06932211,
      "balance_loss_mlp": 1.0285809,
      "epoch": 0.08092346540010822,
      "flos": 21763626526080.0,
      "grad_norm": 2.197025500857738,
      "language_loss": 0.91614586,
      "learning_rate": 3.972917954054368e-06,
      "loss": 0.93966913,
      "num_input_tokens_seen": 14260515,
      "step": 673,
      "time_per_iteration": 3.5203258991241455
    },
    {
      "auxiliary_loss_clip": 0.01358625,
      "auxiliary_loss_mlp": 0.01042499,
      "balance_loss_clip": 1.06962013,
      "balance_loss_mlp": 1.0314604,
      "epoch": 0.08104370829074731,
      "flos": 21032485188480.0,
      "grad_norm": 3.0291848761949223,
      "language_loss": 0.82208902,
      "learning_rate": 3.972790046682013e-06,
      "loss": 0.84610027,
      "num_input_tokens_seen": 14279190,
      "step": 674,
      "time_per_iteration": 2.8011114597320557
    },
    {
      "auxiliary_loss_clip": 0.01409444,
      "auxiliary_loss_mlp": 0.01037354,
      "balance_loss_clip": 1.06315792,
      "balance_loss_mlp": 1.02735853,
      "epoch": 0.0811639511813864,
      "flos": 20083186598400.0,
      "grad_norm": 1.9748189433286991,
      "language_loss": 0.79138386,
      "learning_rate": 3.972661840038977e-06,
      "loss": 0.81585187,
      "num_input_tokens_seen": 14299480,
      "step": 675,
      "time_per_iteration": 2.7794582843780518
    },
    {
      "auxiliary_loss_clip": 0.01314533,
      "auxiliary_loss_mlp": 0.01050385,
      "balance_loss_clip": 1.07207286,
      "balance_loss_mlp": 1.04024649,
      "epoch": 0.08128419407202549,
      "flos": 16836718538880.0,
      "grad_norm": 2.3678113288527345,
      "language_loss": 0.83488858,
      "learning_rate": 3.972533334144707e-06,
      "loss": 0.85853773,
      "num_input_tokens_seen": 14316405,
      "step": 676,
      "time_per_iteration": 2.7211384773254395
    },
    {
      "auxiliary_loss_clip": 0.01314279,
      "auxiliary_loss_mlp": 0.01040187,
      "balance_loss_clip": 1.06811666,
      "balance_loss_mlp": 1.0297507,
      "epoch": 0.08140443696266458,
      "flos": 23769273214080.0,
      "grad_norm": 2.4270072497877297,
      "language_loss": 0.78467476,
      "learning_rate": 3.972404529018699e-06,
      "loss": 0.80821943,
      "num_input_tokens_seen": 14336265,
      "step": 677,
      "time_per_iteration": 2.722174644470215
    },
    {
      "auxiliary_loss_clip": 0.01360077,
      "auxiliary_loss_mlp": 0.0103198,
      "balance_loss_clip": 1.06197524,
      "balance_loss_mlp": 1.02218711,
      "epoch": 0.08152467985330367,
      "flos": 24390132819840.0,
      "grad_norm": 2.839141865278411,
      "language_loss": 0.85579526,
      "learning_rate": 3.972275424680493e-06,
      "loss": 0.8797158,
      "num_input_tokens_seen": 14356375,
      "step": 678,
      "time_per_iteration": 2.739910840988159
    },
    {
      "auxiliary_loss_clip": 0.01259816,
      "auxiliary_loss_mlp": 0.01037663,
      "balance_loss_clip": 1.07028222,
      "balance_loss_mlp": 1.02645707,
      "epoch": 0.08164492274394276,
      "flos": 19317750750720.0,
      "grad_norm": 2.3543173210471373,
      "language_loss": 0.91800451,
      "learning_rate": 3.972146021149673e-06,
      "loss": 0.94097924,
      "num_input_tokens_seen": 14374650,
      "step": 679,
      "time_per_iteration": 2.6989400386810303
    },
    {
      "auxiliary_loss_clip": 0.01404949,
      "auxiliary_loss_mlp": 0.01043599,
      "balance_loss_clip": 1.06414425,
      "balance_loss_mlp": 1.03450322,
      "epoch": 0.08176516563458186,
      "flos": 14830461319680.0,
      "grad_norm": 2.8730445061163237,
      "language_loss": 0.79016995,
      "learning_rate": 3.972016318445868e-06,
      "loss": 0.81465542,
      "num_input_tokens_seen": 14392650,
      "step": 680,
      "time_per_iteration": 2.8389296531677246
    },
    {
      "auxiliary_loss_clip": 0.01313591,
      "auxiliary_loss_mlp": 0.01037415,
      "balance_loss_clip": 1.07077038,
      "balance_loss_mlp": 1.02673948,
      "epoch": 0.08188540852522094,
      "flos": 22602320161920.0,
      "grad_norm": 1.9830962458314851,
      "language_loss": 0.92355907,
      "learning_rate": 3.971886316588757e-06,
      "loss": 0.94706911,
      "num_input_tokens_seen": 14413155,
      "step": 681,
      "time_per_iteration": 2.8532533645629883
    },
    {
      "auxiliary_loss_clip": 0.01401912,
      "auxiliary_loss_mlp": 0.01036309,
      "balance_loss_clip": 1.06584013,
      "balance_loss_mlp": 1.0257647,
      "epoch": 0.08200565141586004,
      "flos": 19463727623040.0,
      "grad_norm": 2.3779268328919008,
      "language_loss": 0.73717785,
      "learning_rate": 3.9717560155980595e-06,
      "loss": 0.76155996,
      "num_input_tokens_seen": 14428805,
      "step": 682,
      "time_per_iteration": 2.83217716217041
    },
    {
      "auxiliary_loss_clip": 0.01308499,
      "auxiliary_loss_mlp": 0.01043426,
      "balance_loss_clip": 1.06761336,
      "balance_loss_mlp": 1.03359127,
      "epoch": 0.08212589430649912,
      "flos": 20594662312320.0,
      "grad_norm": 2.313076678205687,
      "language_loss": 0.9213745,
      "learning_rate": 3.971625415493542e-06,
      "loss": 0.94489378,
      "num_input_tokens_seen": 14447125,
      "step": 683,
      "time_per_iteration": 2.807955265045166
    },
    {
      "auxiliary_loss_clip": 0.01402447,
      "auxiliary_loss_mlp": 0.01034553,
      "balance_loss_clip": 1.06333137,
      "balance_loss_mlp": 1.02419972,
      "epoch": 0.08224613719713822,
      "flos": 25953611086080.0,
      "grad_norm": 2.2640871146991635,
      "language_loss": 0.87610269,
      "learning_rate": 3.971494516295017e-06,
      "loss": 0.9004727,
      "num_input_tokens_seen": 14466575,
      "step": 684,
      "time_per_iteration": 2.9822959899902344
    },
    {
      "auxiliary_loss_clip": 0.01405906,
      "auxiliary_loss_mlp": 0.01036605,
      "balance_loss_clip": 1.06119478,
      "balance_loss_mlp": 1.02743173,
      "epoch": 0.08236638008777732,
      "flos": 23768734510080.0,
      "grad_norm": 38.265939661304444,
      "language_loss": 0.85398424,
      "learning_rate": 3.971363318022341e-06,
      "loss": 0.87840933,
      "num_input_tokens_seen": 14487915,
      "step": 685,
      "time_per_iteration": 2.8582754135131836
    },
    {
      "auxiliary_loss_clip": 0.01362062,
      "auxiliary_loss_mlp": 0.01029425,
      "balance_loss_clip": 1.06243217,
      "balance_loss_mlp": 1.0195843,
      "epoch": 0.0824866229784164,
      "flos": 38799144887040.0,
      "grad_norm": 1.876856375304612,
      "language_loss": 0.68467045,
      "learning_rate": 3.971231820695417e-06,
      "loss": 0.70858538,
      "num_input_tokens_seen": 14511530,
      "step": 686,
      "time_per_iteration": 3.0165061950683594
    },
    {
      "auxiliary_loss_clip": 0.01359633,
      "auxiliary_loss_mlp": 0.01037596,
      "balance_loss_clip": 1.06579626,
      "balance_loss_mlp": 1.02769578,
      "epoch": 0.0826068658690555,
      "flos": 23107762391040.0,
      "grad_norm": 1.9152014785182616,
      "language_loss": 0.80927873,
      "learning_rate": 3.971100024334193e-06,
      "loss": 0.833251,
      "num_input_tokens_seen": 14529050,
      "step": 687,
      "time_per_iteration": 2.8164446353912354
    },
    {
      "auxiliary_loss_clip": 0.01393606,
      "auxiliary_loss_mlp": 0.01040164,
      "balance_loss_clip": 1.05856955,
      "balance_loss_mlp": 1.03076458,
      "epoch": 0.08272710875969458,
      "flos": 21136374299520.0,
      "grad_norm": 2.776459366740247,
      "language_loss": 0.86205864,
      "learning_rate": 3.970967928958663e-06,
      "loss": 0.88639635,
      "num_input_tokens_seen": 14546165,
      "step": 688,
      "time_per_iteration": 2.8166511058807373
    },
    {
      "auxiliary_loss_clip": 0.01458614,
      "auxiliary_loss_mlp": 0.01039283,
      "balance_loss_clip": 1.05963683,
      "balance_loss_mlp": 1.02906644,
      "epoch": 0.08284735165033368,
      "flos": 19063000517760.0,
      "grad_norm": 1.6951955706755704,
      "language_loss": 0.83538759,
      "learning_rate": 3.970835534588865e-06,
      "loss": 0.86036658,
      "num_input_tokens_seen": 14563660,
      "step": 689,
      "time_per_iteration": 2.8753840923309326
    },
    {
      "auxiliary_loss_clip": 0.01363196,
      "auxiliary_loss_mlp": 0.01036394,
      "balance_loss_clip": 1.07137954,
      "balance_loss_mlp": 1.02632046,
      "epoch": 0.08296759454097276,
      "flos": 16727442387840.0,
      "grad_norm": 2.2110592799773277,
      "language_loss": 0.85439324,
      "learning_rate": 3.970702841244883e-06,
      "loss": 0.87838912,
      "num_input_tokens_seen": 14581980,
      "step": 690,
      "time_per_iteration": 2.8695499897003174
    },
    {
      "auxiliary_loss_clip": 0.01312252,
      "auxiliary_loss_mlp": 0.01035803,
      "balance_loss_clip": 1.07136893,
      "balance_loss_mlp": 1.02616477,
      "epoch": 0.08308783743161186,
      "flos": 18004928567040.0,
      "grad_norm": 1.9279812050122374,
      "language_loss": 0.82704985,
      "learning_rate": 3.970569848946847e-06,
      "loss": 0.85053039,
      "num_input_tokens_seen": 14601795,
      "step": 691,
      "time_per_iteration": 2.754331350326538
    },
    {
      "auxiliary_loss_clip": 0.01299622,
      "auxiliary_loss_mlp": 0.01042106,
      "balance_loss_clip": 1.06768799,
      "balance_loss_mlp": 1.03227699,
      "epoch": 0.08320808032225095,
      "flos": 15079788599040.0,
      "grad_norm": 2.109284883161969,
      "language_loss": 0.83064294,
      "learning_rate": 3.970436557714932e-06,
      "loss": 0.85406017,
      "num_input_tokens_seen": 14618315,
      "step": 692,
      "time_per_iteration": 3.618786573410034
    },
    {
      "auxiliary_loss_clip": 0.01352168,
      "auxiliary_loss_mlp": 0.01036316,
      "balance_loss_clip": 1.06718338,
      "balance_loss_mlp": 1.02587318,
      "epoch": 0.08332832321289003,
      "flos": 22383085501440.0,
      "grad_norm": 3.1218122666125074,
      "language_loss": 0.86961806,
      "learning_rate": 3.970302967569358e-06,
      "loss": 0.89350289,
      "num_input_tokens_seen": 14636905,
      "step": 693,
      "time_per_iteration": 2.7313051223754883
    },
    {
      "auxiliary_loss_clip": 0.0131378,
      "auxiliary_loss_mlp": 0.01037024,
      "balance_loss_clip": 1.07343316,
      "balance_loss_mlp": 1.02636659,
      "epoch": 0.08344856610352913,
      "flos": 24717386655360.0,
      "grad_norm": 2.112607004014389,
      "language_loss": 0.68074435,
      "learning_rate": 3.9701690785303896e-06,
      "loss": 0.70425242,
      "num_input_tokens_seen": 14656100,
      "step": 694,
      "time_per_iteration": 2.7107737064361572
    },
    {
      "auxiliary_loss_clip": 0.01310597,
      "auxiliary_loss_mlp": 0.01039408,
      "balance_loss_clip": 1.06735635,
      "balance_loss_mlp": 1.02985334,
      "epoch": 0.08356880899416821,
      "flos": 25370206387200.0,
      "grad_norm": 2.1609038756217744,
      "language_loss": 0.88638783,
      "learning_rate": 3.970034890618339e-06,
      "loss": 0.90988779,
      "num_input_tokens_seen": 14675790,
      "step": 695,
      "time_per_iteration": 3.622317314147949
    },
    {
      "auxiliary_loss_clip": 0.01296117,
      "auxiliary_loss_mlp": 0.01033265,
      "balance_loss_clip": 1.06408632,
      "balance_loss_mlp": 1.02334058,
      "epoch": 0.08368905188480731,
      "flos": 24353072962560.0,
      "grad_norm": 2.0763870966544937,
      "language_loss": 0.87794471,
      "learning_rate": 3.969900403853562e-06,
      "loss": 0.90123856,
      "num_input_tokens_seen": 14694830,
      "step": 696,
      "time_per_iteration": 3.5528438091278076
    },
    {
      "auxiliary_loss_clip": 0.01260266,
      "auxiliary_loss_mlp": 0.0103513,
      "balance_loss_clip": 1.07046449,
      "balance_loss_mlp": 1.02561069,
      "epoch": 0.08380929477544641,
      "flos": 18037319656320.0,
      "grad_norm": 3.612553247367886,
      "language_loss": 0.77904582,
      "learning_rate": 3.96976561825646e-06,
      "loss": 0.80199981,
      "num_input_tokens_seen": 14711920,
      "step": 697,
      "time_per_iteration": 2.5657222270965576
    },
    {
      "auxiliary_loss_clip": 0.0146017,
      "auxiliary_loss_mlp": 0.01040135,
      "balance_loss_clip": 1.06425536,
      "balance_loss_mlp": 1.02991903,
      "epoch": 0.08392953766608549,
      "flos": 26286287875200.0,
      "grad_norm": 2.2245951677955262,
      "language_loss": 0.86985171,
      "learning_rate": 3.969630533847479e-06,
      "loss": 0.89485472,
      "num_input_tokens_seen": 14730880,
      "step": 698,
      "time_per_iteration": 2.834428071975708
    },
    {
      "auxiliary_loss_clip": 0.01309261,
      "auxiliary_loss_mlp": 0.01042104,
      "balance_loss_clip": 1.06710505,
      "balance_loss_mlp": 1.03262651,
      "epoch": 0.08404978055672459,
      "flos": 22492146170880.0,
      "grad_norm": 2.4310948367085854,
      "language_loss": 0.84461898,
      "learning_rate": 3.969495150647113e-06,
      "loss": 0.86813265,
      "num_input_tokens_seen": 14749050,
      "step": 699,
      "time_per_iteration": 3.5609936714172363
    },
    {
      "auxiliary_loss_clip": 0.01404076,
      "auxiliary_loss_mlp": 0.01041372,
      "balance_loss_clip": 1.06693387,
      "balance_loss_mlp": 1.03165627,
      "epoch": 0.08417002344736367,
      "flos": 24826878288000.0,
      "grad_norm": 1.967039175698942,
      "language_loss": 0.76337361,
      "learning_rate": 3.969359468675899e-06,
      "loss": 0.78782809,
      "num_input_tokens_seen": 14769180,
      "step": 700,
      "time_per_iteration": 2.7711572647094727
    },
    {
      "auxiliary_loss_clip": 0.013017,
      "auxiliary_loss_mlp": 0.0103712,
      "balance_loss_clip": 1.06699646,
      "balance_loss_mlp": 1.02748251,
      "epoch": 0.08429026633800277,
      "flos": 16945922862720.0,
      "grad_norm": 2.027036615795237,
      "language_loss": 0.89415979,
      "learning_rate": 3.969223487954418e-06,
      "loss": 0.917548,
      "num_input_tokens_seen": 14786640,
      "step": 701,
      "time_per_iteration": 2.6487808227539062
    },
    {
      "auxiliary_loss_clip": 0.01458497,
      "auxiliary_loss_mlp": 0.01040474,
      "balance_loss_clip": 1.06358862,
      "balance_loss_mlp": 1.03096104,
      "epoch": 0.08441050922864185,
      "flos": 23841920471040.0,
      "grad_norm": 2.164338354856163,
      "language_loss": 0.82637346,
      "learning_rate": 3.969087208503301e-06,
      "loss": 0.85136324,
      "num_input_tokens_seen": 14806720,
      "step": 702,
      "time_per_iteration": 2.8083207607269287
    },
    {
      "auxiliary_loss_clip": 0.01392182,
      "auxiliary_loss_mlp": 0.01035702,
      "balance_loss_clip": 1.06284499,
      "balance_loss_mlp": 1.02566481,
      "epoch": 0.08453075211928095,
      "flos": 25520205582720.0,
      "grad_norm": 2.588409663549193,
      "language_loss": 0.84950829,
      "learning_rate": 3.968950630343219e-06,
      "loss": 0.87378716,
      "num_input_tokens_seen": 14823705,
      "step": 703,
      "time_per_iteration": 2.782013177871704
    },
    {
      "auxiliary_loss_clip": 0.01353019,
      "auxiliary_loss_mlp": 0.01041297,
      "balance_loss_clip": 1.06403077,
      "balance_loss_mlp": 1.0302825,
      "epoch": 0.08465099500992004,
      "flos": 19532496211200.0,
      "grad_norm": 2.1671375730269724,
      "language_loss": 0.9364624,
      "learning_rate": 3.968813753494892e-06,
      "loss": 0.96040553,
      "num_input_tokens_seen": 14841865,
      "step": 704,
      "time_per_iteration": 2.712681770324707
    },
    {
      "auxiliary_loss_clip": 0.01400574,
      "auxiliary_loss_mlp": 0.02582686,
      "balance_loss_clip": 1.05918312,
      "balance_loss_mlp": 1.00011683,
      "epoch": 0.08477123790055913,
      "flos": 29351299403520.0,
      "grad_norm": 2.1762123696503926,
      "language_loss": 0.75796485,
      "learning_rate": 3.968676577979084e-06,
      "loss": 0.79779744,
      "num_input_tokens_seen": 14861415,
      "step": 705,
      "time_per_iteration": 2.8413641452789307
    },
    {
      "auxiliary_loss_clip": 0.01449133,
      "auxiliary_loss_mlp": 0.01041073,
      "balance_loss_clip": 1.06035495,
      "balance_loss_mlp": 1.03153634,
      "epoch": 0.08489148079119822,
      "flos": 18624495283200.0,
      "grad_norm": 2.5156930537807165,
      "language_loss": 0.78596365,
      "learning_rate": 3.968539103816605e-06,
      "loss": 0.81086576,
      "num_input_tokens_seen": 14879215,
      "step": 706,
      "time_per_iteration": 2.8760921955108643
    },
    {
      "auxiliary_loss_clip": 0.01354272,
      "auxiliary_loss_mlp": 0.02582546,
      "balance_loss_clip": 1.06618738,
      "balance_loss_mlp": 1.00014782,
      "epoch": 0.0850117236818373,
      "flos": 23471393725440.0,
      "grad_norm": 2.707649812894315,
      "language_loss": 0.89538515,
      "learning_rate": 3.9684013310283085e-06,
      "loss": 0.9347533,
      "num_input_tokens_seen": 14897900,
      "step": 707,
      "time_per_iteration": 2.7541263103485107
    },
    {
      "auxiliary_loss_clip": 0.01355035,
      "auxiliary_loss_mlp": 0.01039595,
      "balance_loss_clip": 1.0692575,
      "balance_loss_mlp": 1.03001094,
      "epoch": 0.0851319665724764,
      "flos": 40625058896640.0,
      "grad_norm": 2.0822320502986997,
      "language_loss": 0.63641429,
      "learning_rate": 3.9682632596350956e-06,
      "loss": 0.66036057,
      "num_input_tokens_seen": 14919065,
      "step": 708,
      "time_per_iteration": 2.9098339080810547
    },
    {
      "auxiliary_loss_clip": 0.01302547,
      "auxiliary_loss_mlp": 0.01033086,
      "balance_loss_clip": 1.06659746,
      "balance_loss_mlp": 1.02392459,
      "epoch": 0.0852522094631155,
      "flos": 15879554870400.0,
      "grad_norm": 2.0875243227293505,
      "language_loss": 0.78525651,
      "learning_rate": 3.968124889657911e-06,
      "loss": 0.80861282,
      "num_input_tokens_seen": 14934165,
      "step": 709,
      "time_per_iteration": 2.6050527095794678
    },
    {
      "auxiliary_loss_clip": 0.01455483,
      "auxiliary_loss_mlp": 0.01037263,
      "balance_loss_clip": 1.06179881,
      "balance_loss_mlp": 1.02720141,
      "epoch": 0.08537245235375458,
      "flos": 14567091822720.0,
      "grad_norm": 2.5423510809886496,
      "language_loss": 0.90828383,
      "learning_rate": 3.967986221117746e-06,
      "loss": 0.93321133,
      "num_input_tokens_seen": 14950105,
      "step": 710,
      "time_per_iteration": 2.786137580871582
    },
    {
      "auxiliary_loss_clip": 0.0155894,
      "auxiliary_loss_mlp": 0.01038306,
      "balance_loss_clip": 1.05858707,
      "balance_loss_mlp": 1.02811933,
      "epoch": 0.08549269524439368,
      "flos": 26468929555200.0,
      "grad_norm": 2.2721326025294863,
      "language_loss": 0.86787164,
      "learning_rate": 3.967847254035635e-06,
      "loss": 0.89384413,
      "num_input_tokens_seen": 14969490,
      "step": 711,
      "time_per_iteration": 2.8332087993621826
    },
    {
      "auxiliary_loss_clip": 0.01409181,
      "auxiliary_loss_mlp": 0.01040297,
      "balance_loss_clip": 1.06656885,
      "balance_loss_mlp": 1.03042662,
      "epoch": 0.08561293813503276,
      "flos": 13590214565760.0,
      "grad_norm": 2.9004123929465786,
      "language_loss": 0.86519164,
      "learning_rate": 3.967707988432661e-06,
      "loss": 0.88968641,
      "num_input_tokens_seen": 14987195,
      "step": 712,
      "time_per_iteration": 2.7373852729797363
    },
    {
      "auxiliary_loss_clip": 0.01255634,
      "auxiliary_loss_mlp": 0.01033709,
      "balance_loss_clip": 1.06697726,
      "balance_loss_mlp": 1.02327847,
      "epoch": 0.08573318102567186,
      "flos": 26943524979840.0,
      "grad_norm": 2.186666590465645,
      "language_loss": 0.87872839,
      "learning_rate": 3.967568424329949e-06,
      "loss": 0.90162182,
      "num_input_tokens_seen": 15007620,
      "step": 713,
      "time_per_iteration": 2.6712751388549805
    },
    {
      "auxiliary_loss_clip": 0.01289318,
      "auxiliary_loss_mlp": 0.01007289,
      "balance_loss_clip": 1.04996145,
      "balance_loss_mlp": 1.00345039,
      "epoch": 0.08585342391631094,
      "flos": 67302739319040.0,
      "grad_norm": 0.8361809849798827,
      "language_loss": 0.55509484,
      "learning_rate": 3.967428561748671e-06,
      "loss": 0.57806098,
      "num_input_tokens_seen": 15075590,
      "step": 714,
      "time_per_iteration": 3.4481446743011475
    },
    {
      "auxiliary_loss_clip": 0.01443565,
      "auxiliary_loss_mlp": 0.01041699,
      "balance_loss_clip": 1.05588245,
      "balance_loss_mlp": 1.03068399,
      "epoch": 0.08597366680695004,
      "flos": 22456594684800.0,
      "grad_norm": 3.59231187685513,
      "language_loss": 0.87734973,
      "learning_rate": 3.967288400710045e-06,
      "loss": 0.90220237,
      "num_input_tokens_seen": 15095055,
      "step": 715,
      "time_per_iteration": 2.780897617340088
    },
    {
      "auxiliary_loss_clip": 0.01399904,
      "auxiliary_loss_mlp": 0.01033376,
      "balance_loss_clip": 1.0677073,
      "balance_loss_mlp": 1.02401769,
      "epoch": 0.08609390969758914,
      "flos": 23550505430400.0,
      "grad_norm": 2.1290540253941113,
      "language_loss": 0.88680243,
      "learning_rate": 3.9671479412353335e-06,
      "loss": 0.9111352,
      "num_input_tokens_seen": 15113520,
      "step": 716,
      "time_per_iteration": 2.7679481506347656
    },
    {
      "auxiliary_loss_clip": 0.01303525,
      "auxiliary_loss_mlp": 0.01039582,
      "balance_loss_clip": 1.06732881,
      "balance_loss_mlp": 1.02913892,
      "epoch": 0.08621415258822822,
      "flos": 25885848078720.0,
      "grad_norm": 2.5282921618239897,
      "language_loss": 0.7411834,
      "learning_rate": 3.967007183345843e-06,
      "loss": 0.76461452,
      "num_input_tokens_seen": 15133375,
      "step": 717,
      "time_per_iteration": 2.7239208221435547
    },
    {
      "auxiliary_loss_clip": 0.01301153,
      "auxiliary_loss_mlp": 0.01042188,
      "balance_loss_clip": 1.06759024,
      "balance_loss_mlp": 1.03182912,
      "epoch": 0.08633439547886732,
      "flos": 13589568120960.0,
      "grad_norm": 3.8568717781930784,
      "language_loss": 0.89714462,
      "learning_rate": 3.966866127062927e-06,
      "loss": 0.920578,
      "num_input_tokens_seen": 15150500,
      "step": 718,
      "time_per_iteration": 3.5563740730285645
    },
    {
      "auxiliary_loss_clip": 0.0122397,
      "auxiliary_loss_mlp": 0.01004495,
      "balance_loss_clip": 1.05337191,
      "balance_loss_mlp": 1.00070417,
      "epoch": 0.0864546383695064,
      "flos": 57767342434560.0,
      "grad_norm": 0.8653826537885116,
      "language_loss": 0.62683767,
      "learning_rate": 3.966724772407982e-06,
      "loss": 0.6491223,
      "num_input_tokens_seen": 15208015,
      "step": 719,
      "time_per_iteration": 3.0836548805236816
    },
    {
      "auxiliary_loss_clip": 0.01394753,
      "auxiliary_loss_mlp": 0.01037582,
      "balance_loss_clip": 1.06184387,
      "balance_loss_mlp": 1.02821207,
      "epoch": 0.0865748812601455,
      "flos": 20046952753920.0,
      "grad_norm": 2.6773046656118495,
      "language_loss": 0.8850053,
      "learning_rate": 3.966583119402454e-06,
      "loss": 0.90932864,
      "num_input_tokens_seen": 15224780,
      "step": 720,
      "time_per_iteration": 2.748619556427002
    },
    {
      "auxiliary_loss_clip": 0.01300028,
      "auxiliary_loss_mlp": 0.02579728,
      "balance_loss_clip": 1.06684113,
      "balance_loss_mlp": 1.00008512,
      "epoch": 0.08669512415078459,
      "flos": 35262446935680.0,
      "grad_norm": 1.7196564964298402,
      "language_loss": 0.82139409,
      "learning_rate": 3.9664411680678305e-06,
      "loss": 0.86019164,
      "num_input_tokens_seen": 15246535,
      "step": 721,
      "time_per_iteration": 3.7022294998168945
    },
    {
      "auxiliary_loss_clip": 0.01323082,
      "auxiliary_loss_mlp": 0.01004616,
      "balance_loss_clip": 1.04743457,
      "balance_loss_mlp": 1.00077713,
      "epoch": 0.08681536704142367,
      "flos": 65654870048640.0,
      "grad_norm": 0.8435989803055133,
      "language_loss": 0.61431336,
      "learning_rate": 3.966298918425644e-06,
      "loss": 0.63759035,
      "num_input_tokens_seen": 15304025,
      "step": 722,
      "time_per_iteration": 4.195866823196411
    },
    {
      "auxiliary_loss_clip": 0.01310107,
      "auxiliary_loss_mlp": 0.01049261,
      "balance_loss_clip": 1.06789541,
      "balance_loss_mlp": 1.03906298,
      "epoch": 0.08693560993206277,
      "flos": 34529940881280.0,
      "grad_norm": 1.8403463291431765,
      "language_loss": 0.82822466,
      "learning_rate": 3.966156370497476e-06,
      "loss": 0.85181832,
      "num_input_tokens_seen": 15327635,
      "step": 723,
      "time_per_iteration": 2.859050989151001
    },
    {
      "auxiliary_loss_clip": 0.01310252,
      "auxiliary_loss_mlp": 0.01033425,
      "balance_loss_clip": 1.06747699,
      "balance_loss_mlp": 1.02398324,
      "epoch": 0.08705585282270185,
      "flos": 23149419189120.0,
      "grad_norm": 3.084536665437749,
      "language_loss": 0.88859707,
      "learning_rate": 3.96601352430495e-06,
      "loss": 0.91203386,
      "num_input_tokens_seen": 15347405,
      "step": 724,
      "time_per_iteration": 2.7160415649414062
    },
    {
      "auxiliary_loss_clip": 0.01357447,
      "auxiliary_loss_mlp": 0.01039157,
      "balance_loss_clip": 1.07080507,
      "balance_loss_mlp": 1.02987015,
      "epoch": 0.08717609571334095,
      "flos": 29497599498240.0,
      "grad_norm": 1.4745326717067637,
      "language_loss": 0.82751548,
      "learning_rate": 3.965870379869735e-06,
      "loss": 0.85148156,
      "num_input_tokens_seen": 15369450,
      "step": 725,
      "time_per_iteration": 3.714097499847412
    },
    {
      "auxiliary_loss_clip": 0.0130583,
      "auxiliary_loss_mlp": 0.0103607,
      "balance_loss_clip": 1.06499219,
      "balance_loss_mlp": 1.02646172,
      "epoch": 0.08729633860398003,
      "flos": 20667489137280.0,
      "grad_norm": 2.2893605625281834,
      "language_loss": 0.87099385,
      "learning_rate": 3.965726937213547e-06,
      "loss": 0.89441288,
      "num_input_tokens_seen": 15388085,
      "step": 726,
      "time_per_iteration": 2.810105562210083
    },
    {
      "auxiliary_loss_clip": 0.01302987,
      "auxiliary_loss_mlp": 0.0103078,
      "balance_loss_clip": 1.0640558,
      "balance_loss_mlp": 1.01974094,
      "epoch": 0.08741658149461913,
      "flos": 18369493655040.0,
      "grad_norm": 2.1016513308701024,
      "language_loss": 0.81654745,
      "learning_rate": 3.965583196358144e-06,
      "loss": 0.83988512,
      "num_input_tokens_seen": 15407120,
      "step": 727,
      "time_per_iteration": 2.642669677734375
    },
    {
      "auxiliary_loss_clip": 0.01260356,
      "auxiliary_loss_mlp": 0.01035475,
      "balance_loss_clip": 1.07025957,
      "balance_loss_mlp": 1.02559292,
      "epoch": 0.08753682438525823,
      "flos": 18729677283840.0,
      "grad_norm": 2.145429686902692,
      "language_loss": 0.74335688,
      "learning_rate": 3.965439157325335e-06,
      "loss": 0.76631522,
      "num_input_tokens_seen": 15424485,
      "step": 728,
      "time_per_iteration": 2.6900532245635986
    },
    {
      "auxiliary_loss_clip": 0.01354779,
      "auxiliary_loss_mlp": 0.01039442,
      "balance_loss_clip": 1.06310511,
      "balance_loss_mlp": 1.02935112,
      "epoch": 0.08765706727589731,
      "flos": 27776113303680.0,
      "grad_norm": 1.8941370294092865,
      "language_loss": 0.75919729,
      "learning_rate": 3.965294820136968e-06,
      "loss": 0.78313947,
      "num_input_tokens_seen": 15446285,
      "step": 729,
      "time_per_iteration": 2.7971136569976807
    },
    {
      "auxiliary_loss_clip": 0.01348227,
      "auxiliary_loss_mlp": 0.0103766,
      "balance_loss_clip": 1.06530714,
      "balance_loss_mlp": 1.02787328,
      "epoch": 0.08777731016653641,
      "flos": 24389127239040.0,
      "grad_norm": 3.797939621026205,
      "language_loss": 0.87065512,
      "learning_rate": 3.965150184814938e-06,
      "loss": 0.89451396,
      "num_input_tokens_seen": 15465770,
      "step": 730,
      "time_per_iteration": 2.760451316833496
    },
    {
      "auxiliary_loss_clip": 0.01347902,
      "auxiliary_loss_mlp": 0.01036653,
      "balance_loss_clip": 1.06369352,
      "balance_loss_mlp": 1.02640688,
      "epoch": 0.08789755305717549,
      "flos": 21981855605760.0,
      "grad_norm": 2.3761576618973614,
      "language_loss": 0.76776218,
      "learning_rate": 3.965005251381189e-06,
      "loss": 0.79160774,
      "num_input_tokens_seen": 15483705,
      "step": 731,
      "time_per_iteration": 2.7108702659606934
    },
    {
      "auxiliary_loss_clip": 0.01219179,
      "auxiliary_loss_mlp": 0.01008597,
      "balance_loss_clip": 1.0449636,
      "balance_loss_mlp": 1.00497341,
      "epoch": 0.08801779594781459,
      "flos": 58360120583040.0,
      "grad_norm": 0.9518711678504015,
      "language_loss": 0.64535224,
      "learning_rate": 3.964860019857705e-06,
      "loss": 0.66763008,
      "num_input_tokens_seen": 15548620,
      "step": 732,
      "time_per_iteration": 3.2629663944244385
    },
    {
      "auxiliary_loss_clip": 0.01255384,
      "auxiliary_loss_mlp": 0.01035387,
      "balance_loss_clip": 1.07126415,
      "balance_loss_mlp": 1.02574921,
      "epoch": 0.08813803883845367,
      "flos": 23294785530240.0,
      "grad_norm": 1.857181033378721,
      "language_loss": 0.84165937,
      "learning_rate": 3.964714490266518e-06,
      "loss": 0.86456704,
      "num_input_tokens_seen": 15569265,
      "step": 733,
      "time_per_iteration": 2.6731762886047363
    },
    {
      "auxiliary_loss_clip": 0.0121226,
      "auxiliary_loss_mlp": 0.01003133,
      "balance_loss_clip": 1.04130924,
      "balance_loss_mlp": 0.99948531,
      "epoch": 0.08825828172909277,
      "flos": 63424924882560.0,
      "grad_norm": 0.8811400491936832,
      "language_loss": 0.64573991,
      "learning_rate": 3.964568662629706e-06,
      "loss": 0.66789377,
      "num_input_tokens_seen": 15630570,
      "step": 734,
      "time_per_iteration": 3.1388072967529297
    },
    {
      "auxiliary_loss_clip": 0.01303171,
      "auxiliary_loss_mlp": 0.01036064,
      "balance_loss_clip": 1.06472993,
      "balance_loss_mlp": 1.02672434,
      "epoch": 0.08837852461973186,
      "flos": 26720986268160.0,
      "grad_norm": 2.2512747602528447,
      "language_loss": 0.84613991,
      "learning_rate": 3.9644225369693895e-06,
      "loss": 0.86953223,
      "num_input_tokens_seen": 15650870,
      "step": 735,
      "time_per_iteration": 2.7511446475982666
    },
    {
      "auxiliary_loss_clip": 0.01256909,
      "auxiliary_loss_mlp": 0.01033447,
      "balance_loss_clip": 1.07242942,
      "balance_loss_mlp": 1.02328479,
      "epoch": 0.08849876751037095,
      "flos": 27265427688960.0,
      "grad_norm": 4.414156027546238,
      "language_loss": 0.87072128,
      "learning_rate": 3.964276113307735e-06,
      "loss": 0.89362484,
      "num_input_tokens_seen": 15670835,
      "step": 736,
      "time_per_iteration": 2.6838343143463135
    },
    {
      "auxiliary_loss_clip": 0.01396674,
      "auxiliary_loss_mlp": 0.01037603,
      "balance_loss_clip": 1.06355166,
      "balance_loss_mlp": 1.02755404,
      "epoch": 0.08861901040101004,
      "flos": 19828759587840.0,
      "grad_norm": 1.952472192862226,
      "language_loss": 0.80591631,
      "learning_rate": 3.9641293916669574e-06,
      "loss": 0.83025908,
      "num_input_tokens_seen": 15689795,
      "step": 737,
      "time_per_iteration": 2.7900123596191406
    },
    {
      "auxiliary_loss_clip": 0.01401693,
      "auxiliary_loss_mlp": 0.01040277,
      "balance_loss_clip": 1.06600142,
      "balance_loss_mlp": 1.02941763,
      "epoch": 0.08873925329164913,
      "flos": 23658704173440.0,
      "grad_norm": 3.1346043280852465,
      "language_loss": 0.83007139,
      "learning_rate": 3.9639823720693115e-06,
      "loss": 0.85449106,
      "num_input_tokens_seen": 15711650,
      "step": 738,
      "time_per_iteration": 2.7639260292053223
    },
    {
      "auxiliary_loss_clip": 0.01336932,
      "auxiliary_loss_mlp": 0.01027234,
      "balance_loss_clip": 1.06296682,
      "balance_loss_mlp": 1.02401495,
      "epoch": 0.08885949618228822,
      "flos": 71831541893760.0,
      "grad_norm": 0.8493393548171678,
      "language_loss": 0.60027277,
      "learning_rate": 3.963835054537102e-06,
      "loss": 0.62391436,
      "num_input_tokens_seen": 15780615,
      "step": 739,
      "time_per_iteration": 3.313080310821533
    },
    {
      "auxiliary_loss_clip": 0.01353929,
      "auxiliary_loss_mlp": 0.0104956,
      "balance_loss_clip": 1.06317592,
      "balance_loss_mlp": 1.03880191,
      "epoch": 0.08897973907292732,
      "flos": 22346169298560.0,
      "grad_norm": 2.1966853973686424,
      "language_loss": 0.60749972,
      "learning_rate": 3.963687439092676e-06,
      "loss": 0.63153464,
      "num_input_tokens_seen": 15801300,
      "step": 740,
      "time_per_iteration": 2.5952529907226562
    },
    {
      "auxiliary_loss_clip": 0.01302784,
      "auxiliary_loss_mlp": 0.01038374,
      "balance_loss_clip": 1.06880653,
      "balance_loss_mlp": 1.0280025,
      "epoch": 0.0890999819635664,
      "flos": 21251827589760.0,
      "grad_norm": 1.9541694641005884,
      "language_loss": 0.80422556,
      "learning_rate": 3.963539525758427e-06,
      "loss": 0.8276372,
      "num_input_tokens_seen": 15820860,
      "step": 741,
      "time_per_iteration": 2.561311960220337
    },
    {
      "auxiliary_loss_clip": 0.01357794,
      "auxiliary_loss_mlp": 0.01048324,
      "balance_loss_clip": 1.06715453,
      "balance_loss_mlp": 1.03781009,
      "epoch": 0.0892202248542055,
      "flos": 25370888745600.0,
      "grad_norm": 2.254177087945771,
      "language_loss": 0.68021184,
      "learning_rate": 3.9633913145567925e-06,
      "loss": 0.70427299,
      "num_input_tokens_seen": 15841350,
      "step": 742,
      "time_per_iteration": 2.698265552520752
    },
    {
      "auxiliary_loss_clip": 0.01353127,
      "auxiliary_loss_mlp": 0.01037298,
      "balance_loss_clip": 1.06548202,
      "balance_loss_mlp": 1.02773774,
      "epoch": 0.08934046774484458,
      "flos": 24457895827200.0,
      "grad_norm": 2.05393218000211,
      "language_loss": 0.81722689,
      "learning_rate": 3.9632428055102575e-06,
      "loss": 0.84113109,
      "num_input_tokens_seen": 15861360,
      "step": 743,
      "time_per_iteration": 2.7311532497406006
    },
    {
      "auxiliary_loss_clip": 0.01311594,
      "auxiliary_loss_mlp": 0.01044098,
      "balance_loss_clip": 1.0718348,
      "balance_loss_mlp": 1.03367901,
      "epoch": 0.08946071063548368,
      "flos": 35772773414400.0,
      "grad_norm": 7.129120560807403,
      "language_loss": 0.6686846,
      "learning_rate": 3.9630939986413495e-06,
      "loss": 0.69224155,
      "num_input_tokens_seen": 15883160,
      "step": 744,
      "time_per_iteration": 3.6818323135375977
    },
    {
      "auxiliary_loss_clip": 0.01394289,
      "auxiliary_loss_mlp": 0.01043466,
      "balance_loss_clip": 1.0604589,
      "balance_loss_mlp": 1.0335176,
      "epoch": 0.08958095352612276,
      "flos": 14356584167040.0,
      "grad_norm": 1.7190951814224327,
      "language_loss": 0.78193617,
      "learning_rate": 3.962944893972643e-06,
      "loss": 0.80631369,
      "num_input_tokens_seen": 15901610,
      "step": 745,
      "time_per_iteration": 2.730952739715576
    },
    {
      "auxiliary_loss_clip": 0.01350475,
      "auxiliary_loss_mlp": 0.01042963,
      "balance_loss_clip": 1.06693494,
      "balance_loss_mlp": 1.03283024,
      "epoch": 0.08970119641676186,
      "flos": 17853277345920.0,
      "grad_norm": 2.594676716641194,
      "language_loss": 0.90837628,
      "learning_rate": 3.962795491526756e-06,
      "loss": 0.93231058,
      "num_input_tokens_seen": 15918770,
      "step": 746,
      "time_per_iteration": 2.683987855911255
    },
    {
      "auxiliary_loss_clip": 0.01260606,
      "auxiliary_loss_mlp": 0.01047111,
      "balance_loss_clip": 1.07087564,
      "balance_loss_mlp": 1.0361079,
      "epoch": 0.08982143930740095,
      "flos": 20811670329600.0,
      "grad_norm": 2.479035372331964,
      "language_loss": 0.88961422,
      "learning_rate": 3.962645791326354e-06,
      "loss": 0.91269147,
      "num_input_tokens_seen": 15938025,
      "step": 747,
      "time_per_iteration": 2.590918779373169
    },
    {
      "auxiliary_loss_clip": 0.01302054,
      "auxiliary_loss_mlp": 0.0105199,
      "balance_loss_clip": 1.06787634,
      "balance_loss_mlp": 1.04164267,
      "epoch": 0.08994168219804004,
      "flos": 24097712198400.0,
      "grad_norm": 3.071554934680499,
      "language_loss": 0.83462262,
      "learning_rate": 3.962495793394146e-06,
      "loss": 0.85816312,
      "num_input_tokens_seen": 15957215,
      "step": 748,
      "time_per_iteration": 4.373162031173706
    },
    {
      "auxiliary_loss_clip": 0.01155813,
      "auxiliary_loss_mlp": 0.01018394,
      "balance_loss_clip": 1.03262281,
      "balance_loss_mlp": 1.01434124,
      "epoch": 0.09006192508867913,
      "flos": 57188893812480.0,
      "grad_norm": 0.7424081652029518,
      "language_loss": 0.61192942,
      "learning_rate": 3.9623454977528864e-06,
      "loss": 0.63367146,
      "num_input_tokens_seen": 16015870,
      "step": 749,
      "time_per_iteration": 3.0555105209350586
    },
    {
      "auxiliary_loss_clip": 0.0141058,
      "auxiliary_loss_mlp": 0.01041296,
      "balance_loss_clip": 1.06495988,
      "balance_loss_mlp": 1.03069258,
      "epoch": 0.09018216797931822,
      "flos": 20487505063680.0,
      "grad_norm": 7.108364220458012,
      "language_loss": 0.85116076,
      "learning_rate": 3.962194904425375e-06,
      "loss": 0.87567955,
      "num_input_tokens_seen": 16036500,
      "step": 750,
      "time_per_iteration": 2.7749440670013428
    },
    {
      "auxiliary_loss_clip": 0.01300232,
      "auxiliary_loss_mlp": 0.01031096,
      "balance_loss_clip": 1.06624711,
      "balance_loss_mlp": 1.02108228,
      "epoch": 0.09030241086995731,
      "flos": 22638123043200.0,
      "grad_norm": 1.9502881523619102,
      "language_loss": 0.67904025,
      "learning_rate": 3.9620440134344566e-06,
      "loss": 0.70235354,
      "num_input_tokens_seen": 16054655,
      "step": 751,
      "time_per_iteration": 3.659010171890259
    },
    {
      "auxiliary_loss_clip": 0.01403664,
      "auxiliary_loss_mlp": 0.01047113,
      "balance_loss_clip": 1.06683469,
      "balance_loss_mlp": 1.03625298,
      "epoch": 0.09042265376059641,
      "flos": 21871502046720.0,
      "grad_norm": 4.876023168403352,
      "language_loss": 0.82464576,
      "learning_rate": 3.9618928248030215e-06,
      "loss": 0.84915352,
      "num_input_tokens_seen": 16074165,
      "step": 752,
      "time_per_iteration": 2.7422611713409424
    },
    {
      "auxiliary_loss_clip": 0.01303294,
      "auxiliary_loss_mlp": 0.01038701,
      "balance_loss_clip": 1.06926775,
      "balance_loss_mlp": 1.0286994,
      "epoch": 0.0905428966512355,
      "flos": 24316192673280.0,
      "grad_norm": 4.975457258030548,
      "language_loss": 0.82889283,
      "learning_rate": 3.961741338554005e-06,
      "loss": 0.85231268,
      "num_input_tokens_seen": 16092505,
      "step": 753,
      "time_per_iteration": 2.6946282386779785
    },
    {
      "auxiliary_loss_clip": 0.01360675,
      "auxiliary_loss_mlp": 0.01035536,
      "balance_loss_clip": 1.06714964,
      "balance_loss_mlp": 1.02530801,
      "epoch": 0.09066313954187459,
      "flos": 35845061535360.0,
      "grad_norm": 2.242171888305962,
      "language_loss": 0.75739336,
      "learning_rate": 3.9615895547103865e-06,
      "loss": 0.7813555,
      "num_input_tokens_seen": 16116150,
      "step": 754,
      "time_per_iteration": 2.8749704360961914
    },
    {
      "auxiliary_loss_clip": 0.01354448,
      "auxiliary_loss_mlp": 0.01036715,
      "balance_loss_clip": 1.06517148,
      "balance_loss_mlp": 1.02643931,
      "epoch": 0.09078338243251367,
      "flos": 29168729550720.0,
      "grad_norm": 2.526811948931708,
      "language_loss": 0.77857608,
      "learning_rate": 3.961437473295193e-06,
      "loss": 0.80248767,
      "num_input_tokens_seen": 16136295,
      "step": 755,
      "time_per_iteration": 2.7294087409973145
    },
    {
      "auxiliary_loss_clip": 0.01445579,
      "auxiliary_loss_mlp": 0.01036596,
      "balance_loss_clip": 1.056566,
      "balance_loss_mlp": 1.02639771,
      "epoch": 0.09090362532315277,
      "flos": 21907699977600.0,
      "grad_norm": 2.39161622791147,
      "language_loss": 0.72365761,
      "learning_rate": 3.961285094331495e-06,
      "loss": 0.74847937,
      "num_input_tokens_seen": 16154210,
      "step": 756,
      "time_per_iteration": 2.837576150894165
    },
    {
      "auxiliary_loss_clip": 0.01252016,
      "auxiliary_loss_mlp": 0.01037057,
      "balance_loss_clip": 1.06826508,
      "balance_loss_mlp": 1.02702546,
      "epoch": 0.09102386821379185,
      "flos": 27344503480320.0,
      "grad_norm": 2.149406199634861,
      "language_loss": 0.85738033,
      "learning_rate": 3.961132417842406e-06,
      "loss": 0.88027108,
      "num_input_tokens_seen": 16173995,
      "step": 757,
      "time_per_iteration": 2.699751615524292
    },
    {
      "auxiliary_loss_clip": 0.0129628,
      "auxiliary_loss_mlp": 0.01039954,
      "balance_loss_clip": 1.06446385,
      "balance_loss_mlp": 1.0297792,
      "epoch": 0.09114411110443095,
      "flos": 20813501923200.0,
      "grad_norm": 2.5560026975325316,
      "language_loss": 0.75476611,
      "learning_rate": 3.960979443851089e-06,
      "loss": 0.77812839,
      "num_input_tokens_seen": 16191020,
      "step": 758,
      "time_per_iteration": 2.688504219055176
    },
    {
      "auxiliary_loss_clip": 0.01354498,
      "auxiliary_loss_mlp": 0.01038725,
      "balance_loss_clip": 1.06376648,
      "balance_loss_mlp": 1.02823424,
      "epoch": 0.09126435399507005,
      "flos": 26145949438080.0,
      "grad_norm": 1.6302794559122626,
      "language_loss": 0.78947717,
      "learning_rate": 3.96082617238075e-06,
      "loss": 0.81340939,
      "num_input_tokens_seen": 16213645,
      "step": 759,
      "time_per_iteration": 2.7945070266723633
    },
    {
      "auxiliary_loss_clip": 0.01356489,
      "auxiliary_loss_mlp": 0.01038865,
      "balance_loss_clip": 1.06719697,
      "balance_loss_mlp": 1.02886891,
      "epoch": 0.09138459688570913,
      "flos": 24388911757440.0,
      "grad_norm": 2.5863569032400675,
      "language_loss": 0.796229,
      "learning_rate": 3.960672603454639e-06,
      "loss": 0.82018256,
      "num_input_tokens_seen": 16233625,
      "step": 760,
      "time_per_iteration": 2.842303514480591
    },
    {
      "auxiliary_loss_clip": 0.0130027,
      "auxiliary_loss_mlp": 0.01040685,
      "balance_loss_clip": 1.06557357,
      "balance_loss_mlp": 1.03084445,
      "epoch": 0.09150483977634823,
      "flos": 21032664756480.0,
      "grad_norm": 4.103497911934647,
      "language_loss": 0.77315772,
      "learning_rate": 3.960518737096054e-06,
      "loss": 0.79656726,
      "num_input_tokens_seen": 16253255,
      "step": 761,
      "time_per_iteration": 2.6801748275756836
    },
    {
      "auxiliary_loss_clip": 0.0130732,
      "auxiliary_loss_mlp": 0.0103558,
      "balance_loss_clip": 1.06855977,
      "balance_loss_mlp": 1.02530396,
      "epoch": 0.09162508266698731,
      "flos": 22856998567680.0,
      "grad_norm": 2.764526456365886,
      "language_loss": 0.72817451,
      "learning_rate": 3.960364573328334e-06,
      "loss": 0.75160348,
      "num_input_tokens_seen": 16272580,
      "step": 762,
      "time_per_iteration": 2.6751911640167236
    },
    {
      "auxiliary_loss_clip": 0.01407289,
      "auxiliary_loss_mlp": 0.01041637,
      "balance_loss_clip": 1.06383526,
      "balance_loss_mlp": 1.03180814,
      "epoch": 0.0917453255576264,
      "flos": 21724411852800.0,
      "grad_norm": 3.918601558582504,
      "language_loss": 0.88640869,
      "learning_rate": 3.9602101121748675e-06,
      "loss": 0.91089785,
      "num_input_tokens_seen": 16293075,
      "step": 763,
      "time_per_iteration": 2.7737317085266113
    },
    {
      "auxiliary_loss_clip": 0.01353482,
      "auxiliary_loss_mlp": 0.01033903,
      "balance_loss_clip": 1.0684706,
      "balance_loss_mlp": 1.02429521,
      "epoch": 0.0918655684482655,
      "flos": 14609215497600.0,
      "grad_norm": 3.512128758614238,
      "language_loss": 0.72186911,
      "learning_rate": 3.960055353659085e-06,
      "loss": 0.74574298,
      "num_input_tokens_seen": 16310185,
      "step": 764,
      "time_per_iteration": 2.6308038234710693
    },
    {
      "auxiliary_loss_clip": 0.01405598,
      "auxiliary_loss_mlp": 0.01039037,
      "balance_loss_clip": 1.06620336,
      "balance_loss_mlp": 1.02917266,
      "epoch": 0.09198581133890459,
      "flos": 23435016226560.0,
      "grad_norm": 1.9330620763254391,
      "language_loss": 0.83718765,
      "learning_rate": 3.959900297804465e-06,
      "loss": 0.86163396,
      "num_input_tokens_seen": 16330355,
      "step": 765,
      "time_per_iteration": 2.7982285022735596
    },
    {
      "auxiliary_loss_clip": 0.01347831,
      "auxiliary_loss_mlp": 0.01037276,
      "balance_loss_clip": 1.06505632,
      "balance_loss_mlp": 1.02766144,
      "epoch": 0.09210605422954368,
      "flos": 16795887753600.0,
      "grad_norm": 2.015871726975309,
      "language_loss": 0.77593559,
      "learning_rate": 3.9597449446345276e-06,
      "loss": 0.79978669,
      "num_input_tokens_seen": 16347600,
      "step": 766,
      "time_per_iteration": 2.694493532180786
    },
    {
      "auxiliary_loss_clip": 0.01341549,
      "auxiliary_loss_mlp": 0.01035572,
      "balance_loss_clip": 1.06068921,
      "balance_loss_mlp": 1.02641046,
      "epoch": 0.09222629712018277,
      "flos": 22674249146880.0,
      "grad_norm": 2.779871629715447,
      "language_loss": 0.83478492,
      "learning_rate": 3.95958929417284e-06,
      "loss": 0.85855615,
      "num_input_tokens_seen": 16365755,
      "step": 767,
      "time_per_iteration": 2.687399387359619
    },
    {
      "auxiliary_loss_clip": 0.01211678,
      "auxiliary_loss_mlp": 0.01014034,
      "balance_loss_clip": 1.03795266,
      "balance_loss_mlp": 1.01033807,
      "epoch": 0.09234654001082186,
      "flos": 69976756327680.0,
      "grad_norm": 0.7659338506635649,
      "language_loss": 0.5875811,
      "learning_rate": 3.9594333464430145e-06,
      "loss": 0.60983825,
      "num_input_tokens_seen": 16435245,
      "step": 768,
      "time_per_iteration": 3.376556873321533
    },
    {
      "auxiliary_loss_clip": 0.01537862,
      "auxiliary_loss_mlp": 0.01037366,
      "balance_loss_clip": 1.05359113,
      "balance_loss_mlp": 1.02773428,
      "epoch": 0.09246678290146094,
      "flos": 20011437181440.0,
      "grad_norm": 1.8732742692720303,
      "language_loss": 0.88046199,
      "learning_rate": 3.959277101468709e-06,
      "loss": 0.9062143,
      "num_input_tokens_seen": 16454795,
      "step": 769,
      "time_per_iteration": 2.849626064300537
    },
    {
      "auxiliary_loss_clip": 0.01342295,
      "auxiliary_loss_mlp": 0.01032882,
      "balance_loss_clip": 1.06192172,
      "balance_loss_mlp": 1.02283907,
      "epoch": 0.09258702579210004,
      "flos": 17747448900480.0,
      "grad_norm": 3.159480831027179,
      "language_loss": 0.78738081,
      "learning_rate": 3.959120559273624e-06,
      "loss": 0.81113255,
      "num_input_tokens_seen": 16472580,
      "step": 770,
      "time_per_iteration": 3.7400615215301514
    },
    {
      "auxiliary_loss_clip": 0.01340639,
      "auxiliary_loss_mlp": 0.01038448,
      "balance_loss_clip": 1.06508923,
      "balance_loss_mlp": 1.02885747,
      "epoch": 0.09270726868273914,
      "flos": 20886544229760.0,
      "grad_norm": 1.9269695621306626,
      "language_loss": 0.83555305,
      "learning_rate": 3.958963719881509e-06,
      "loss": 0.85934401,
      "num_input_tokens_seen": 16490670,
      "step": 771,
      "time_per_iteration": 2.670522451400757
    },
    {
      "auxiliary_loss_clip": 0.01303509,
      "auxiliary_loss_mlp": 0.0103509,
      "balance_loss_clip": 1.06878364,
      "balance_loss_mlp": 1.02583325,
      "epoch": 0.09282751157337822,
      "flos": 17015697031680.0,
      "grad_norm": 2.005401622591965,
      "language_loss": 0.93735003,
      "learning_rate": 3.958806583316154e-06,
      "loss": 0.96073592,
      "num_input_tokens_seen": 16508640,
      "step": 772,
      "time_per_iteration": 2.6314430236816406
    },
    {
      "auxiliary_loss_clip": 0.01256161,
      "auxiliary_loss_mlp": 0.01045672,
      "balance_loss_clip": 1.07044625,
      "balance_loss_mlp": 1.03589094,
      "epoch": 0.09294775446401732,
      "flos": 32523647748480.0,
      "grad_norm": 3.8376339795513155,
      "language_loss": 0.78951025,
      "learning_rate": 3.9586491496013985e-06,
      "loss": 0.81252855,
      "num_input_tokens_seen": 16531035,
      "step": 773,
      "time_per_iteration": 3.629610776901245
    },
    {
      "auxiliary_loss_clip": 0.01306247,
      "auxiliary_loss_mlp": 0.01041585,
      "balance_loss_clip": 1.06667233,
      "balance_loss_mlp": 1.03184509,
      "epoch": 0.0930679973546564,
      "flos": 18259750627200.0,
      "grad_norm": 2.0153414805847074,
      "language_loss": 0.82961088,
      "learning_rate": 3.958491418761124e-06,
      "loss": 0.85308921,
      "num_input_tokens_seen": 16548605,
      "step": 774,
      "time_per_iteration": 3.608821392059326
    },
    {
      "auxiliary_loss_clip": 0.01351853,
      "auxiliary_loss_mlp": 0.01039041,
      "balance_loss_clip": 1.06184685,
      "balance_loss_mlp": 1.02861643,
      "epoch": 0.0931882402452955,
      "flos": 21099745405440.0,
      "grad_norm": 2.1691343485274164,
      "language_loss": 0.72433114,
      "learning_rate": 3.958333390819258e-06,
      "loss": 0.74824011,
      "num_input_tokens_seen": 16565535,
      "step": 775,
      "time_per_iteration": 2.753573417663574
    },
    {
      "auxiliary_loss_clip": 0.0125264,
      "auxiliary_loss_mlp": 0.01036662,
      "balance_loss_clip": 1.06867456,
      "balance_loss_mlp": 1.0263865,
      "epoch": 0.0933084831359346,
      "flos": 24207275658240.0,
      "grad_norm": 2.062264720239107,
      "language_loss": 0.80244809,
      "learning_rate": 3.9581750657997754e-06,
      "loss": 0.82534111,
      "num_input_tokens_seen": 16584900,
      "step": 776,
      "time_per_iteration": 2.655562162399292
    },
    {
      "auxiliary_loss_clip": 0.01344805,
      "auxiliary_loss_mlp": 0.01029036,
      "balance_loss_clip": 1.06048727,
      "balance_loss_mlp": 1.01971364,
      "epoch": 0.09342872602657368,
      "flos": 25480272637440.0,
      "grad_norm": 2.394105679354208,
      "language_loss": 0.89398283,
      "learning_rate": 3.95801644372669e-06,
      "loss": 0.91772127,
      "num_input_tokens_seen": 16604805,
      "step": 777,
      "time_per_iteration": 3.6469974517822266
    },
    {
      "auxiliary_loss_clip": 0.01357896,
      "auxiliary_loss_mlp": 0.01046417,
      "balance_loss_clip": 1.06303811,
      "balance_loss_mlp": 1.03660059,
      "epoch": 0.09354896891721277,
      "flos": 23149060053120.0,
      "grad_norm": 3.3083158959730037,
      "language_loss": 0.84337449,
      "learning_rate": 3.957857524624068e-06,
      "loss": 0.86741757,
      "num_input_tokens_seen": 16623685,
      "step": 778,
      "time_per_iteration": 2.7452552318573
    },
    {
      "auxiliary_loss_clip": 0.01349341,
      "auxiliary_loss_mlp": 0.01038679,
      "balance_loss_clip": 1.06181586,
      "balance_loss_mlp": 1.0291605,
      "epoch": 0.09366921180785186,
      "flos": 24279563779200.0,
      "grad_norm": 1.7330541563695752,
      "language_loss": 0.89551222,
      "learning_rate": 3.957698308516016e-06,
      "loss": 0.91939247,
      "num_input_tokens_seen": 16644985,
      "step": 779,
      "time_per_iteration": 2.7227189540863037
    },
    {
      "auxiliary_loss_clip": 0.01298793,
      "auxiliary_loss_mlp": 0.02578303,
      "balance_loss_clip": 1.06932855,
      "balance_loss_mlp": 1.00015235,
      "epoch": 0.09378945469849095,
      "flos": 18730036419840.0,
      "grad_norm": 2.0063095413566665,
      "language_loss": 0.82348573,
      "learning_rate": 3.957538795426688e-06,
      "loss": 0.86225671,
      "num_input_tokens_seen": 16662410,
      "step": 780,
      "time_per_iteration": 2.6857783794403076
    },
    {
      "auxiliary_loss_clip": 0.01354879,
      "auxiliary_loss_mlp": 0.01035411,
      "balance_loss_clip": 1.06599164,
      "balance_loss_mlp": 1.02592802,
      "epoch": 0.09390969758913004,
      "flos": 23218834222080.0,
      "grad_norm": 5.292314124655492,
      "language_loss": 0.77017176,
      "learning_rate": 3.9573789853802804e-06,
      "loss": 0.79407465,
      "num_input_tokens_seen": 16680885,
      "step": 781,
      "time_per_iteration": 2.6873743534088135
    },
    {
      "auxiliary_loss_clip": 0.0135271,
      "auxiliary_loss_mlp": 0.02580956,
      "balance_loss_clip": 1.06863463,
      "balance_loss_mlp": 1.00016832,
      "epoch": 0.09402994047976913,
      "flos": 19646728439040.0,
      "grad_norm": 1.9796106928531685,
      "language_loss": 0.74743629,
      "learning_rate": 3.957218878401037e-06,
      "loss": 0.78677291,
      "num_input_tokens_seen": 16699375,
      "step": 782,
      "time_per_iteration": 2.748136520385742
    },
    {
      "auxiliary_loss_clip": 0.01258772,
      "auxiliary_loss_mlp": 0.01039207,
      "balance_loss_clip": 1.07331872,
      "balance_loss_mlp": 1.02903223,
      "epoch": 0.09415018337040823,
      "flos": 29420463041280.0,
      "grad_norm": 1.8937080079344857,
      "language_loss": 0.89350581,
      "learning_rate": 3.957058474513246e-06,
      "loss": 0.91648555,
      "num_input_tokens_seen": 16719230,
      "step": 783,
      "time_per_iteration": 2.6662962436676025
    },
    {
      "auxiliary_loss_clip": 0.0130446,
      "auxiliary_loss_mlp": 0.01039295,
      "balance_loss_clip": 1.07030725,
      "balance_loss_mlp": 1.02987766,
      "epoch": 0.09427042626104731,
      "flos": 24572092141440.0,
      "grad_norm": 1.9074537137907868,
      "language_loss": 0.78553396,
      "learning_rate": 3.956897773741241e-06,
      "loss": 0.80897152,
      "num_input_tokens_seen": 16738220,
      "step": 784,
      "time_per_iteration": 2.8393797874450684
    },
    {
      "auxiliary_loss_clip": 0.01341039,
      "auxiliary_loss_mlp": 0.01037614,
      "balance_loss_clip": 1.06089544,
      "balance_loss_mlp": 1.02786303,
      "epoch": 0.09439066915168641,
      "flos": 26359581576960.0,
      "grad_norm": 1.8169635397708785,
      "language_loss": 0.71213341,
      "learning_rate": 3.956736776109398e-06,
      "loss": 0.73591995,
      "num_input_tokens_seen": 16759395,
      "step": 785,
      "time_per_iteration": 2.759185314178467
    },
    {
      "auxiliary_loss_clip": 0.01289307,
      "auxiliary_loss_mlp": 0.0258298,
      "balance_loss_clip": 1.0600791,
      "balance_loss_mlp": 1.0001893,
      "epoch": 0.09451091204232549,
      "flos": 19427278296960.0,
      "grad_norm": 2.7290088573052933,
      "language_loss": 0.83541107,
      "learning_rate": 3.956575481642143e-06,
      "loss": 0.87413394,
      "num_input_tokens_seen": 16778285,
      "step": 786,
      "time_per_iteration": 2.692680597305298
    },
    {
      "auxiliary_loss_clip": 0.01442805,
      "auxiliary_loss_mlp": 0.01037114,
      "balance_loss_clip": 1.05516315,
      "balance_loss_mlp": 1.0278337,
      "epoch": 0.09463115493296459,
      "flos": 25368051571200.0,
      "grad_norm": 2.3465075328999365,
      "language_loss": 0.74849319,
      "learning_rate": 3.956413890363943e-06,
      "loss": 0.77329236,
      "num_input_tokens_seen": 16795265,
      "step": 787,
      "time_per_iteration": 2.8058485984802246
    },
    {
      "auxiliary_loss_clip": 0.01302729,
      "auxiliary_loss_mlp": 0.01034894,
      "balance_loss_clip": 1.06897068,
      "balance_loss_mlp": 1.02450466,
      "epoch": 0.09475139782360369,
      "flos": 10123254869760.0,
      "grad_norm": 2.559183535337093,
      "language_loss": 0.81780982,
      "learning_rate": 3.956252002299312e-06,
      "loss": 0.84118605,
      "num_input_tokens_seen": 16811165,
      "step": 788,
      "time_per_iteration": 2.635739803314209
    },
    {
      "auxiliary_loss_clip": 0.01253591,
      "auxiliary_loss_mlp": 0.01039898,
      "balance_loss_clip": 1.07018113,
      "balance_loss_mlp": 1.03068948,
      "epoch": 0.09487164071424277,
      "flos": 17231088936960.0,
      "grad_norm": 1.8891441524584396,
      "language_loss": 0.91263819,
      "learning_rate": 3.956089817472807e-06,
      "loss": 0.93557298,
      "num_input_tokens_seen": 16828470,
      "step": 789,
      "time_per_iteration": 2.6143836975097656
    },
    {
      "auxiliary_loss_clip": 0.01347876,
      "auxiliary_loss_mlp": 0.01039898,
      "balance_loss_clip": 1.06557083,
      "balance_loss_mlp": 1.02942574,
      "epoch": 0.09499188360488187,
      "flos": 30849564528000.0,
      "grad_norm": 3.121485379887564,
      "language_loss": 0.85626978,
      "learning_rate": 3.955927335909032e-06,
      "loss": 0.88014746,
      "num_input_tokens_seen": 16851680,
      "step": 790,
      "time_per_iteration": 2.7574801445007324
    },
    {
      "auxiliary_loss_clip": 0.01445303,
      "auxiliary_loss_mlp": 0.01044821,
      "balance_loss_clip": 1.0629952,
      "balance_loss_mlp": 1.03477192,
      "epoch": 0.09511212649552095,
      "flos": 29351694453120.0,
      "grad_norm": 2.142713441826949,
      "language_loss": 0.75920862,
      "learning_rate": 3.955764557632634e-06,
      "loss": 0.78410983,
      "num_input_tokens_seen": 16871490,
      "step": 791,
      "time_per_iteration": 2.8307130336761475
    },
    {
      "auxiliary_loss_clip": 0.01343906,
      "auxiliary_loss_mlp": 0.01034695,
      "balance_loss_clip": 1.06266606,
      "balance_loss_mlp": 1.02496219,
      "epoch": 0.09523236938616005,
      "flos": 10378687461120.0,
      "grad_norm": 2.821064854403045,
      "language_loss": 0.94545352,
      "learning_rate": 3.955601482668309e-06,
      "loss": 0.96923959,
      "num_input_tokens_seen": 16889350,
      "step": 792,
      "time_per_iteration": 2.652146816253662
    },
    {
      "auxiliary_loss_clip": 0.01447796,
      "auxiliary_loss_mlp": 0.01032097,
      "balance_loss_clip": 1.05687332,
      "balance_loss_mlp": 1.0218749,
      "epoch": 0.09535261227679913,
      "flos": 19061815368960.0,
      "grad_norm": 1.9037487233876924,
      "language_loss": 0.88375258,
      "learning_rate": 3.955438111040794e-06,
      "loss": 0.90855151,
      "num_input_tokens_seen": 16907625,
      "step": 793,
      "time_per_iteration": 2.8065950870513916
    },
    {
      "auxiliary_loss_clip": 0.01440078,
      "auxiliary_loss_mlp": 0.01031836,
      "balance_loss_clip": 1.05818176,
      "balance_loss_mlp": 1.02223969,
      "epoch": 0.09547285516743823,
      "flos": 20922993555840.0,
      "grad_norm": 2.1363304538330277,
      "language_loss": 0.80234361,
      "learning_rate": 3.955274442774873e-06,
      "loss": 0.82706273,
      "num_input_tokens_seen": 16926205,
      "step": 794,
      "time_per_iteration": 2.7760157585144043
    },
    {
      "auxiliary_loss_clip": 0.01297406,
      "auxiliary_loss_mlp": 0.01047135,
      "balance_loss_clip": 1.06263733,
      "balance_loss_mlp": 1.03722906,
      "epoch": 0.09559309805807732,
      "flos": 30154405639680.0,
      "grad_norm": 2.3626513442612205,
      "language_loss": 0.70433891,
      "learning_rate": 3.9551104778953725e-06,
      "loss": 0.72778428,
      "num_input_tokens_seen": 16946500,
      "step": 795,
      "time_per_iteration": 2.734409809112549
    },
    {
      "auxiliary_loss_clip": 0.01395315,
      "auxiliary_loss_mlp": 0.01034249,
      "balance_loss_clip": 1.05897343,
      "balance_loss_mlp": 1.02460504,
      "epoch": 0.0957133409487164,
      "flos": 21066743784960.0,
      "grad_norm": 1.855307641719697,
      "language_loss": 0.85444975,
      "learning_rate": 3.954946216427167e-06,
      "loss": 0.87874532,
      "num_input_tokens_seen": 16966960,
      "step": 796,
      "time_per_iteration": 3.796165943145752
    },
    {
      "auxiliary_loss_clip": 0.01307157,
      "auxiliary_loss_mlp": 0.01007643,
      "balance_loss_clip": 1.03402925,
      "balance_loss_mlp": 1.00456762,
      "epoch": 0.0958335838393555,
      "flos": 71297979315840.0,
      "grad_norm": 0.8781008428262876,
      "language_loss": 0.61572999,
      "learning_rate": 3.954781658395176e-06,
      "loss": 0.63887799,
      "num_input_tokens_seen": 17023215,
      "step": 797,
      "time_per_iteration": 3.224285125732422
    },
    {
      "auxiliary_loss_clip": 0.01353458,
      "auxiliary_loss_mlp": 0.0103977,
      "balance_loss_clip": 1.06078446,
      "balance_loss_mlp": 1.02917254,
      "epoch": 0.09595382672999458,
      "flos": 21872974504320.0,
      "grad_norm": 2.2251914122058305,
      "language_loss": 0.92381334,
      "learning_rate": 3.95461680382436e-06,
      "loss": 0.94774556,
      "num_input_tokens_seen": 17042140,
      "step": 798,
      "time_per_iteration": 2.744856834411621
    },
    {
      "auxiliary_loss_clip": 0.01302539,
      "auxiliary_loss_mlp": 0.01044018,
      "balance_loss_clip": 1.07004166,
      "balance_loss_mlp": 1.03370094,
      "epoch": 0.09607406962063368,
      "flos": 18695562341760.0,
      "grad_norm": 3.4683293536497057,
      "language_loss": 0.86272025,
      "learning_rate": 3.9544516527397295e-06,
      "loss": 0.88618582,
      "num_input_tokens_seen": 17058490,
      "step": 799,
      "time_per_iteration": 3.5638248920440674
    },
    {
      "auxiliary_loss_clip": 0.01336385,
      "auxiliary_loss_mlp": 0.01035011,
      "balance_loss_clip": 1.06079924,
      "balance_loss_mlp": 1.02567673,
      "epoch": 0.09619431251127276,
      "flos": 22568456615040.0,
      "grad_norm": 1.9224407063806832,
      "language_loss": 0.80537695,
      "learning_rate": 3.954286205166338e-06,
      "loss": 0.82909095,
      "num_input_tokens_seen": 17079655,
      "step": 800,
      "time_per_iteration": 3.633012533187866
    },
    {
      "auxiliary_loss_clip": 0.01305729,
      "auxiliary_loss_mlp": 0.0103605,
      "balance_loss_clip": 1.07083166,
      "balance_loss_mlp": 1.02580392,
      "epoch": 0.09631455540191186,
      "flos": 14246230608000.0,
      "grad_norm": 5.020085940932602,
      "language_loss": 0.84101719,
      "learning_rate": 3.954120461129282e-06,
      "loss": 0.86443496,
      "num_input_tokens_seen": 17097065,
      "step": 801,
      "time_per_iteration": 2.6699647903442383
    },
    {
      "auxiliary_loss_clip": 0.01253412,
      "auxiliary_loss_mlp": 0.01033982,
      "balance_loss_clip": 1.07040751,
      "balance_loss_mlp": 1.02451706,
      "epoch": 0.09643479829255096,
      "flos": 20740387789440.0,
      "grad_norm": 2.112756336900727,
      "language_loss": 0.83810461,
      "learning_rate": 3.953954420653706e-06,
      "loss": 0.8609786,
      "num_input_tokens_seen": 17114090,
      "step": 802,
      "time_per_iteration": 2.6397109031677246
    },
    {
      "auxiliary_loss_clip": 0.01302618,
      "auxiliary_loss_mlp": 0.01037963,
      "balance_loss_clip": 1.06729817,
      "balance_loss_mlp": 1.02750862,
      "epoch": 0.09655504118319004,
      "flos": 24420476833920.0,
      "grad_norm": 2.2183713137845733,
      "language_loss": 0.88147485,
      "learning_rate": 3.953788083764798e-06,
      "loss": 0.90488064,
      "num_input_tokens_seen": 17133325,
      "step": 803,
      "time_per_iteration": 3.5655534267425537
    },
    {
      "auxiliary_loss_clip": 0.01446458,
      "auxiliary_loss_mlp": 0.01033517,
      "balance_loss_clip": 1.06038308,
      "balance_loss_mlp": 1.02262712,
      "epoch": 0.09667528407382914,
      "flos": 18441961344000.0,
      "grad_norm": 2.5626233109635557,
      "language_loss": 0.92146331,
      "learning_rate": 3.953621450487792e-06,
      "loss": 0.94626302,
      "num_input_tokens_seen": 17151945,
      "step": 804,
      "time_per_iteration": 2.769533634185791
    },
    {
      "auxiliary_loss_clip": 0.01158109,
      "auxiliary_loss_mlp": 0.0100909,
      "balance_loss_clip": 1.04054379,
      "balance_loss_mlp": 1.0054419,
      "epoch": 0.09679552696446822,
      "flos": 70816455544320.0,
      "grad_norm": 0.8661937028060824,
      "language_loss": 0.61236465,
      "learning_rate": 3.953454520847964e-06,
      "loss": 0.6340366,
      "num_input_tokens_seen": 17216790,
      "step": 805,
      "time_per_iteration": 3.3543875217437744
    },
    {
      "auxiliary_loss_clip": 0.01348135,
      "auxiliary_loss_mlp": 0.01038352,
      "balance_loss_clip": 1.06193972,
      "balance_loss_mlp": 1.02745605,
      "epoch": 0.09691576985510732,
      "flos": 21945514020480.0,
      "grad_norm": 2.0630903265838394,
      "language_loss": 0.73673344,
      "learning_rate": 3.9532872948706395e-06,
      "loss": 0.76059824,
      "num_input_tokens_seen": 17236285,
      "step": 806,
      "time_per_iteration": 2.71807599067688
    },
    {
      "auxiliary_loss_clip": 0.01352046,
      "auxiliary_loss_mlp": 0.01042137,
      "balance_loss_clip": 1.06504726,
      "balance_loss_mlp": 1.0320282,
      "epoch": 0.09703601274574641,
      "flos": 17965211103360.0,
      "grad_norm": 3.114135835783417,
      "language_loss": 0.82817769,
      "learning_rate": 3.9531197725811845e-06,
      "loss": 0.85211957,
      "num_input_tokens_seen": 17251670,
      "step": 807,
      "time_per_iteration": 2.6773998737335205
    },
    {
      "auxiliary_loss_clip": 0.012524,
      "auxiliary_loss_mlp": 0.01033983,
      "balance_loss_clip": 1.07073212,
      "balance_loss_mlp": 1.02441096,
      "epoch": 0.0971562556363855,
      "flos": 22162162901760.0,
      "grad_norm": 2.005967189399882,
      "language_loss": 0.87951058,
      "learning_rate": 3.952951954005013e-06,
      "loss": 0.90237439,
      "num_input_tokens_seen": 17271355,
      "step": 808,
      "time_per_iteration": 2.6130824089050293
    },
    {
      "auxiliary_loss_clip": 0.01342179,
      "auxiliary_loss_mlp": 0.01038281,
      "balance_loss_clip": 1.05819821,
      "balance_loss_mlp": 1.02893496,
      "epoch": 0.0972764985270246,
      "flos": 25848716394240.0,
      "grad_norm": 1.7409043381265648,
      "language_loss": 0.84615773,
      "learning_rate": 3.952783839167584e-06,
      "loss": 0.86996233,
      "num_input_tokens_seen": 17291400,
      "step": 809,
      "time_per_iteration": 2.735697031021118
    },
    {
      "auxiliary_loss_clip": 0.01298295,
      "auxiliary_loss_mlp": 0.01041456,
      "balance_loss_clip": 1.06500268,
      "balance_loss_mlp": 1.03069711,
      "epoch": 0.09739674141766368,
      "flos": 20339373375360.0,
      "grad_norm": 2.720032169766388,
      "language_loss": 0.74442279,
      "learning_rate": 3.952615428094398e-06,
      "loss": 0.76782024,
      "num_input_tokens_seen": 17310920,
      "step": 810,
      "time_per_iteration": 2.6495962142944336
    },
    {
      "auxiliary_loss_clip": 0.01437086,
      "auxiliary_loss_mlp": 0.0104259,
      "balance_loss_clip": 1.05481482,
      "balance_loss_mlp": 1.0331372,
      "epoch": 0.09751698430830277,
      "flos": 15743059188480.0,
      "grad_norm": 1.8250951979139092,
      "language_loss": 0.7343688,
      "learning_rate": 3.952446720811004e-06,
      "loss": 0.75916553,
      "num_input_tokens_seen": 17329245,
      "step": 811,
      "time_per_iteration": 2.8345518112182617
    },
    {
      "auxiliary_loss_clip": 0.0130363,
      "auxiliary_loss_mlp": 0.01005574,
      "balance_loss_clip": 1.03610396,
      "balance_loss_mlp": 1.00223577,
      "epoch": 0.09763722719894186,
      "flos": 63716806800000.0,
      "grad_norm": 0.8322198178218507,
      "language_loss": 0.63597763,
      "learning_rate": 3.952277717342995e-06,
      "loss": 0.65906972,
      "num_input_tokens_seen": 17395680,
      "step": 812,
      "time_per_iteration": 3.3960976600646973
    },
    {
      "auxiliary_loss_clip": 0.01354859,
      "auxiliary_loss_mlp": 0.01045796,
      "balance_loss_clip": 1.06398451,
      "balance_loss_mlp": 1.0353893,
      "epoch": 0.09775747008958095,
      "flos": 22090916275200.0,
      "grad_norm": 3.300995286679296,
      "language_loss": 0.8561461,
      "learning_rate": 3.952108417716009e-06,
      "loss": 0.88015264,
      "num_input_tokens_seen": 17415135,
      "step": 813,
      "time_per_iteration": 2.694577693939209
    },
    {
      "auxiliary_loss_clip": 0.01305712,
      "auxiliary_loss_mlp": 0.01043499,
      "balance_loss_clip": 1.06960356,
      "balance_loss_mlp": 1.03262138,
      "epoch": 0.09787771298022005,
      "flos": 21286050272640.0,
      "grad_norm": 2.67947657467663,
      "language_loss": 0.85130829,
      "learning_rate": 3.951938821955727e-06,
      "loss": 0.87480044,
      "num_input_tokens_seen": 17434535,
      "step": 814,
      "time_per_iteration": 2.697467088699341
    },
    {
      "auxiliary_loss_clip": 0.01350048,
      "auxiliary_loss_mlp": 0.01041953,
      "balance_loss_clip": 1.06405711,
      "balance_loss_mlp": 1.03183746,
      "epoch": 0.09799795587085913,
      "flos": 22054574689920.0,
      "grad_norm": 1.6939892030102104,
      "language_loss": 0.76625752,
      "learning_rate": 3.9517689300878786e-06,
      "loss": 0.79017746,
      "num_input_tokens_seen": 17454270,
      "step": 815,
      "time_per_iteration": 2.7347590923309326
    },
    {
      "auxiliary_loss_clip": 0.01249764,
      "auxiliary_loss_mlp": 0.01035997,
      "balance_loss_clip": 1.06611609,
      "balance_loss_mlp": 1.02624011,
      "epoch": 0.09811819876149823,
      "flos": 22163743100160.0,
      "grad_norm": 1.756564118561201,
      "language_loss": 0.78680634,
      "learning_rate": 3.951598742138236e-06,
      "loss": 0.80966395,
      "num_input_tokens_seen": 17472995,
      "step": 816,
      "time_per_iteration": 2.607825517654419
    },
    {
      "auxiliary_loss_clip": 0.0135286,
      "auxiliary_loss_mlp": 0.01039513,
      "balance_loss_clip": 1.06011462,
      "balance_loss_mlp": 1.02833164,
      "epoch": 0.09823844165213731,
      "flos": 22231111057920.0,
      "grad_norm": 2.3816179409227987,
      "language_loss": 0.79656738,
      "learning_rate": 3.951428258132615e-06,
      "loss": 0.82049108,
      "num_input_tokens_seen": 17491115,
      "step": 817,
      "time_per_iteration": 2.6606271266937256
    },
    {
      "auxiliary_loss_clip": 0.01352758,
      "auxiliary_loss_mlp": 0.01042977,
      "balance_loss_clip": 1.06827521,
      "balance_loss_mlp": 1.0333153,
      "epoch": 0.09835868454277641,
      "flos": 22487728798080.0,
      "grad_norm": 1.878565184958034,
      "language_loss": 0.84609395,
      "learning_rate": 3.951257478096879e-06,
      "loss": 0.87005132,
      "num_input_tokens_seen": 17509480,
      "step": 818,
      "time_per_iteration": 2.831615686416626
    },
    {
      "auxiliary_loss_clip": 0.01353153,
      "auxiliary_loss_mlp": 0.02586715,
      "balance_loss_clip": 1.06674218,
      "balance_loss_mlp": 1.00010407,
      "epoch": 0.0984789274334155,
      "flos": 16362554077440.0,
      "grad_norm": 4.5417733200795976,
      "language_loss": 0.68203533,
      "learning_rate": 3.951086402056936e-06,
      "loss": 0.721434,
      "num_input_tokens_seen": 17524080,
      "step": 819,
      "time_per_iteration": 2.766151189804077
    },
    {
      "auxiliary_loss_clip": 0.01545048,
      "auxiliary_loss_mlp": 0.02582232,
      "balance_loss_clip": 1.06100988,
      "balance_loss_mlp": 1.00014377,
      "epoch": 0.09859917032405459,
      "flos": 24243545416320.0,
      "grad_norm": 1.705610350307804,
      "language_loss": 0.83839619,
      "learning_rate": 3.950915030038735e-06,
      "loss": 0.87966901,
      "num_input_tokens_seen": 17543875,
      "step": 820,
      "time_per_iteration": 3.245574474334717
    },
    {
      "auxiliary_loss_clip": 0.01301975,
      "auxiliary_loss_mlp": 0.010363,
      "balance_loss_clip": 1.06888199,
      "balance_loss_mlp": 1.0273416,
      "epoch": 0.09871941321469369,
      "flos": 17420195064960.0,
      "grad_norm": 2.0042840614351065,
      "language_loss": 0.83153558,
      "learning_rate": 3.9507433620682765e-06,
      "loss": 0.85491836,
      "num_input_tokens_seen": 17560810,
      "step": 821,
      "time_per_iteration": 2.759399890899658
    },
    {
      "auxiliary_loss_clip": 0.01397249,
      "auxiliary_loss_mlp": 0.01042341,
      "balance_loss_clip": 1.06214011,
      "balance_loss_mlp": 1.03096819,
      "epoch": 0.09883965610533277,
      "flos": 28477341590400.0,
      "grad_norm": 2.3302838217099957,
      "language_loss": 0.88266319,
      "learning_rate": 3.9505713981716e-06,
      "loss": 0.90705907,
      "num_input_tokens_seen": 17583640,
      "step": 822,
      "time_per_iteration": 3.7172889709472656
    },
    {
      "auxiliary_loss_clip": 0.01341136,
      "auxiliary_loss_mlp": 0.01036175,
      "balance_loss_clip": 1.06578994,
      "balance_loss_mlp": 1.02569079,
      "epoch": 0.09895989899597187,
      "flos": 23693932437120.0,
      "grad_norm": 1.7423176327569767,
      "language_loss": 0.81422752,
      "learning_rate": 3.950399138374795e-06,
      "loss": 0.83800066,
      "num_input_tokens_seen": 17602720,
      "step": 823,
      "time_per_iteration": 2.733043909072876
    },
    {
      "auxiliary_loss_clip": 0.01298087,
      "auxiliary_loss_mlp": 0.01041376,
      "balance_loss_clip": 1.0634973,
      "balance_loss_mlp": 1.03150535,
      "epoch": 0.09908014188661095,
      "flos": 24679608526080.0,
      "grad_norm": 1.8032670349109168,
      "language_loss": 0.74445665,
      "learning_rate": 3.95022658270399e-06,
      "loss": 0.76785123,
      "num_input_tokens_seen": 17623085,
      "step": 824,
      "time_per_iteration": 2.7748830318450928
    },
    {
      "auxiliary_loss_clip": 0.01348419,
      "auxiliary_loss_mlp": 0.01038867,
      "balance_loss_clip": 1.06459451,
      "balance_loss_mlp": 1.0284121,
      "epoch": 0.09920038477725004,
      "flos": 14064307200000.0,
      "grad_norm": 2.0002958338262187,
      "language_loss": 0.78307974,
      "learning_rate": 3.9500537311853635e-06,
      "loss": 0.8069526,
      "num_input_tokens_seen": 17641040,
      "step": 825,
      "time_per_iteration": 3.572065830230713
    },
    {
      "auxiliary_loss_clip": 0.01300943,
      "auxiliary_loss_mlp": 0.01033861,
      "balance_loss_clip": 1.06352174,
      "balance_loss_mlp": 1.02316749,
      "epoch": 0.09932062766788914,
      "flos": 13407070095360.0,
      "grad_norm": 3.9795129805510787,
      "language_loss": 0.82857579,
      "learning_rate": 3.949880583845136e-06,
      "loss": 0.85192382,
      "num_input_tokens_seen": 17659115,
      "step": 826,
      "time_per_iteration": 2.658402681350708
    },
    {
      "auxiliary_loss_clip": 0.0135094,
      "auxiliary_loss_mlp": 0.01042451,
      "balance_loss_clip": 1.06425178,
      "balance_loss_mlp": 1.0315969,
      "epoch": 0.09944087055852822,
      "flos": 19500751566720.0,
      "grad_norm": 2.300144202827262,
      "language_loss": 0.81137884,
      "learning_rate": 3.949707140709575e-06,
      "loss": 0.83531272,
      "num_input_tokens_seen": 17678845,
      "step": 827,
      "time_per_iteration": 2.728961706161499
    },
    {
      "auxiliary_loss_clip": 0.01301776,
      "auxiliary_loss_mlp": 0.0103221,
      "balance_loss_clip": 1.06339192,
      "balance_loss_mlp": 1.02232754,
      "epoch": 0.09956111344916732,
      "flos": 17749100926080.0,
      "grad_norm": 3.698818506576039,
      "language_loss": 0.83373415,
      "learning_rate": 3.949533401804991e-06,
      "loss": 0.85707396,
      "num_input_tokens_seen": 17695750,
      "step": 828,
      "time_per_iteration": 2.7012412548065186
    },
    {
      "auxiliary_loss_clip": 0.01297643,
      "auxiliary_loss_mlp": 0.02586823,
      "balance_loss_clip": 1.0649246,
      "balance_loss_mlp": 1.00017715,
      "epoch": 0.0996813563398064,
      "flos": 17967581400960.0,
      "grad_norm": 2.104135310068123,
      "language_loss": 0.9090116,
      "learning_rate": 3.949359367157739e-06,
      "loss": 0.94785631,
      "num_input_tokens_seen": 17714445,
      "step": 829,
      "time_per_iteration": 3.4236602783203125
    },
    {
      "auxiliary_loss_clip": 0.01303851,
      "auxiliary_loss_mlp": 0.01035541,
      "balance_loss_clip": 1.06641686,
      "balance_loss_mlp": 1.02463317,
      "epoch": 0.0998015992304455,
      "flos": 17457039440640.0,
      "grad_norm": 2.1256782320122336,
      "language_loss": 0.7586543,
      "learning_rate": 3.949185036794222e-06,
      "loss": 0.78204823,
      "num_input_tokens_seen": 17732455,
      "step": 830,
      "time_per_iteration": 2.7445285320281982
    },
    {
      "auxiliary_loss_clip": 0.01250084,
      "auxiliary_loss_mlp": 0.01038,
      "balance_loss_clip": 1.06826544,
      "balance_loss_mlp": 1.02792645,
      "epoch": 0.0999218421210846,
      "flos": 25888757080320.0,
      "grad_norm": 1.6700352833308703,
      "language_loss": 0.7871033,
      "learning_rate": 3.949010410740884e-06,
      "loss": 0.80998409,
      "num_input_tokens_seen": 17755280,
      "step": 831,
      "time_per_iteration": 2.7063138484954834
    },
    {
      "auxiliary_loss_clip": 0.01340093,
      "auxiliary_loss_mlp": 0.02586968,
      "balance_loss_clip": 1.06156838,
      "balance_loss_mlp": 1.00012934,
      "epoch": 0.10004208501172368,
      "flos": 21215916967680.0,
      "grad_norm": 1.6929425756514753,
      "language_loss": 0.86559629,
      "learning_rate": 3.948835489024216e-06,
      "loss": 0.90486687,
      "num_input_tokens_seen": 17775015,
      "step": 832,
      "time_per_iteration": 2.7119762897491455
    },
    {
      "auxiliary_loss_clip": 0.01302414,
      "auxiliary_loss_mlp": 0.01038415,
      "balance_loss_clip": 1.06465936,
      "balance_loss_mlp": 1.02864015,
      "epoch": 0.10016232790236278,
      "flos": 17348409734400.0,
      "grad_norm": 2.065389451346853,
      "language_loss": 0.9021489,
      "learning_rate": 3.948660271670755e-06,
      "loss": 0.9255572,
      "num_input_tokens_seen": 17792165,
      "step": 833,
      "time_per_iteration": 2.6568450927734375
    },
    {
      "auxiliary_loss_clip": 0.01343759,
      "auxiliary_loss_mlp": 0.01044636,
      "balance_loss_clip": 1.06412292,
      "balance_loss_mlp": 1.03397894,
      "epoch": 0.10028257079300186,
      "flos": 25666541591040.0,
      "grad_norm": 2.513647871963231,
      "language_loss": 0.84353143,
      "learning_rate": 3.948484758707079e-06,
      "loss": 0.86741543,
      "num_input_tokens_seen": 17811765,
      "step": 834,
      "time_per_iteration": 2.7861578464508057
    },
    {
      "auxiliary_loss_clip": 0.01390252,
      "auxiliary_loss_mlp": 0.01042374,
      "balance_loss_clip": 1.0600909,
      "balance_loss_mlp": 1.03211594,
      "epoch": 0.10040281368364096,
      "flos": 25156035544320.0,
      "grad_norm": 1.9055475613971458,
      "language_loss": 0.83512115,
      "learning_rate": 3.948308950159815e-06,
      "loss": 0.85944742,
      "num_input_tokens_seen": 17830445,
      "step": 835,
      "time_per_iteration": 2.7812860012054443
    },
    {
      "auxiliary_loss_clip": 0.01389568,
      "auxiliary_loss_mlp": 0.01049471,
      "balance_loss_clip": 1.05799627,
      "balance_loss_mlp": 1.03776419,
      "epoch": 0.10052305657428004,
      "flos": 17603303621760.0,
      "grad_norm": 2.773156265467036,
      "language_loss": 0.75865924,
      "learning_rate": 3.9481328460556326e-06,
      "loss": 0.78304964,
      "num_input_tokens_seen": 17847665,
      "step": 836,
      "time_per_iteration": 2.721379041671753
    },
    {
      "auxiliary_loss_clip": 0.01342897,
      "auxiliary_loss_mlp": 0.01036679,
      "balance_loss_clip": 1.06079221,
      "balance_loss_mlp": 1.02624822,
      "epoch": 0.10064329946491914,
      "flos": 18660154510080.0,
      "grad_norm": 2.111303597573002,
      "language_loss": 0.89695066,
      "learning_rate": 3.9479564464212455e-06,
      "loss": 0.92074645,
      "num_input_tokens_seen": 17866825,
      "step": 837,
      "time_per_iteration": 2.7622225284576416
    },
    {
      "auxiliary_loss_clip": 0.0125548,
      "auxiliary_loss_mlp": 0.01045309,
      "balance_loss_clip": 1.06680083,
      "balance_loss_mlp": 1.03432977,
      "epoch": 0.10076354235555823,
      "flos": 17199056983680.0,
      "grad_norm": 2.66184649373113,
      "language_loss": 0.76359916,
      "learning_rate": 3.947779751283414e-06,
      "loss": 0.78660697,
      "num_input_tokens_seen": 17883995,
      "step": 838,
      "time_per_iteration": 2.601921796798706
    },
    {
      "auxiliary_loss_clip": 0.01306564,
      "auxiliary_loss_mlp": 0.02587888,
      "balance_loss_clip": 1.0735085,
      "balance_loss_mlp": 1.00014162,
      "epoch": 0.10088378524619732,
      "flos": 22962252395520.0,
      "grad_norm": 1.9090757641667346,
      "language_loss": 0.76190615,
      "learning_rate": 3.947602760668944e-06,
      "loss": 0.80085063,
      "num_input_tokens_seen": 17903785,
      "step": 839,
      "time_per_iteration": 2.6983680725097656
    },
    {
      "auxiliary_loss_clip": 0.01304404,
      "auxiliary_loss_mlp": 0.01043458,
      "balance_loss_clip": 1.070297,
      "balance_loss_mlp": 1.0320015,
      "epoch": 0.10100402813683641,
      "flos": 37885828746240.0,
      "grad_norm": 1.747946456929717,
      "language_loss": 0.71343321,
      "learning_rate": 3.947425474604684e-06,
      "loss": 0.73691177,
      "num_input_tokens_seen": 17927720,
      "step": 840,
      "time_per_iteration": 2.7661752700805664
    },
    {
      "auxiliary_loss_clip": 0.01352032,
      "auxiliary_loss_mlp": 0.01044561,
      "balance_loss_clip": 1.06433249,
      "balance_loss_mlp": 1.03419018,
      "epoch": 0.1011242710274755,
      "flos": 21543458112000.0,
      "grad_norm": 3.0637580333462004,
      "language_loss": 0.92237216,
      "learning_rate": 3.947247893117528e-06,
      "loss": 0.94633806,
      "num_input_tokens_seen": 17946225,
      "step": 841,
      "time_per_iteration": 2.7693183422088623
    },
    {
      "auxiliary_loss_clip": 0.01296589,
      "auxiliary_loss_mlp": 0.01036179,
      "balance_loss_clip": 1.06247568,
      "balance_loss_mlp": 1.02397168,
      "epoch": 0.10124451391811459,
      "flos": 13621456419840.0,
      "grad_norm": 3.00290162648237,
      "language_loss": 0.69262087,
      "learning_rate": 3.947070016234413e-06,
      "loss": 0.71594852,
      "num_input_tokens_seen": 17962015,
      "step": 842,
      "time_per_iteration": 2.5775561332702637
    },
    {
      "auxiliary_loss_clip": 0.01363807,
      "auxiliary_loss_mlp": 0.01041379,
      "balance_loss_clip": 1.06748295,
      "balance_loss_mlp": 1.03045893,
      "epoch": 0.10136475680875369,
      "flos": 16649228522880.0,
      "grad_norm": 2.3370184714401376,
      "language_loss": 0.74689561,
      "learning_rate": 3.946891843982326e-06,
      "loss": 0.77094746,
      "num_input_tokens_seen": 17979680,
      "step": 843,
      "time_per_iteration": 2.7018189430236816
    },
    {
      "auxiliary_loss_clip": 0.01306154,
      "auxiliary_loss_mlp": 0.0104131,
      "balance_loss_clip": 1.07071114,
      "balance_loss_mlp": 1.03074265,
      "epoch": 0.10148499969939277,
      "flos": 19461034103040.0,
      "grad_norm": 5.000096303466017,
      "language_loss": 0.7441259,
      "learning_rate": 3.9467133763882935e-06,
      "loss": 0.76760054,
      "num_input_tokens_seen": 17998145,
      "step": 844,
      "time_per_iteration": 2.6081578731536865
    },
    {
      "auxiliary_loss_clip": 0.01293608,
      "auxiliary_loss_mlp": 0.01041343,
      "balance_loss_clip": 1.06418085,
      "balance_loss_mlp": 1.03026271,
      "epoch": 0.10160524259003187,
      "flos": 21104988791040.0,
      "grad_norm": 2.160297694559054,
      "language_loss": 0.86166215,
      "learning_rate": 3.9465346134793905e-06,
      "loss": 0.88501167,
      "num_input_tokens_seen": 18017955,
      "step": 845,
      "time_per_iteration": 2.6279892921447754
    },
    {
      "auxiliary_loss_clip": 0.01400756,
      "auxiliary_loss_mlp": 0.01037706,
      "balance_loss_clip": 1.06657743,
      "balance_loss_mlp": 1.02648282,
      "epoch": 0.10172548548067095,
      "flos": 17712687513600.0,
      "grad_norm": 1.9810872742552859,
      "language_loss": 0.79641867,
      "learning_rate": 3.9463555552827335e-06,
      "loss": 0.82080328,
      "num_input_tokens_seen": 18035125,
      "step": 846,
      "time_per_iteration": 2.7496979236602783
    },
    {
      "auxiliary_loss_clip": 0.01293261,
      "auxiliary_loss_mlp": 0.01034873,
      "balance_loss_clip": 1.06492686,
      "balance_loss_mlp": 1.02458572,
      "epoch": 0.10184572837131005,
      "flos": 21104845136640.0,
      "grad_norm": 2.466958964085333,
      "language_loss": 0.86365438,
      "learning_rate": 3.946176201825487e-06,
      "loss": 0.88693571,
      "num_input_tokens_seen": 18053160,
      "step": 847,
      "time_per_iteration": 2.685763120651245
    },
    {
      "auxiliary_loss_clip": 0.01352523,
      "auxiliary_loss_mlp": 0.01038489,
      "balance_loss_clip": 1.06782162,
      "balance_loss_mlp": 1.0277009,
      "epoch": 0.10196597126194913,
      "flos": 26067591918720.0,
      "grad_norm": 2.093672568118895,
      "language_loss": 0.83569312,
      "learning_rate": 3.9459965531348575e-06,
      "loss": 0.85960323,
      "num_input_tokens_seen": 18072815,
      "step": 848,
      "time_per_iteration": 3.571596145629883
    },
    {
      "auxiliary_loss_clip": 0.0135507,
      "auxiliary_loss_mlp": 0.02589704,
      "balance_loss_clip": 1.06685448,
      "balance_loss_mlp": 1.00022531,
      "epoch": 0.10208621415258823,
      "flos": 29314634595840.0,
      "grad_norm": 2.3532693893168766,
      "language_loss": 0.85170591,
      "learning_rate": 3.945816609238098e-06,
      "loss": 0.89115369,
      "num_input_tokens_seen": 18092225,
      "step": 849,
      "time_per_iteration": 2.709230899810791
    },
    {
      "auxiliary_loss_clip": 0.0144598,
      "auxiliary_loss_mlp": 0.01039489,
      "balance_loss_clip": 1.0630784,
      "balance_loss_mlp": 1.02849197,
      "epoch": 0.10220645704322733,
      "flos": 23805794367360.0,
      "grad_norm": 2.602493067714094,
      "language_loss": 0.85317552,
      "learning_rate": 3.945636370162507e-06,
      "loss": 0.87803018,
      "num_input_tokens_seen": 18112335,
      "step": 850,
      "time_per_iteration": 2.79087233543396
    },
    {
      "auxiliary_loss_clip": 0.01299931,
      "auxiliary_loss_mlp": 0.01033223,
      "balance_loss_clip": 1.06664443,
      "balance_loss_mlp": 1.02294147,
      "epoch": 0.10232669993386641,
      "flos": 23218546913280.0,
      "grad_norm": 3.0138658758730537,
      "language_loss": 0.79482681,
      "learning_rate": 3.945455835935425e-06,
      "loss": 0.81815839,
      "num_input_tokens_seen": 18131520,
      "step": 851,
      "time_per_iteration": 3.6043152809143066
    },
    {
      "auxiliary_loss_clip": 0.01343116,
      "auxiliary_loss_mlp": 0.01042786,
      "balance_loss_clip": 1.064237,
      "balance_loss_mlp": 1.03140163,
      "epoch": 0.1024469428245055,
      "flos": 22922929981440.0,
      "grad_norm": 2.9926095511414985,
      "language_loss": 0.75070095,
      "learning_rate": 3.94527500658424e-06,
      "loss": 0.77455997,
      "num_input_tokens_seen": 18149185,
      "step": 852,
      "time_per_iteration": 2.7230799198150635
    },
    {
      "auxiliary_loss_clip": 0.01438397,
      "auxiliary_loss_mlp": 0.01035912,
      "balance_loss_clip": 1.06229687,
      "balance_loss_mlp": 1.02632117,
      "epoch": 0.10256718571514459,
      "flos": 31359495957120.0,
      "grad_norm": 1.9174218095066955,
      "language_loss": 0.81454813,
      "learning_rate": 3.945093882136382e-06,
      "loss": 0.83929127,
      "num_input_tokens_seen": 18172960,
      "step": 853,
      "time_per_iteration": 2.890918016433716
    },
    {
      "auxiliary_loss_clip": 0.01353718,
      "auxiliary_loss_mlp": 0.02583719,
      "balance_loss_clip": 1.07037961,
      "balance_loss_mlp": 1.00022626,
      "epoch": 0.10268742860578368,
      "flos": 23474877344640.0,
      "grad_norm": 2.174854289909286,
      "language_loss": 0.84821534,
      "learning_rate": 3.944912462619329e-06,
      "loss": 0.88758969,
      "num_input_tokens_seen": 18191925,
      "step": 854,
      "time_per_iteration": 2.7103817462921143
    },
    {
      "auxiliary_loss_clip": 0.01354252,
      "auxiliary_loss_mlp": 0.01037453,
      "balance_loss_clip": 1.06519496,
      "balance_loss_mlp": 1.02601469,
      "epoch": 0.10280767149642277,
      "flos": 25520313323520.0,
      "grad_norm": 2.915545647260689,
      "language_loss": 0.80504239,
      "learning_rate": 3.9447307480606025e-06,
      "loss": 0.82895947,
      "num_input_tokens_seen": 18212010,
      "step": 855,
      "time_per_iteration": 3.6548516750335693
    },
    {
      "auxiliary_loss_clip": 0.01347285,
      "auxiliary_loss_mlp": 0.0103845,
      "balance_loss_clip": 1.06437457,
      "balance_loss_mlp": 1.02812028,
      "epoch": 0.10292791438706186,
      "flos": 17347691462400.0,
      "grad_norm": 4.090083955652681,
      "language_loss": 0.90432471,
      "learning_rate": 3.944548738487767e-06,
      "loss": 0.92818213,
      "num_input_tokens_seen": 18229525,
      "step": 856,
      "time_per_iteration": 2.6541080474853516
    },
    {
      "auxiliary_loss_clip": 0.01259482,
      "auxiliary_loss_mlp": 0.01044471,
      "balance_loss_clip": 1.07301736,
      "balance_loss_mlp": 1.03346813,
      "epoch": 0.10304815727770096,
      "flos": 27052693390080.0,
      "grad_norm": 2.9288207290056323,
      "language_loss": 0.90937489,
      "learning_rate": 3.944366433928434e-06,
      "loss": 0.93241447,
      "num_input_tokens_seen": 18249505,
      "step": 857,
      "time_per_iteration": 2.6882293224334717
    },
    {
      "auxiliary_loss_clip": 0.01348875,
      "auxiliary_loss_mlp": 0.01037213,
      "balance_loss_clip": 1.06459022,
      "balance_loss_mlp": 1.02760446,
      "epoch": 0.10316840016834004,
      "flos": 22782591544320.0,
      "grad_norm": 1.476033550260478,
      "language_loss": 0.83588958,
      "learning_rate": 3.9441838344102594e-06,
      "loss": 0.85975045,
      "num_input_tokens_seen": 18269230,
      "step": 858,
      "time_per_iteration": 2.7407398223876953
    },
    {
      "auxiliary_loss_clip": 0.01348059,
      "auxiliary_loss_mlp": 0.01049101,
      "balance_loss_clip": 1.0692687,
      "balance_loss_mlp": 1.03813958,
      "epoch": 0.10328864305897914,
      "flos": 20704584908160.0,
      "grad_norm": 2.570264125567191,
      "language_loss": 0.67277664,
      "learning_rate": 3.944000939960943e-06,
      "loss": 0.69674826,
      "num_input_tokens_seen": 18287955,
      "step": 859,
      "time_per_iteration": 2.7064414024353027
    },
    {
      "auxiliary_loss_clip": 0.01301122,
      "auxiliary_loss_mlp": 0.01048256,
      "balance_loss_clip": 1.06491089,
      "balance_loss_mlp": 1.03786051,
      "epoch": 0.10340888594961822,
      "flos": 28478814048000.0,
      "grad_norm": 2.1099928274487354,
      "language_loss": 0.80099273,
      "learning_rate": 3.943817750608229e-06,
      "loss": 0.82448649,
      "num_input_tokens_seen": 18310505,
      "step": 860,
      "time_per_iteration": 2.7620561122894287
    },
    {
      "auxiliary_loss_clip": 0.01305565,
      "auxiliary_loss_mlp": 0.01045252,
      "balance_loss_clip": 1.07082915,
      "balance_loss_mlp": 1.03547716,
      "epoch": 0.10352912884025732,
      "flos": 13370333460480.0,
      "grad_norm": 3.3914448296861703,
      "language_loss": 0.81796426,
      "learning_rate": 3.943634266379908e-06,
      "loss": 0.84147251,
      "num_input_tokens_seen": 18327400,
      "step": 861,
      "time_per_iteration": 2.6324148178100586
    },
    {
      "auxiliary_loss_clip": 0.01306078,
      "auxiliary_loss_mlp": 0.01034072,
      "balance_loss_clip": 1.06863582,
      "balance_loss_mlp": 1.02303934,
      "epoch": 0.10364937173089642,
      "flos": 25558558329600.0,
      "grad_norm": 1.8228171277431984,
      "language_loss": 0.84820378,
      "learning_rate": 3.943450487303815e-06,
      "loss": 0.87160528,
      "num_input_tokens_seen": 18347895,
      "step": 862,
      "time_per_iteration": 2.725874900817871
    },
    {
      "auxiliary_loss_clip": 0.01303008,
      "auxiliary_loss_mlp": 0.01040817,
      "balance_loss_clip": 1.06975269,
      "balance_loss_mlp": 1.03109539,
      "epoch": 0.1037696146215355,
      "flos": 21215486004480.0,
      "grad_norm": 1.9351552918347021,
      "language_loss": 0.85734904,
      "learning_rate": 3.943266413407827e-06,
      "loss": 0.88078725,
      "num_input_tokens_seen": 18367170,
      "step": 863,
      "time_per_iteration": 2.6590418815612793
    },
    {
      "auxiliary_loss_clip": 0.01306361,
      "auxiliary_loss_mlp": 0.01041022,
      "balance_loss_clip": 1.06937075,
      "balance_loss_mlp": 1.03106189,
      "epoch": 0.1038898575121746,
      "flos": 25807382818560.0,
      "grad_norm": 2.153519410291218,
      "language_loss": 0.85054392,
      "learning_rate": 3.94308204471987e-06,
      "loss": 0.87401772,
      "num_input_tokens_seen": 18386185,
      "step": 864,
      "time_per_iteration": 2.7443456649780273
    },
    {
      "auxiliary_loss_clip": 0.01382104,
      "auxiliary_loss_mlp": 0.01035388,
      "balance_loss_clip": 1.06143212,
      "balance_loss_mlp": 1.02535045,
      "epoch": 0.10401010040281368,
      "flos": 19062425900160.0,
      "grad_norm": 11.008755945457288,
      "language_loss": 0.74760067,
      "learning_rate": 3.942897381267912e-06,
      "loss": 0.7717756,
      "num_input_tokens_seen": 18402550,
      "step": 865,
      "time_per_iteration": 2.7408041954040527
    },
    {
      "auxiliary_loss_clip": 0.01303018,
      "auxiliary_loss_mlp": 0.01037687,
      "balance_loss_clip": 1.06917453,
      "balance_loss_mlp": 1.02650535,
      "epoch": 0.10413034329345278,
      "flos": 16355119962240.0,
      "grad_norm": 3.221429661219543,
      "language_loss": 0.6607008,
      "learning_rate": 3.942712423079965e-06,
      "loss": 0.68410784,
      "num_input_tokens_seen": 18418940,
      "step": 866,
      "time_per_iteration": 2.6933605670928955
    },
    {
      "auxiliary_loss_clip": 0.01387212,
      "auxiliary_loss_mlp": 0.01031468,
      "balance_loss_clip": 1.05687702,
      "balance_loss_mlp": 1.02167463,
      "epoch": 0.10425058618409186,
      "flos": 17236511890560.0,
      "grad_norm": 2.380158919463994,
      "language_loss": 0.90033174,
      "learning_rate": 3.942527170184088e-06,
      "loss": 0.92451859,
      "num_input_tokens_seen": 18435560,
      "step": 867,
      "time_per_iteration": 2.7243762016296387
    },
    {
      "auxiliary_loss_clip": 0.01254056,
      "auxiliary_loss_mlp": 0.01039697,
      "balance_loss_clip": 1.06963706,
      "balance_loss_mlp": 1.02982068,
      "epoch": 0.10437082907473096,
      "flos": 17967365919360.0,
      "grad_norm": 2.895594894395919,
      "language_loss": 0.77428818,
      "learning_rate": 3.942341622608385e-06,
      "loss": 0.79722571,
      "num_input_tokens_seen": 18452590,
      "step": 868,
      "time_per_iteration": 2.6456151008605957
    },
    {
      "auxiliary_loss_clip": 0.01353222,
      "auxiliary_loss_mlp": 0.01041592,
      "balance_loss_clip": 1.07053971,
      "balance_loss_mlp": 1.03128088,
      "epoch": 0.10449107196537005,
      "flos": 36283315374720.0,
      "grad_norm": 1.5070420982343802,
      "language_loss": 0.77549112,
      "learning_rate": 3.942155780381001e-06,
      "loss": 0.79943919,
      "num_input_tokens_seen": 18476325,
      "step": 869,
      "time_per_iteration": 2.876343011856079
    },
    {
      "auxiliary_loss_clip": 0.01356272,
      "auxiliary_loss_mlp": 0.0104173,
      "balance_loss_clip": 1.0665282,
      "balance_loss_mlp": 1.0313468,
      "epoch": 0.10461131485600914,
      "flos": 23802095266560.0,
      "grad_norm": 2.4674539349140363,
      "language_loss": 0.75990868,
      "learning_rate": 3.94196964353013e-06,
      "loss": 0.7838887,
      "num_input_tokens_seen": 18495775,
      "step": 870,
      "time_per_iteration": 2.6793954372406006
    },
    {
      "auxiliary_loss_clip": 0.01337256,
      "auxiliary_loss_mlp": 0.02581252,
      "balance_loss_clip": 1.0646832,
      "balance_loss_mlp": 1.00022149,
      "epoch": 0.10473155774664823,
      "flos": 18405476104320.0,
      "grad_norm": 1.9733478114754022,
      "language_loss": 0.80862761,
      "learning_rate": 3.941783212084008e-06,
      "loss": 0.84781265,
      "num_input_tokens_seen": 18513530,
      "step": 871,
      "time_per_iteration": 2.706925868988037
    },
    {
      "auxiliary_loss_clip": 0.01338429,
      "auxiliary_loss_mlp": 0.01042873,
      "balance_loss_clip": 1.06592727,
      "balance_loss_mlp": 1.03160191,
      "epoch": 0.10485180063728732,
      "flos": 25592637358080.0,
      "grad_norm": 2.5133753991572845,
      "language_loss": 0.7929424,
      "learning_rate": 3.941596486070916e-06,
      "loss": 0.81675535,
      "num_input_tokens_seen": 18531575,
      "step": 872,
      "time_per_iteration": 2.7938344478607178
    },
    {
      "auxiliary_loss_clip": 0.01441097,
      "auxiliary_loss_mlp": 0.01035388,
      "balance_loss_clip": 1.06493735,
      "balance_loss_mlp": 1.02515984,
      "epoch": 0.10497204352792641,
      "flos": 27088747666560.0,
      "grad_norm": 4.045320762522045,
      "language_loss": 0.58563644,
      "learning_rate": 3.941409465519182e-06,
      "loss": 0.61040133,
      "num_input_tokens_seen": 18552100,
      "step": 873,
      "time_per_iteration": 2.7772326469421387
    },
    {
      "auxiliary_loss_clip": 0.01298892,
      "auxiliary_loss_mlp": 0.01035669,
      "balance_loss_clip": 1.06715918,
      "balance_loss_mlp": 1.02555442,
      "epoch": 0.10509228641856551,
      "flos": 32858479353600.0,
      "grad_norm": 1.8877114873019658,
      "language_loss": 0.85442543,
      "learning_rate": 3.941222150457176e-06,
      "loss": 0.87777102,
      "num_input_tokens_seen": 18575355,
      "step": 874,
      "time_per_iteration": 3.7072579860687256
    },
    {
      "auxiliary_loss_clip": 0.01306494,
      "auxiliary_loss_mlp": 0.01043968,
      "balance_loss_clip": 1.06779838,
      "balance_loss_mlp": 1.03397274,
      "epoch": 0.10521252930920459,
      "flos": 14319165173760.0,
      "grad_norm": 2.5392838585738007,
      "language_loss": 0.71609318,
      "learning_rate": 3.941034540913311e-06,
      "loss": 0.7395978,
      "num_input_tokens_seen": 18592885,
      "step": 875,
      "time_per_iteration": 2.659313440322876
    },
    {
      "auxiliary_loss_clip": 0.01305398,
      "auxiliary_loss_mlp": 0.02584788,
      "balance_loss_clip": 1.07000494,
      "balance_loss_mlp": 1.00018096,
      "epoch": 0.10533277219984369,
      "flos": 21687028773120.0,
      "grad_norm": 1.756753641284867,
      "language_loss": 0.82390672,
      "learning_rate": 3.940846636916051e-06,
      "loss": 0.86280859,
      "num_input_tokens_seen": 18612920,
      "step": 876,
      "time_per_iteration": 2.7088160514831543
    },
    {
      "auxiliary_loss_clip": 0.01351354,
      "auxiliary_loss_mlp": 0.01038978,
      "balance_loss_clip": 1.07007122,
      "balance_loss_mlp": 1.02857065,
      "epoch": 0.10545301509048277,
      "flos": 22269787027200.0,
      "grad_norm": 2.2313118991040506,
      "language_loss": 0.8656652,
      "learning_rate": 3.940658438493899e-06,
      "loss": 0.88956857,
      "num_input_tokens_seen": 18630765,
      "step": 877,
      "time_per_iteration": 4.1792895793914795
    },
    {
      "auxiliary_loss_clip": 0.01256471,
      "auxiliary_loss_mlp": 0.01034991,
      "balance_loss_clip": 1.06894863,
      "balance_loss_mlp": 1.02447128,
      "epoch": 0.10557325798112187,
      "flos": 22199725549440.0,
      "grad_norm": 3.264153738361369,
      "language_loss": 0.76168668,
      "learning_rate": 3.940469945675405e-06,
      "loss": 0.78460133,
      "num_input_tokens_seen": 18649150,
      "step": 878,
      "time_per_iteration": 3.4910080432891846
    },
    {
      "auxiliary_loss_clip": 0.01430099,
      "auxiliary_loss_mlp": 0.01039579,
      "balance_loss_clip": 1.05929315,
      "balance_loss_mlp": 1.02998304,
      "epoch": 0.10569350087176095,
      "flos": 25775889569280.0,
      "grad_norm": 1.9744735553123471,
      "language_loss": 0.91745734,
      "learning_rate": 3.940281158489163e-06,
      "loss": 0.94215417,
      "num_input_tokens_seen": 18668380,
      "step": 879,
      "time_per_iteration": 2.8823490142822266
    },
    {
      "auxiliary_loss_clip": 0.01496624,
      "auxiliary_loss_mlp": 0.01034615,
      "balance_loss_clip": 1.05835819,
      "balance_loss_mlp": 1.02481604,
      "epoch": 0.10581374376240005,
      "flos": 17311385790720.0,
      "grad_norm": 1.9712096341276695,
      "language_loss": 0.82552469,
      "learning_rate": 3.940092076963812e-06,
      "loss": 0.85083711,
      "num_input_tokens_seen": 18685875,
      "step": 880,
      "time_per_iteration": 2.883492946624756
    },
    {
      "auxiliary_loss_clip": 0.0135262,
      "auxiliary_loss_mlp": 0.01034728,
      "balance_loss_clip": 1.06509137,
      "balance_loss_mlp": 1.02413607,
      "epoch": 0.10593398665303914,
      "flos": 34349454017280.0,
      "grad_norm": 2.223088259957407,
      "language_loss": 0.7897352,
      "learning_rate": 3.9399027011280355e-06,
      "loss": 0.81360865,
      "num_input_tokens_seen": 18707970,
      "step": 881,
      "time_per_iteration": 3.9809439182281494
    },
    {
      "auxiliary_loss_clip": 0.01350961,
      "auxiliary_loss_mlp": 0.0104135,
      "balance_loss_clip": 1.06838095,
      "balance_loss_mlp": 1.0306747,
      "epoch": 0.10605422954367823,
      "flos": 23257977068160.0,
      "grad_norm": 2.0186631158563375,
      "language_loss": 0.77477062,
      "learning_rate": 3.939713031010561e-06,
      "loss": 0.79869378,
      "num_input_tokens_seen": 18726335,
      "step": 882,
      "time_per_iteration": 2.7577338218688965
    },
    {
      "auxiliary_loss_clip": 0.01401742,
      "auxiliary_loss_mlp": 0.01040346,
      "balance_loss_clip": 1.06661057,
      "balance_loss_mlp": 1.02924144,
      "epoch": 0.10617447243431732,
      "flos": 22820118278400.0,
      "grad_norm": 2.5893077292787297,
      "language_loss": 0.78048599,
      "learning_rate": 3.939523066640163e-06,
      "loss": 0.80490685,
      "num_input_tokens_seen": 18745230,
      "step": 883,
      "time_per_iteration": 2.725170373916626
    },
    {
      "auxiliary_loss_clip": 0.01288289,
      "auxiliary_loss_mlp": 0.01038096,
      "balance_loss_clip": 1.06752968,
      "balance_loss_mlp": 1.02765334,
      "epoch": 0.10629471532495641,
      "flos": 24386577373440.0,
      "grad_norm": 2.06984676202614,
      "language_loss": 0.81154305,
      "learning_rate": 3.939332808045657e-06,
      "loss": 0.83480692,
      "num_input_tokens_seen": 18764880,
      "step": 884,
      "time_per_iteration": 2.726757764816284
    },
    {
      "auxiliary_loss_clip": 0.01399036,
      "auxiliary_loss_mlp": 0.01044102,
      "balance_loss_clip": 1.06401467,
      "balance_loss_mlp": 1.03298545,
      "epoch": 0.1064149582155955,
      "flos": 21105491581440.0,
      "grad_norm": 12.649308647013715,
      "language_loss": 0.84463727,
      "learning_rate": 3.939142255255906e-06,
      "loss": 0.86906862,
      "num_input_tokens_seen": 18785765,
      "step": 885,
      "time_per_iteration": 2.7322347164154053
    },
    {
      "auxiliary_loss_clip": 0.01299615,
      "auxiliary_loss_mlp": 0.0103791,
      "balance_loss_clip": 1.06775844,
      "balance_loss_mlp": 1.02842712,
      "epoch": 0.1065352011062346,
      "flos": 20702035042560.0,
      "grad_norm": 1.813703040013049,
      "language_loss": 0.86977959,
      "learning_rate": 3.938951408299817e-06,
      "loss": 0.89315486,
      "num_input_tokens_seen": 18804605,
      "step": 886,
      "time_per_iteration": 2.7396903038024902
    },
    {
      "auxiliary_loss_clip": 0.0137153,
      "auxiliary_loss_mlp": 0.01042594,
      "balance_loss_clip": 1.05712545,
      "balance_loss_mlp": 1.03968525,
      "epoch": 0.10665544399687368,
      "flos": 62659632689280.0,
      "grad_norm": 0.7983060358596532,
      "language_loss": 0.54427731,
      "learning_rate": 3.938760267206342e-06,
      "loss": 0.5684185,
      "num_input_tokens_seen": 18866425,
      "step": 887,
      "time_per_iteration": 3.38413667678833
    },
    {
      "auxiliary_loss_clip": 0.01252577,
      "auxiliary_loss_mlp": 0.01034485,
      "balance_loss_clip": 1.07047093,
      "balance_loss_mlp": 1.02364874,
      "epoch": 0.10677568688751278,
      "flos": 26140382830080.0,
      "grad_norm": 2.1781077724017046,
      "language_loss": 0.78759515,
      "learning_rate": 3.938568832004475e-06,
      "loss": 0.81046569,
      "num_input_tokens_seen": 18885130,
      "step": 888,
      "time_per_iteration": 2.794687509536743
    },
    {
      "auxiliary_loss_clip": 0.01339379,
      "auxiliary_loss_mlp": 0.01041576,
      "balance_loss_clip": 1.060691,
      "balance_loss_mlp": 1.0306505,
      "epoch": 0.10689592977815186,
      "flos": 12786533712000.0,
      "grad_norm": 2.042285640585899,
      "language_loss": 0.75300163,
      "learning_rate": 3.938377102723257e-06,
      "loss": 0.77681112,
      "num_input_tokens_seen": 18902265,
      "step": 889,
      "time_per_iteration": 2.735180377960205
    },
    {
      "auxiliary_loss_clip": 0.01436911,
      "auxiliary_loss_mlp": 0.01036884,
      "balance_loss_clip": 1.06015277,
      "balance_loss_mlp": 1.02579105,
      "epoch": 0.10701617266879096,
      "flos": 22126683242880.0,
      "grad_norm": 4.156283339285816,
      "language_loss": 0.83418858,
      "learning_rate": 3.938185079391774e-06,
      "loss": 0.85892648,
      "num_input_tokens_seen": 18919310,
      "step": 890,
      "time_per_iteration": 2.8454277515411377
    },
    {
      "auxiliary_loss_clip": 0.01249651,
      "auxiliary_loss_mlp": 0.01035494,
      "balance_loss_clip": 1.06715846,
      "balance_loss_mlp": 1.02519393,
      "epoch": 0.10713641555943004,
      "flos": 19745625559680.0,
      "grad_norm": 3.2083081356734438,
      "language_loss": 1.05870104,
      "learning_rate": 3.937992762039157e-06,
      "loss": 1.08155251,
      "num_input_tokens_seen": 18932635,
      "step": 891,
      "time_per_iteration": 0.005930185317993164
    },
    {
      "auxiliary_loss_clip": 0.01297979,
      "auxiliary_loss_mlp": 0.01043442,
      "balance_loss_clip": 1.06634617,
      "balance_loss_mlp": 1.03342199,
      "epoch": 0.10725665845006914,
      "flos": 23952992302080.0,
      "grad_norm": 1.7548662569013032,
      "language_loss": 0.80428374,
      "learning_rate": 3.937800150694577e-06,
      "loss": 0.82769799,
      "num_input_tokens_seen": 18953810,
      "step": 892,
      "time_per_iteration": 2.703380823135376
    },
    {
      "auxiliary_loss_clip": 0.01446861,
      "auxiliary_loss_mlp": 0.01033506,
      "balance_loss_clip": 1.06270194,
      "balance_loss_mlp": 1.02312875,
      "epoch": 0.10737690134070824,
      "flos": 18551704371840.0,
      "grad_norm": 2.4541229606539363,
      "language_loss": 0.76317835,
      "learning_rate": 3.937607245387255e-06,
      "loss": 0.78798205,
      "num_input_tokens_seen": 18973175,
      "step": 893,
      "time_per_iteration": 2.785367965698242
    },
    {
      "auxiliary_loss_clip": 0.01355567,
      "auxiliary_loss_mlp": 0.01043047,
      "balance_loss_clip": 1.06224954,
      "balance_loss_mlp": 1.03402328,
      "epoch": 0.10749714423134732,
      "flos": 22707609903360.0,
      "grad_norm": 2.02342478890887,
      "language_loss": 0.72013092,
      "learning_rate": 3.937414046146455e-06,
      "loss": 0.74411708,
      "num_input_tokens_seen": 18991130,
      "step": 894,
      "time_per_iteration": 2.687171459197998
    },
    {
      "auxiliary_loss_clip": 0.01257222,
      "auxiliary_loss_mlp": 0.01046536,
      "balance_loss_clip": 1.071931,
      "balance_loss_mlp": 1.03655815,
      "epoch": 0.10761738712198642,
      "flos": 21106066199040.0,
      "grad_norm": 2.5719204794241484,
      "language_loss": 0.75634623,
      "learning_rate": 3.9372205530014845e-06,
      "loss": 0.77938378,
      "num_input_tokens_seen": 19009610,
      "step": 895,
      "time_per_iteration": 2.7022111415863037
    },
    {
      "auxiliary_loss_clip": 0.01252011,
      "auxiliary_loss_mlp": 0.01040888,
      "balance_loss_clip": 1.06602263,
      "balance_loss_mlp": 1.02930081,
      "epoch": 0.1077376300126255,
      "flos": 23766723348480.0,
      "grad_norm": 4.61498356580739,
      "language_loss": 0.71150112,
      "learning_rate": 3.937026765981696e-06,
      "loss": 0.73443007,
      "num_input_tokens_seen": 19029680,
      "step": 896,
      "time_per_iteration": 2.6587748527526855
    },
    {
      "auxiliary_loss_clip": 0.01404176,
      "auxiliary_loss_mlp": 0.01054295,
      "balance_loss_clip": 1.06795788,
      "balance_loss_mlp": 1.04330361,
      "epoch": 0.1078578729032646,
      "flos": 20919581763840.0,
      "grad_norm": 1.9289846450607246,
      "language_loss": 0.79881126,
      "learning_rate": 3.936832685116488e-06,
      "loss": 0.82339597,
      "num_input_tokens_seen": 19047775,
      "step": 897,
      "time_per_iteration": 2.7055270671844482
    },
    {
      "auxiliary_loss_clip": 0.01253639,
      "auxiliary_loss_mlp": 0.0104183,
      "balance_loss_clip": 1.06968713,
      "balance_loss_mlp": 1.03195953,
      "epoch": 0.10797811579390369,
      "flos": 14829886702080.0,
      "grad_norm": 3.3024750994554832,
      "language_loss": 0.90090346,
      "learning_rate": 3.936638310435301e-06,
      "loss": 0.92385823,
      "num_input_tokens_seen": 19065640,
      "step": 898,
      "time_per_iteration": 2.61527681350708
    },
    {
      "auxiliary_loss_clip": 0.01306109,
      "auxiliary_loss_mlp": 0.01039513,
      "balance_loss_clip": 1.06809473,
      "balance_loss_mlp": 1.0296005,
      "epoch": 0.10809835868454278,
      "flos": 19536985411200.0,
      "grad_norm": 2.0540341080333335,
      "language_loss": 0.81639665,
      "learning_rate": 3.936443641967623e-06,
      "loss": 0.83985287,
      "num_input_tokens_seen": 19084470,
      "step": 899,
      "time_per_iteration": 2.733175754547119
    },
    {
      "auxiliary_loss_clip": 0.01354343,
      "auxiliary_loss_mlp": 0.01040532,
      "balance_loss_clip": 1.06639707,
      "balance_loss_mlp": 1.02992201,
      "epoch": 0.10821860157518187,
      "flos": 18442320480000.0,
      "grad_norm": 2.117914242693852,
      "language_loss": 0.83914173,
      "learning_rate": 3.936248679742983e-06,
      "loss": 0.86309052,
      "num_input_tokens_seen": 19102965,
      "step": 900,
      "time_per_iteration": 3.7367053031921387
    },
    {
      "auxiliary_loss_clip": 0.01308046,
      "auxiliary_loss_mlp": 0.01009966,
      "balance_loss_clip": 1.04276538,
      "balance_loss_mlp": 1.00724781,
      "epoch": 0.10833884446582095,
      "flos": 49359468447360.0,
      "grad_norm": 1.2811927196034485,
      "language_loss": 0.70163208,
      "learning_rate": 3.936053423790959e-06,
      "loss": 0.72481221,
      "num_input_tokens_seen": 19151285,
      "step": 901,
      "time_per_iteration": 3.0485939979553223
    },
    {
      "auxiliary_loss_clip": 0.01253927,
      "auxiliary_loss_mlp": 0.01042886,
      "balance_loss_clip": 1.07057929,
      "balance_loss_mlp": 1.03256285,
      "epoch": 0.10845908735646005,
      "flos": 20411912891520.0,
      "grad_norm": 7.687201216450959,
      "language_loss": 0.77794921,
      "learning_rate": 3.935857874141168e-06,
      "loss": 0.80091739,
      "num_input_tokens_seen": 19170120,
      "step": 902,
      "time_per_iteration": 2.7428719997406006
    },
    {
      "auxiliary_loss_clip": 0.01348146,
      "auxiliary_loss_mlp": 0.01044077,
      "balance_loss_clip": 1.06641579,
      "balance_loss_mlp": 1.0340575,
      "epoch": 0.10857933024709913,
      "flos": 14027750133120.0,
      "grad_norm": 2.678966799820402,
      "language_loss": 0.83846188,
      "learning_rate": 3.935662030823279e-06,
      "loss": 0.86238408,
      "num_input_tokens_seen": 19186305,
      "step": 903,
      "time_per_iteration": 4.71454119682312
    },
    {
      "auxiliary_loss_clip": 0.0130413,
      "auxiliary_loss_mlp": 0.01033813,
      "balance_loss_clip": 1.06641448,
      "balance_loss_mlp": 1.0237875,
      "epoch": 0.10869957313773823,
      "flos": 13369004657280.0,
      "grad_norm": 3.310856095184647,
      "language_loss": 0.72771907,
      "learning_rate": 3.935465893866998e-06,
      "loss": 0.75109857,
      "num_input_tokens_seen": 19204530,
      "step": 904,
      "time_per_iteration": 2.6407790184020996
    },
    {
      "auxiliary_loss_clip": 0.01349835,
      "auxiliary_loss_mlp": 0.01037543,
      "balance_loss_clip": 1.06442356,
      "balance_loss_mlp": 1.027035,
      "epoch": 0.10881981602837733,
      "flos": 25807095509760.0,
      "grad_norm": 4.04348022385036,
      "language_loss": 0.8005845,
      "learning_rate": 3.935269463302079e-06,
      "loss": 0.82445836,
      "num_input_tokens_seen": 19222735,
      "step": 905,
      "time_per_iteration": 2.873183012008667
    },
    {
      "auxiliary_loss_clip": 0.01308453,
      "auxiliary_loss_mlp": 0.01043504,
      "balance_loss_clip": 1.0701263,
      "balance_loss_mlp": 1.03274548,
      "epoch": 0.10894005891901641,
      "flos": 20777555387520.0,
      "grad_norm": 2.996045955200057,
      "language_loss": 0.76714158,
      "learning_rate": 3.935072739158322e-06,
      "loss": 0.7906611,
      "num_input_tokens_seen": 19242445,
      "step": 906,
      "time_per_iteration": 2.6403756141662598
    },
    {
      "auxiliary_loss_clip": 0.01352057,
      "auxiliary_loss_mlp": 0.01037713,
      "balance_loss_clip": 1.06362033,
      "balance_loss_mlp": 1.02671564,
      "epoch": 0.10906030180965551,
      "flos": 26649883296000.0,
      "grad_norm": 1.71955212295564,
      "language_loss": 0.8003453,
      "learning_rate": 3.934875721465569e-06,
      "loss": 0.82424307,
      "num_input_tokens_seen": 19262865,
      "step": 907,
      "time_per_iteration": 3.7000839710235596
    },
    {
      "auxiliary_loss_clip": 0.01349796,
      "auxiliary_loss_mlp": 0.01037671,
      "balance_loss_clip": 1.06330407,
      "balance_loss_mlp": 1.02692461,
      "epoch": 0.10918054470029459,
      "flos": 36534402420480.0,
      "grad_norm": 4.759305846650769,
      "language_loss": 0.72060323,
      "learning_rate": 3.9346784102537076e-06,
      "loss": 0.74447793,
      "num_input_tokens_seen": 19285000,
      "step": 908,
      "time_per_iteration": 2.8910715579986572
    },
    {
      "auxiliary_loss_clip": 0.01252803,
      "auxiliary_loss_mlp": 0.01039202,
      "balance_loss_clip": 1.06769538,
      "balance_loss_mlp": 1.02926588,
      "epoch": 0.10930078759093369,
      "flos": 21762549118080.0,
      "grad_norm": 2.65967196188039,
      "language_loss": 0.78405321,
      "learning_rate": 3.934480805552669e-06,
      "loss": 0.80697328,
      "num_input_tokens_seen": 19306010,
      "step": 909,
      "time_per_iteration": 2.6137969493865967
    },
    {
      "auxiliary_loss_clip": 0.01254546,
      "auxiliary_loss_mlp": 0.02587315,
      "balance_loss_clip": 1.07030749,
      "balance_loss_mlp": 1.00029302,
      "epoch": 0.10942103048157277,
      "flos": 22601781457920.0,
      "grad_norm": 2.0403150471959886,
      "language_loss": 0.88165593,
      "learning_rate": 3.93428290739243e-06,
      "loss": 0.92007458,
      "num_input_tokens_seen": 19325380,
      "step": 910,
      "time_per_iteration": 2.737029552459717
    },
    {
      "auxiliary_loss_clip": 0.01354058,
      "auxiliary_loss_mlp": 0.01044498,
      "balance_loss_clip": 1.06546724,
      "balance_loss_mlp": 1.0332799,
      "epoch": 0.10954127337221187,
      "flos": 15045781397760.0,
      "grad_norm": 2.408002519192761,
      "language_loss": 0.79609638,
      "learning_rate": 3.9340847158030125e-06,
      "loss": 0.82008195,
      "num_input_tokens_seen": 19338960,
      "step": 911,
      "time_per_iteration": 2.6169111728668213
    },
    {
      "auxiliary_loss_clip": 0.01304539,
      "auxiliary_loss_mlp": 0.01034871,
      "balance_loss_clip": 1.06643391,
      "balance_loss_mlp": 1.02443993,
      "epoch": 0.10966151626285096,
      "flos": 21650974496640.0,
      "grad_norm": 1.7248700382725897,
      "language_loss": 0.75394684,
      "learning_rate": 3.9338862308144814e-06,
      "loss": 0.77734089,
      "num_input_tokens_seen": 19357780,
      "step": 912,
      "time_per_iteration": 2.6591861248016357
    },
    {
      "auxiliary_loss_clip": 0.01250936,
      "auxiliary_loss_mlp": 0.01035504,
      "balance_loss_clip": 1.06947553,
      "balance_loss_mlp": 1.02498984,
      "epoch": 0.10978175915349005,
      "flos": 20121359777280.0,
      "grad_norm": 1.6797692910536006,
      "language_loss": 0.84625626,
      "learning_rate": 3.933687452456946e-06,
      "loss": 0.8691206,
      "num_input_tokens_seen": 19377680,
      "step": 913,
      "time_per_iteration": 2.5765480995178223
    },
    {
      "auxiliary_loss_clip": 0.01403028,
      "auxiliary_loss_mlp": 0.01039924,
      "balance_loss_clip": 1.06268191,
      "balance_loss_mlp": 1.02856314,
      "epoch": 0.10990200204412914,
      "flos": 20412667077120.0,
      "grad_norm": 2.6919168172385097,
      "language_loss": 0.86240536,
      "learning_rate": 3.933488380760562e-06,
      "loss": 0.88683492,
      "num_input_tokens_seen": 19397040,
      "step": 914,
      "time_per_iteration": 2.7172205448150635
    },
    {
      "auxiliary_loss_clip": 0.01252563,
      "auxiliary_loss_mlp": 0.02588384,
      "balance_loss_clip": 1.06820977,
      "balance_loss_mlp": 1.00037944,
      "epoch": 0.11002224493476823,
      "flos": 17530117660800.0,
      "grad_norm": 2.1949284673590483,
      "language_loss": 0.87353969,
      "learning_rate": 3.9332890157555286e-06,
      "loss": 0.9119491,
      "num_input_tokens_seen": 19413975,
      "step": 915,
      "time_per_iteration": 2.6271986961364746
    },
    {
      "auxiliary_loss_clip": 0.01357455,
      "auxiliary_loss_mlp": 0.01043983,
      "balance_loss_clip": 1.06599438,
      "balance_loss_mlp": 1.03277731,
      "epoch": 0.11014248782540732,
      "flos": 12203093099520.0,
      "grad_norm": 1.9644286166609504,
      "language_loss": 0.76540124,
      "learning_rate": 3.933089357472088e-06,
      "loss": 0.78941566,
      "num_input_tokens_seen": 19432005,
      "step": 916,
      "time_per_iteration": 2.679608106613159
    },
    {
      "auxiliary_loss_clip": 0.01258447,
      "auxiliary_loss_mlp": 0.01043883,
      "balance_loss_clip": 1.07374263,
      "balance_loss_mlp": 1.03411984,
      "epoch": 0.11026273071604642,
      "flos": 22382977760640.0,
      "grad_norm": 2.9629813802026717,
      "language_loss": 0.85751271,
      "learning_rate": 3.932889405940529e-06,
      "loss": 0.88053596,
      "num_input_tokens_seen": 19450100,
      "step": 917,
      "time_per_iteration": 2.6873793601989746
    },
    {
      "auxiliary_loss_clip": 0.01354103,
      "auxiliary_loss_mlp": 0.0104218,
      "balance_loss_clip": 1.06997216,
      "balance_loss_mlp": 1.03103983,
      "epoch": 0.1103829736066855,
      "flos": 19829046896640.0,
      "grad_norm": 2.159129516461089,
      "language_loss": 0.80326855,
      "learning_rate": 3.932689161191184e-06,
      "loss": 0.82723135,
      "num_input_tokens_seen": 19467805,
      "step": 918,
      "time_per_iteration": 2.709218978881836
    },
    {
      "auxiliary_loss_clip": 0.01302088,
      "auxiliary_loss_mlp": 0.01039708,
      "balance_loss_clip": 1.06604373,
      "balance_loss_mlp": 1.02837729,
      "epoch": 0.1105032164973246,
      "flos": 22669616292480.0,
      "grad_norm": 2.3362183205206644,
      "language_loss": 0.87991738,
      "learning_rate": 3.93248862325443e-06,
      "loss": 0.90333539,
      "num_input_tokens_seen": 19486710,
      "step": 919,
      "time_per_iteration": 2.742094039916992
    },
    {
      "auxiliary_loss_clip": 0.01215488,
      "auxiliary_loss_mlp": 0.010044,
      "balance_loss_clip": 1.04336655,
      "balance_loss_mlp": 1.00147939,
      "epoch": 0.11062345938796368,
      "flos": 66483507876480.0,
      "grad_norm": 0.9358174680393039,
      "language_loss": 0.64536786,
      "learning_rate": 3.932287792160688e-06,
      "loss": 0.66756666,
      "num_input_tokens_seen": 19545170,
      "step": 920,
      "time_per_iteration": 3.1785061359405518
    },
    {
      "auxiliary_loss_clip": 0.01299929,
      "auxiliary_loss_mlp": 0.01038685,
      "balance_loss_clip": 1.06241608,
      "balance_loss_mlp": 1.02800417,
      "epoch": 0.11074370227860278,
      "flos": 21907771804800.0,
      "grad_norm": 3.3532758440619856,
      "language_loss": 0.80729139,
      "learning_rate": 3.932086667940424e-06,
      "loss": 0.83067751,
      "num_input_tokens_seen": 19561875,
      "step": 921,
      "time_per_iteration": 2.675508499145508
    },
    {
      "auxiliary_loss_clip": 0.01298808,
      "auxiliary_loss_mlp": 0.02586591,
      "balance_loss_clip": 1.06883168,
      "balance_loss_mlp": 1.00032222,
      "epoch": 0.11086394516924186,
      "flos": 28658115763200.0,
      "grad_norm": 1.9328651123693088,
      "language_loss": 0.81754071,
      "learning_rate": 3.93188525062415e-06,
      "loss": 0.85639471,
      "num_input_tokens_seen": 19582340,
      "step": 922,
      "time_per_iteration": 2.7350330352783203
    },
    {
      "auxiliary_loss_clip": 0.01299129,
      "auxiliary_loss_mlp": 0.01037398,
      "balance_loss_clip": 1.06542182,
      "balance_loss_mlp": 1.0263536,
      "epoch": 0.11098418805988096,
      "flos": 24535247765760.0,
      "grad_norm": 3.4743555845746177,
      "language_loss": 0.8586489,
      "learning_rate": 3.931683540242418e-06,
      "loss": 0.88201416,
      "num_input_tokens_seen": 19603405,
      "step": 923,
      "time_per_iteration": 2.680359363555908
    },
    {
      "auxiliary_loss_clip": 0.01297452,
      "auxiliary_loss_mlp": 0.01036827,
      "balance_loss_clip": 1.06566823,
      "balance_loss_mlp": 1.02632523,
      "epoch": 0.11110443095052006,
      "flos": 22960384888320.0,
      "grad_norm": 2.4994308453705467,
      "language_loss": 0.90992671,
      "learning_rate": 3.9314815368258295e-06,
      "loss": 0.9332695,
      "num_input_tokens_seen": 19619885,
      "step": 924,
      "time_per_iteration": 2.6304702758789062
    },
    {
      "auxiliary_loss_clip": 0.01305023,
      "auxiliary_loss_mlp": 0.01045593,
      "balance_loss_clip": 1.06990492,
      "balance_loss_mlp": 1.03580034,
      "epoch": 0.11122467384115914,
      "flos": 18950025265920.0,
      "grad_norm": 1.8093520709172253,
      "language_loss": 0.78959405,
      "learning_rate": 3.9312792404050275e-06,
      "loss": 0.81310022,
      "num_input_tokens_seen": 19637940,
      "step": 925,
      "time_per_iteration": 2.6697819232940674
    },
    {
      "auxiliary_loss_clip": 0.0125165,
      "auxiliary_loss_mlp": 0.01031898,
      "balance_loss_clip": 1.07032323,
      "balance_loss_mlp": 1.02244508,
      "epoch": 0.11134491673179824,
      "flos": 25082957324160.0,
      "grad_norm": 2.2350297273811406,
      "language_loss": 0.77398568,
      "learning_rate": 3.9310766510107e-06,
      "loss": 0.79682112,
      "num_input_tokens_seen": 19657115,
      "step": 926,
      "time_per_iteration": 3.56718373298645
    },
    {
      "auxiliary_loss_clip": 0.01397904,
      "auxiliary_loss_mlp": 0.01038653,
      "balance_loss_clip": 1.06008863,
      "balance_loss_mlp": 1.02721477,
      "epoch": 0.11146515962243732,
      "flos": 24499121662080.0,
      "grad_norm": 1.9892741705515435,
      "language_loss": 0.91848159,
      "learning_rate": 3.9308737686735806e-06,
      "loss": 0.94284719,
      "num_input_tokens_seen": 19677075,
      "step": 927,
      "time_per_iteration": 2.754479169845581
    },
    {
      "auxiliary_loss_clip": 0.01254744,
      "auxiliary_loss_mlp": 0.01037818,
      "balance_loss_clip": 1.07011282,
      "balance_loss_mlp": 1.02683854,
      "epoch": 0.11158540251307641,
      "flos": 22343763087360.0,
      "grad_norm": 2.8640778254264827,
      "language_loss": 0.82948834,
      "learning_rate": 3.9306705934244455e-06,
      "loss": 0.85241389,
      "num_input_tokens_seen": 19697155,
      "step": 928,
      "time_per_iteration": 2.616848945617676
    },
    {
      "auxiliary_loss_clip": 0.01338124,
      "auxiliary_loss_mlp": 0.01039975,
      "balance_loss_clip": 1.06263351,
      "balance_loss_mlp": 1.03052139,
      "epoch": 0.11170564540371551,
      "flos": 19902304684800.0,
      "grad_norm": 2.8047353156369828,
      "language_loss": 0.88405967,
      "learning_rate": 3.930467125294116e-06,
      "loss": 0.90784073,
      "num_input_tokens_seen": 19716705,
      "step": 929,
      "time_per_iteration": 3.7702300548553467
    },
    {
      "auxiliary_loss_clip": 0.01406965,
      "auxiliary_loss_mlp": 0.01007914,
      "balance_loss_clip": 1.03575706,
      "balance_loss_mlp": 1.00538635,
      "epoch": 0.1118258882943546,
      "flos": 64586239499520.0,
      "grad_norm": 0.9266534493837569,
      "language_loss": 0.60532421,
      "learning_rate": 3.930263364313458e-06,
      "loss": 0.62947297,
      "num_input_tokens_seen": 19767275,
      "step": 930,
      "time_per_iteration": 4.147348880767822
    },
    {
      "auxiliary_loss_clip": 0.01381977,
      "auxiliary_loss_mlp": 0.01049578,
      "balance_loss_clip": 1.06200063,
      "balance_loss_mlp": 1.03805602,
      "epoch": 0.11194613118499369,
      "flos": 17201965985280.0,
      "grad_norm": 1.9138689549311494,
      "language_loss": 0.82737923,
      "learning_rate": 3.930059310513384e-06,
      "loss": 0.85169482,
      "num_input_tokens_seen": 19786315,
      "step": 931,
      "time_per_iteration": 2.7907538414001465
    },
    {
      "auxiliary_loss_clip": 0.0138774,
      "auxiliary_loss_mlp": 0.02585835,
      "balance_loss_clip": 1.0601269,
      "balance_loss_mlp": 1.00035965,
      "epoch": 0.11206637407563277,
      "flos": 31863465728640.0,
      "grad_norm": 1.8509555225997754,
      "language_loss": 0.84270066,
      "learning_rate": 3.929854963924846e-06,
      "loss": 0.88243639,
      "num_input_tokens_seen": 19806580,
      "step": 932,
      "time_per_iteration": 2.8362209796905518
    },
    {
      "auxiliary_loss_clip": 0.01382428,
      "auxiliary_loss_mlp": 0.01044081,
      "balance_loss_clip": 1.06136036,
      "balance_loss_mlp": 1.0343833,
      "epoch": 0.11218661696627187,
      "flos": 21945621761280.0,
      "grad_norm": 2.1952153759155975,
      "language_loss": 0.77816975,
      "learning_rate": 3.929650324578845e-06,
      "loss": 0.80243486,
      "num_input_tokens_seen": 19826045,
      "step": 933,
      "time_per_iteration": 3.8158721923828125
    },
    {
      "auxiliary_loss_clip": 0.01354869,
      "auxiliary_loss_mlp": 0.01037489,
      "balance_loss_clip": 1.06653619,
      "balance_loss_mlp": 1.02600348,
      "epoch": 0.11230685985691095,
      "flos": 25878198481920.0,
      "grad_norm": 2.336349293100345,
      "language_loss": 0.82603884,
      "learning_rate": 3.929445392506423e-06,
      "loss": 0.84996247,
      "num_input_tokens_seen": 19843985,
      "step": 934,
      "time_per_iteration": 2.7915406227111816
    },
    {
      "auxiliary_loss_clip": 0.01300007,
      "auxiliary_loss_mlp": 0.01040095,
      "balance_loss_clip": 1.06908929,
      "balance_loss_mlp": 1.02998042,
      "epoch": 0.11242710274755005,
      "flos": 22231506107520.0,
      "grad_norm": 4.115367789100844,
      "language_loss": 0.76038957,
      "learning_rate": 3.92924016773867e-06,
      "loss": 0.78379059,
      "num_input_tokens_seen": 19860480,
      "step": 935,
      "time_per_iteration": 2.631925344467163
    },
    {
      "auxiliary_loss_clip": 0.01343505,
      "auxiliary_loss_mlp": 0.02581806,
      "balance_loss_clip": 1.06093228,
      "balance_loss_mlp": 1.00039411,
      "epoch": 0.11254734563818915,
      "flos": 17712184723200.0,
      "grad_norm": 2.4182269981929774,
      "language_loss": 0.73939502,
      "learning_rate": 3.9290346503067175e-06,
      "loss": 0.77864814,
      "num_input_tokens_seen": 19877145,
      "step": 936,
      "time_per_iteration": 2.6903791427612305
    },
    {
      "auxiliary_loss_clip": 0.01301275,
      "auxiliary_loss_mlp": 0.01055257,
      "balance_loss_clip": 1.06337106,
      "balance_loss_mlp": 1.04448092,
      "epoch": 0.11266758852882823,
      "flos": 54930397334400.0,
      "grad_norm": 2.7735922374672946,
      "language_loss": 0.78785551,
      "learning_rate": 3.9288288402417415e-06,
      "loss": 0.8114208,
      "num_input_tokens_seen": 19903405,
      "step": 937,
      "time_per_iteration": 2.935434341430664
    },
    {
      "auxiliary_loss_clip": 0.01297928,
      "auxiliary_loss_mlp": 0.01041564,
      "balance_loss_clip": 1.06534338,
      "balance_loss_mlp": 1.03049505,
      "epoch": 0.11278783141946733,
      "flos": 18878132194560.0,
      "grad_norm": 2.7759500971653805,
      "language_loss": 0.70588887,
      "learning_rate": 3.928622737574964e-06,
      "loss": 0.72928381,
      "num_input_tokens_seen": 19918740,
      "step": 938,
      "time_per_iteration": 2.66365122795105
    },
    {
      "auxiliary_loss_clip": 0.01349009,
      "auxiliary_loss_mlp": 0.01038214,
      "balance_loss_clip": 1.06198382,
      "balance_loss_mlp": 1.02767634,
      "epoch": 0.11290807431010641,
      "flos": 26469252777600.0,
      "grad_norm": 2.0556386216432414,
      "language_loss": 0.91194969,
      "learning_rate": 3.928416342337652e-06,
      "loss": 0.93582189,
      "num_input_tokens_seen": 19938475,
      "step": 939,
      "time_per_iteration": 2.772516965866089
    },
    {
      "auxiliary_loss_clip": 0.01326717,
      "auxiliary_loss_mlp": 0.01036958,
      "balance_loss_clip": 1.06284881,
      "balance_loss_mlp": 1.02695084,
      "epoch": 0.1130283172007455,
      "flos": 22710590732160.0,
      "grad_norm": 1.98440891036142,
      "language_loss": 0.82979137,
      "learning_rate": 3.928209654561113e-06,
      "loss": 0.85342813,
      "num_input_tokens_seen": 19959310,
      "step": 940,
      "time_per_iteration": 2.7124621868133545
    },
    {
      "auxiliary_loss_clip": 0.01342596,
      "auxiliary_loss_mlp": 0.01039945,
      "balance_loss_clip": 1.06473947,
      "balance_loss_mlp": 1.0308013,
      "epoch": 0.1131485600913846,
      "flos": 23219911630080.0,
      "grad_norm": 3.360713355129871,
      "language_loss": 0.81731641,
      "learning_rate": 3.928002674276703e-06,
      "loss": 0.84114182,
      "num_input_tokens_seen": 19978700,
      "step": 941,
      "time_per_iteration": 2.716331958770752
    },
    {
      "auxiliary_loss_clip": 0.01426759,
      "auxiliary_loss_mlp": 0.01034156,
      "balance_loss_clip": 1.05110753,
      "balance_loss_mlp": 1.02463746,
      "epoch": 0.11326880298202369,
      "flos": 14064271286400.0,
      "grad_norm": 2.743292811870845,
      "language_loss": 0.76105726,
      "learning_rate": 3.92779540151582e-06,
      "loss": 0.78566641,
      "num_input_tokens_seen": 19995785,
      "step": 942,
      "time_per_iteration": 2.704352378845215
    },
    {
      "auxiliary_loss_clip": 0.01345558,
      "auxiliary_loss_mlp": 0.0104216,
      "balance_loss_clip": 1.06354511,
      "balance_loss_mlp": 1.0319972,
      "epoch": 0.11338904587266278,
      "flos": 16325386479360.0,
      "grad_norm": 1.7849195130141053,
      "language_loss": 0.85699081,
      "learning_rate": 3.927587836309907e-06,
      "loss": 0.88086796,
      "num_input_tokens_seen": 20013615,
      "step": 943,
      "time_per_iteration": 2.6844851970672607
    },
    {
      "auxiliary_loss_clip": 0.0133807,
      "auxiliary_loss_mlp": 0.01036427,
      "balance_loss_clip": 1.05982256,
      "balance_loss_mlp": 1.02607322,
      "epoch": 0.11350928876330187,
      "flos": 24426258923520.0,
      "grad_norm": 2.065523851801602,
      "language_loss": 0.78108442,
      "learning_rate": 3.927379978690452e-06,
      "loss": 0.80482936,
      "num_input_tokens_seen": 20032880,
      "step": 944,
      "time_per_iteration": 2.7149362564086914
    },
    {
      "auxiliary_loss_clip": 0.01387649,
      "auxiliary_loss_mlp": 0.01033203,
      "balance_loss_clip": 1.054883,
      "balance_loss_mlp": 1.02371383,
      "epoch": 0.11362953165394096,
      "flos": 24497074586880.0,
      "grad_norm": 3.35780923108966,
      "language_loss": 0.87341297,
      "learning_rate": 3.927171828688987e-06,
      "loss": 0.89762151,
      "num_input_tokens_seen": 20052405,
      "step": 945,
      "time_per_iteration": 2.792544364929199
    },
    {
      "auxiliary_loss_clip": 0.0124917,
      "auxiliary_loss_mlp": 0.01037912,
      "balance_loss_clip": 1.06823444,
      "balance_loss_mlp": 1.02778506,
      "epoch": 0.11374977454458005,
      "flos": 24060831909120.0,
      "grad_norm": 2.015304643851002,
      "language_loss": 0.82252961,
      "learning_rate": 3.926963386337088e-06,
      "loss": 0.84540039,
      "num_input_tokens_seen": 20070635,
      "step": 946,
      "time_per_iteration": 2.6608996391296387
    },
    {
      "auxiliary_loss_clip": 0.01249864,
      "auxiliary_loss_mlp": 0.01038116,
      "balance_loss_clip": 1.06800818,
      "balance_loss_mlp": 1.02733946,
      "epoch": 0.11387001743521914,
      "flos": 39457638967680.0,
      "grad_norm": 2.718466988154677,
      "language_loss": 0.7013818,
      "learning_rate": 3.926754651666375e-06,
      "loss": 0.72426164,
      "num_input_tokens_seen": 20091195,
      "step": 947,
      "time_per_iteration": 2.773157835006714
    },
    {
      "auxiliary_loss_clip": 0.01375241,
      "auxiliary_loss_mlp": 0.01039547,
      "balance_loss_clip": 1.06284499,
      "balance_loss_mlp": 1.02912164,
      "epoch": 0.11399026032585824,
      "flos": 25082454533760.0,
      "grad_norm": 2.443921416376691,
      "language_loss": 0.78027958,
      "learning_rate": 3.926545624708513e-06,
      "loss": 0.80442744,
      "num_input_tokens_seen": 20110435,
      "step": 948,
      "time_per_iteration": 2.741490125656128
    },
    {
      "auxiliary_loss_clip": 0.01393597,
      "auxiliary_loss_mlp": 0.01034675,
      "balance_loss_clip": 1.05948734,
      "balance_loss_mlp": 1.02458978,
      "epoch": 0.11411050321649732,
      "flos": 17961835224960.0,
      "grad_norm": 2.7358797762620553,
      "language_loss": 0.85522538,
      "learning_rate": 3.926336305495213e-06,
      "loss": 0.87950814,
      "num_input_tokens_seen": 20128995,
      "step": 949,
      "time_per_iteration": 2.7508292198181152
    },
    {
      "auxiliary_loss_clip": 0.01391163,
      "auxiliary_loss_mlp": 0.01039156,
      "balance_loss_clip": 1.06106138,
      "balance_loss_mlp": 1.0291543,
      "epoch": 0.11423074610713642,
      "flos": 22455409536000.0,
      "grad_norm": 2.3064357631925274,
      "language_loss": 0.89140981,
      "learning_rate": 3.926126694058226e-06,
      "loss": 0.91571301,
      "num_input_tokens_seen": 20148145,
      "step": 950,
      "time_per_iteration": 2.8023548126220703
    },
    {
      "auxiliary_loss_clip": 0.01440016,
      "auxiliary_loss_mlp": 0.01034602,
      "balance_loss_clip": 1.06248295,
      "balance_loss_mlp": 1.02488673,
      "epoch": 0.1143509889977755,
      "flos": 19717687756800.0,
      "grad_norm": 1.5640929664457042,
      "language_loss": 0.82092214,
      "learning_rate": 3.92591679042935e-06,
      "loss": 0.84566832,
      "num_input_tokens_seen": 20168035,
      "step": 951,
      "time_per_iteration": 2.8165667057037354
    },
    {
      "auxiliary_loss_clip": 0.01298285,
      "auxiliary_loss_mlp": 0.01038142,
      "balance_loss_clip": 1.06765199,
      "balance_loss_mlp": 1.02779436,
      "epoch": 0.1144712318884146,
      "flos": 19822869757440.0,
      "grad_norm": 2.63934890929552,
      "language_loss": 0.82050836,
      "learning_rate": 3.92570659464043e-06,
      "loss": 0.84387267,
      "num_input_tokens_seen": 20186095,
      "step": 952,
      "time_per_iteration": 3.5639734268188477
    },
    {
      "auxiliary_loss_clip": 0.01295415,
      "auxiliary_loss_mlp": 0.02583641,
      "balance_loss_clip": 1.06744957,
      "balance_loss_mlp": 1.00030112,
      "epoch": 0.1145914747790537,
      "flos": 14939198766720.0,
      "grad_norm": 1.9574977883334987,
      "language_loss": 0.8002255,
      "learning_rate": 3.925496106723349e-06,
      "loss": 0.83901608,
      "num_input_tokens_seen": 20203535,
      "step": 953,
      "time_per_iteration": 2.6000442504882812
    },
    {
      "auxiliary_loss_clip": 0.01299873,
      "auxiliary_loss_mlp": 0.010404,
      "balance_loss_clip": 1.06606269,
      "balance_loss_mlp": 1.030249,
      "epoch": 0.11471171766969278,
      "flos": 19865029345920.0,
      "grad_norm": 4.152941620204811,
      "language_loss": 0.83817637,
      "learning_rate": 3.9252853267100405e-06,
      "loss": 0.86157912,
      "num_input_tokens_seen": 20222780,
      "step": 954,
      "time_per_iteration": 2.695502281188965
    },
    {
      "auxiliary_loss_clip": 0.01387751,
      "auxiliary_loss_mlp": 0.0103965,
      "balance_loss_clip": 1.06066716,
      "balance_loss_mlp": 1.03048277,
      "epoch": 0.11483196056033187,
      "flos": 22526476594560.0,
      "grad_norm": 4.298973238795976,
      "language_loss": 0.8370533,
      "learning_rate": 3.9250742546324786e-06,
      "loss": 0.86132735,
      "num_input_tokens_seen": 20243015,
      "step": 955,
      "time_per_iteration": 3.6716456413269043
    },
    {
      "auxiliary_loss_clip": 0.01344091,
      "auxiliary_loss_mlp": 0.01046641,
      "balance_loss_clip": 1.06067109,
      "balance_loss_mlp": 1.03683639,
      "epoch": 0.11495220345097096,
      "flos": 28220292887040.0,
      "grad_norm": 1.8357602498808363,
      "language_loss": 0.86737049,
      "learning_rate": 3.924862890522683e-06,
      "loss": 0.89127779,
      "num_input_tokens_seen": 20263025,
      "step": 956,
      "time_per_iteration": 3.6321189403533936
    },
    {
      "auxiliary_loss_clip": 0.01299107,
      "auxiliary_loss_mlp": 0.01044323,
      "balance_loss_clip": 1.06559253,
      "balance_loss_mlp": 1.03359985,
      "epoch": 0.11507244634161005,
      "flos": 17492267704320.0,
      "grad_norm": 2.112150001004441,
      "language_loss": 0.86056089,
      "learning_rate": 3.9246512344127174e-06,
      "loss": 0.88399523,
      "num_input_tokens_seen": 20280685,
      "step": 957,
      "time_per_iteration": 2.6986746788024902
    },
    {
      "auxiliary_loss_clip": 0.01475284,
      "auxiliary_loss_mlp": 0.01046149,
      "balance_loss_clip": 1.05355847,
      "balance_loss_mlp": 1.0356344,
      "epoch": 0.11519268923224914,
      "flos": 22564937082240.0,
      "grad_norm": 2.1456747362369946,
      "language_loss": 0.81974339,
      "learning_rate": 3.9244392863346895e-06,
      "loss": 0.84495771,
      "num_input_tokens_seen": 20300090,
      "step": 958,
      "time_per_iteration": 2.7901992797851562
    },
    {
      "auxiliary_loss_clip": 0.01351107,
      "auxiliary_loss_mlp": 0.0104524,
      "balance_loss_clip": 1.06710267,
      "balance_loss_mlp": 1.03456521,
      "epoch": 0.11531293212288823,
      "flos": 16982839065600.0,
      "grad_norm": 1.9924834759980157,
      "language_loss": 0.92308819,
      "learning_rate": 3.9242270463207524e-06,
      "loss": 0.94705164,
      "num_input_tokens_seen": 20318480,
      "step": 959,
      "time_per_iteration": 3.6206157207489014
    },
    {
      "auxiliary_loss_clip": 0.01430773,
      "auxiliary_loss_mlp": 0.01040875,
      "balance_loss_clip": 1.05670798,
      "balance_loss_mlp": 1.03087962,
      "epoch": 0.11543317501352733,
      "flos": 12422004537600.0,
      "grad_norm": 2.6525342704501482,
      "language_loss": 0.852036,
      "learning_rate": 3.924014514403102e-06,
      "loss": 0.8767525,
      "num_input_tokens_seen": 20334635,
      "step": 960,
      "time_per_iteration": 2.78056001663208
    },
    {
      "auxiliary_loss_clip": 0.01435381,
      "auxiliary_loss_mlp": 0.01037067,
      "balance_loss_clip": 1.05815315,
      "balance_loss_mlp": 1.02589762,
      "epoch": 0.11555341790416641,
      "flos": 19821648695040.0,
      "grad_norm": 2.333066277874281,
      "language_loss": 0.91453505,
      "learning_rate": 3.92380169061398e-06,
      "loss": 0.93925947,
      "num_input_tokens_seen": 20352415,
      "step": 961,
      "time_per_iteration": 2.7320709228515625
    },
    {
      "auxiliary_loss_clip": 0.01385034,
      "auxiliary_loss_mlp": 0.02582366,
      "balance_loss_clip": 1.05436361,
      "balance_loss_mlp": 1.00032258,
      "epoch": 0.11567366079480551,
      "flos": 25738865625600.0,
      "grad_norm": 1.9749347459797657,
      "language_loss": 0.84159851,
      "learning_rate": 3.9235885749856705e-06,
      "loss": 0.88127249,
      "num_input_tokens_seen": 20371095,
      "step": 962,
      "time_per_iteration": 2.8205618858337402
    },
    {
      "auxiliary_loss_clip": 0.01341162,
      "auxiliary_loss_mlp": 0.01041957,
      "balance_loss_clip": 1.06419349,
      "balance_loss_mlp": 1.03163397,
      "epoch": 0.1157939036854446,
      "flos": 18223301301120.0,
      "grad_norm": 1.9472174103942141,
      "language_loss": 0.82664955,
      "learning_rate": 3.9233751675505035e-06,
      "loss": 0.85048079,
      "num_input_tokens_seen": 20389805,
      "step": 963,
      "time_per_iteration": 2.6750898361206055
    },
    {
      "auxiliary_loss_clip": 0.01351986,
      "auxiliary_loss_mlp": 0.0103712,
      "balance_loss_clip": 1.07004118,
      "balance_loss_mlp": 1.02702928,
      "epoch": 0.11591414657608369,
      "flos": 23073755189760.0,
      "grad_norm": 3.6312600889808446,
      "language_loss": 0.85303044,
      "learning_rate": 3.923161468340853e-06,
      "loss": 0.87692147,
      "num_input_tokens_seen": 20409640,
      "step": 964,
      "time_per_iteration": 2.6828417778015137
    },
    {
      "auxiliary_loss_clip": 0.01436744,
      "auxiliary_loss_mlp": 0.01042089,
      "balance_loss_clip": 1.05400097,
      "balance_loss_mlp": 1.03075266,
      "epoch": 0.11603438946672277,
      "flos": 19461716461440.0,
      "grad_norm": 1.7240451646107409,
      "language_loss": 0.81613648,
      "learning_rate": 3.9229474773891374e-06,
      "loss": 0.84092486,
      "num_input_tokens_seen": 20428180,
      "step": 965,
      "time_per_iteration": 2.766862154006958
    },
    {
      "auxiliary_loss_clip": 0.01400963,
      "auxiliary_loss_mlp": 0.01035064,
      "balance_loss_clip": 1.05882752,
      "balance_loss_mlp": 1.02462173,
      "epoch": 0.11615463235736187,
      "flos": 26831986272000.0,
      "grad_norm": 1.976973075956202,
      "language_loss": 0.83586073,
      "learning_rate": 3.922733194727818e-06,
      "loss": 0.86022103,
      "num_input_tokens_seen": 20447975,
      "step": 966,
      "time_per_iteration": 2.7730624675750732
    },
    {
      "auxiliary_loss_clip": 0.01299521,
      "auxiliary_loss_mlp": 0.01037239,
      "balance_loss_clip": 1.06662512,
      "balance_loss_mlp": 1.02696276,
      "epoch": 0.11627487524800097,
      "flos": 18580324533120.0,
      "grad_norm": 2.6808299479858557,
      "language_loss": 0.87582767,
      "learning_rate": 3.922518620389402e-06,
      "loss": 0.89919531,
      "num_input_tokens_seen": 20464840,
      "step": 967,
      "time_per_iteration": 2.6121182441711426
    },
    {
      "auxiliary_loss_clip": 0.01522015,
      "auxiliary_loss_mlp": 0.01045597,
      "balance_loss_clip": 1.05368376,
      "balance_loss_mlp": 1.03622687,
      "epoch": 0.11639511813864005,
      "flos": 18150474476160.0,
      "grad_norm": 1.7355461590200734,
      "language_loss": 0.89496952,
      "learning_rate": 3.922303754406439e-06,
      "loss": 0.92064565,
      "num_input_tokens_seen": 20482680,
      "step": 968,
      "time_per_iteration": 3.053508758544922
    },
    {
      "auxiliary_loss_clip": 0.01379253,
      "auxiliary_loss_mlp": 0.01036488,
      "balance_loss_clip": 1.05303955,
      "balance_loss_mlp": 1.02584815,
      "epoch": 0.11651536102927915,
      "flos": 20922023888640.0,
      "grad_norm": 1.8721983851070452,
      "language_loss": 0.79133415,
      "learning_rate": 3.922088596811526e-06,
      "loss": 0.81549156,
      "num_input_tokens_seen": 20501810,
      "step": 969,
      "time_per_iteration": 2.9730894565582275
    },
    {
      "auxiliary_loss_clip": 0.01289138,
      "auxiliary_loss_mlp": 0.01040348,
      "balance_loss_clip": 1.06461978,
      "balance_loss_mlp": 1.03024507,
      "epoch": 0.11663560391991823,
      "flos": 16508602776960.0,
      "grad_norm": 3.410259027726554,
      "language_loss": 0.8675102,
      "learning_rate": 3.9218731476373e-06,
      "loss": 0.89080501,
      "num_input_tokens_seen": 20517995,
      "step": 970,
      "time_per_iteration": 2.6441404819488525
    },
    {
      "auxiliary_loss_clip": 0.01303018,
      "auxiliary_loss_mlp": 0.01043084,
      "balance_loss_clip": 1.06910992,
      "balance_loss_mlp": 1.0314312,
      "epoch": 0.11675584681055733,
      "flos": 19865029345920.0,
      "grad_norm": 1.9213309372151195,
      "language_loss": 0.84530175,
      "learning_rate": 3.9216574069164455e-06,
      "loss": 0.86876273,
      "num_input_tokens_seen": 20536970,
      "step": 971,
      "time_per_iteration": 2.7098774909973145
    },
    {
      "auxiliary_loss_clip": 0.01245158,
      "auxiliary_loss_mlp": 0.01037939,
      "balance_loss_clip": 1.06823993,
      "balance_loss_mlp": 1.02820015,
      "epoch": 0.11687608970119642,
      "flos": 21944364785280.0,
      "grad_norm": 1.7724453994663179,
      "language_loss": 0.79902267,
      "learning_rate": 3.921441374681691e-06,
      "loss": 0.82185364,
      "num_input_tokens_seen": 20557030,
      "step": 972,
      "time_per_iteration": 2.655426025390625
    },
    {
      "auxiliary_loss_clip": 0.01340726,
      "auxiliary_loss_mlp": 0.01039599,
      "balance_loss_clip": 1.06498873,
      "balance_loss_mlp": 1.03050923,
      "epoch": 0.1169963325918355,
      "flos": 24061155131520.0,
      "grad_norm": 1.9599601951144572,
      "language_loss": 0.6481657,
      "learning_rate": 3.921225050965808e-06,
      "loss": 0.67196894,
      "num_input_tokens_seen": 20576915,
      "step": 973,
      "time_per_iteration": 2.731170177459717
    },
    {
      "auxiliary_loss_clip": 0.01392906,
      "auxiliary_loss_mlp": 0.01034822,
      "balance_loss_clip": 1.06260562,
      "balance_loss_mlp": 1.02460587,
      "epoch": 0.1171165754824746,
      "flos": 23368151059200.0,
      "grad_norm": 1.998440154152543,
      "language_loss": 0.75075018,
      "learning_rate": 3.921008435801612e-06,
      "loss": 0.77502751,
      "num_input_tokens_seen": 20596000,
      "step": 974,
      "time_per_iteration": 2.762303352355957
    },
    {
      "auxiliary_loss_clip": 0.01288288,
      "auxiliary_loss_mlp": 0.01035025,
      "balance_loss_clip": 1.06354105,
      "balance_loss_mlp": 1.02501702,
      "epoch": 0.11723681837311369,
      "flos": 18552243075840.0,
      "grad_norm": 2.0966722694374735,
      "language_loss": 0.75541955,
      "learning_rate": 3.920791529221963e-06,
      "loss": 0.77865267,
      "num_input_tokens_seen": 20614675,
      "step": 975,
      "time_per_iteration": 2.6846323013305664
    },
    {
      "auxiliary_loss_clip": 0.0134597,
      "auxiliary_loss_mlp": 0.02582581,
      "balance_loss_clip": 1.06344736,
      "balance_loss_mlp": 1.00033689,
      "epoch": 0.11735706126375278,
      "flos": 23550541344000.0,
      "grad_norm": 1.7053129909385294,
      "language_loss": 0.76786637,
      "learning_rate": 3.920574331259768e-06,
      "loss": 0.80715191,
      "num_input_tokens_seen": 20635875,
      "step": 976,
      "time_per_iteration": 2.8217179775238037
    },
    {
      "auxiliary_loss_clip": 0.01333384,
      "auxiliary_loss_mlp": 0.0104544,
      "balance_loss_clip": 1.06053913,
      "balance_loss_mlp": 1.03543222,
      "epoch": 0.11747730415439187,
      "flos": 22381541216640.0,
      "grad_norm": 2.4338673890569114,
      "language_loss": 0.80027485,
      "learning_rate": 3.9203568419479716e-06,
      "loss": 0.82406306,
      "num_input_tokens_seen": 20656430,
      "step": 977,
      "time_per_iteration": 4.004366636276245
    },
    {
      "auxiliary_loss_clip": 0.01318425,
      "auxiliary_loss_mlp": 0.0103215,
      "balance_loss_clip": 1.06452751,
      "balance_loss_mlp": 1.02276218,
      "epoch": 0.11759754704503096,
      "flos": 22200731130240.0,
      "grad_norm": 1.803150030159134,
      "language_loss": 0.75081146,
      "learning_rate": 3.92013906131957e-06,
      "loss": 0.7743172,
      "num_input_tokens_seen": 20675360,
      "step": 978,
      "time_per_iteration": 2.820523500442505
    },
    {
      "auxiliary_loss_clip": 0.01393745,
      "auxiliary_loss_mlp": 0.01039788,
      "balance_loss_clip": 1.06206059,
      "balance_loss_mlp": 1.03015018,
      "epoch": 0.11771778993567006,
      "flos": 22309755886080.0,
      "grad_norm": 3.5798691031246963,
      "language_loss": 0.82059491,
      "learning_rate": 3.9199209894076e-06,
      "loss": 0.84493023,
      "num_input_tokens_seen": 20695675,
      "step": 979,
      "time_per_iteration": 2.8071773052215576
    },
    {
      "auxiliary_loss_clip": 0.01245779,
      "auxiliary_loss_mlp": 0.01036261,
      "balance_loss_clip": 1.06479585,
      "balance_loss_mlp": 1.026057,
      "epoch": 0.11783803282630914,
      "flos": 21288169175040.0,
      "grad_norm": 1.9029937296986277,
      "language_loss": 0.90398026,
      "learning_rate": 3.919702626245142e-06,
      "loss": 0.92680061,
      "num_input_tokens_seen": 20715330,
      "step": 980,
      "time_per_iteration": 2.648801565170288
    },
    {
      "auxiliary_loss_clip": 0.01331912,
      "auxiliary_loss_mlp": 0.01041752,
      "balance_loss_clip": 1.05808067,
      "balance_loss_mlp": 1.03161287,
      "epoch": 0.11795827571694824,
      "flos": 25371535190400.0,
      "grad_norm": 2.204839533090296,
      "language_loss": 0.66159213,
      "learning_rate": 3.919483971865322e-06,
      "loss": 0.68532872,
      "num_input_tokens_seen": 20735325,
      "step": 981,
      "time_per_iteration": 3.7246944904327393
    },
    {
      "auxiliary_loss_clip": 0.01344606,
      "auxiliary_loss_mlp": 0.01046283,
      "balance_loss_clip": 1.06710267,
      "balance_loss_mlp": 1.036412,
      "epoch": 0.11807851860758732,
      "flos": 23622218933760.0,
      "grad_norm": 4.963403555527744,
      "language_loss": 0.87475848,
      "learning_rate": 3.91926502630131e-06,
      "loss": 0.89866734,
      "num_input_tokens_seen": 20755940,
      "step": 982,
      "time_per_iteration": 2.68692946434021
    },
    {
      "auxiliary_loss_clip": 0.01297968,
      "auxiliary_loss_mlp": 0.01044138,
      "balance_loss_clip": 1.06936026,
      "balance_loss_mlp": 1.03451765,
      "epoch": 0.11819876149822642,
      "flos": 24972496024320.0,
      "grad_norm": 1.7667037418856715,
      "language_loss": 0.72240376,
      "learning_rate": 3.91904578958632e-06,
      "loss": 0.74582481,
      "num_input_tokens_seen": 20775355,
      "step": 983,
      "time_per_iteration": 2.748490333557129
    },
    {
      "auxiliary_loss_clip": 0.01247141,
      "auxiliary_loss_mlp": 0.01038992,
      "balance_loss_clip": 1.06711888,
      "balance_loss_mlp": 1.02873993,
      "epoch": 0.11831900438886551,
      "flos": 23003226835200.0,
      "grad_norm": 2.028104802745636,
      "language_loss": 0.84126282,
      "learning_rate": 3.918826261753608e-06,
      "loss": 0.86412418,
      "num_input_tokens_seen": 20794935,
      "step": 984,
      "time_per_iteration": 2.6228108406066895
    },
    {
      "auxiliary_loss_clip": 0.01342457,
      "auxiliary_loss_mlp": 0.01042555,
      "balance_loss_clip": 1.06367612,
      "balance_loss_mlp": 1.03381729,
      "epoch": 0.1184392472795046,
      "flos": 27965147604480.0,
      "grad_norm": 2.752864549590666,
      "language_loss": 0.71450686,
      "learning_rate": 3.918606442836478e-06,
      "loss": 0.73835695,
      "num_input_tokens_seen": 20817155,
      "step": 985,
      "time_per_iteration": 3.6273419857025146
    },
    {
      "auxiliary_loss_clip": 0.01295527,
      "auxiliary_loss_mlp": 0.01032982,
      "balance_loss_clip": 1.06925452,
      "balance_loss_mlp": 1.02304566,
      "epoch": 0.1185594901701437,
      "flos": 19898497843200.0,
      "grad_norm": 1.7983104691525906,
      "language_loss": 0.77599549,
      "learning_rate": 3.918386332868277e-06,
      "loss": 0.79928058,
      "num_input_tokens_seen": 20835125,
      "step": 986,
      "time_per_iteration": 2.6986594200134277
    },
    {
      "auxiliary_loss_clip": 0.01288109,
      "auxiliary_loss_mlp": 0.01037349,
      "balance_loss_clip": 1.06375456,
      "balance_loss_mlp": 1.0277046,
      "epoch": 0.11867973306078278,
      "flos": 18912354877440.0,
      "grad_norm": 2.1975139163730986,
      "language_loss": 0.94674289,
      "learning_rate": 3.918165931882394e-06,
      "loss": 0.96999753,
      "num_input_tokens_seen": 20853525,
      "step": 987,
      "time_per_iteration": 2.6833527088165283
    },
    {
      "auxiliary_loss_clip": 0.01482065,
      "auxiliary_loss_mlp": 0.01039819,
      "balance_loss_clip": 1.05339384,
      "balance_loss_mlp": 1.03036571,
      "epoch": 0.11879997595142187,
      "flos": 16982803152000.0,
      "grad_norm": 2.813460551244486,
      "language_loss": 0.75454199,
      "learning_rate": 3.917945239912264e-06,
      "loss": 0.77976084,
      "num_input_tokens_seen": 20871000,
      "step": 988,
      "time_per_iteration": 2.865227699279785
    },
    {
      "auxiliary_loss_clip": 0.01437958,
      "auxiliary_loss_mlp": 0.01036448,
      "balance_loss_clip": 1.05950737,
      "balance_loss_mlp": 1.02760267,
      "epoch": 0.11892021884206096,
      "flos": 17530369056000.0,
      "grad_norm": 2.16015198006088,
      "language_loss": 0.75424045,
      "learning_rate": 3.917724256991367e-06,
      "loss": 0.77898455,
      "num_input_tokens_seen": 20889745,
      "step": 989,
      "time_per_iteration": 2.923170328140259
    },
    {
      "auxiliary_loss_clip": 0.01334385,
      "auxiliary_loss_mlp": 0.01037677,
      "balance_loss_clip": 1.06268024,
      "balance_loss_mlp": 1.02790189,
      "epoch": 0.11904046173270005,
      "flos": 30955895763840.0,
      "grad_norm": 1.9134824509477115,
      "language_loss": 0.81484413,
      "learning_rate": 3.9175029831532245e-06,
      "loss": 0.83856475,
      "num_input_tokens_seen": 20909260,
      "step": 990,
      "time_per_iteration": 2.776963233947754
    },
    {
      "auxiliary_loss_clip": 0.0139553,
      "auxiliary_loss_mlp": 0.01035932,
      "balance_loss_clip": 1.06298769,
      "balance_loss_mlp": 1.02626419,
      "epoch": 0.11916070462333915,
      "flos": 20157234485760.0,
      "grad_norm": 1.998877878452502,
      "language_loss": 0.88408726,
      "learning_rate": 3.917281418431404e-06,
      "loss": 0.90840197,
      "num_input_tokens_seen": 20928305,
      "step": 991,
      "time_per_iteration": 2.6764466762542725
    },
    {
      "auxiliary_loss_clip": 0.01342878,
      "auxiliary_loss_mlp": 0.01039267,
      "balance_loss_clip": 1.06689668,
      "balance_loss_mlp": 1.02998018,
      "epoch": 0.11928094751397823,
      "flos": 23551115961600.0,
      "grad_norm": 2.8704474798227673,
      "language_loss": 0.76994747,
      "learning_rate": 3.917059562859516e-06,
      "loss": 0.79376888,
      "num_input_tokens_seen": 20947630,
      "step": 992,
      "time_per_iteration": 2.7464077472686768
    },
    {
      "auxiliary_loss_clip": 0.01335828,
      "auxiliary_loss_mlp": 0.01036728,
      "balance_loss_clip": 1.06329429,
      "balance_loss_mlp": 1.02782345,
      "epoch": 0.11940119040461733,
      "flos": 23908426502400.0,
      "grad_norm": 2.12705092378322,
      "language_loss": 0.8850342,
      "learning_rate": 3.916837416471218e-06,
      "loss": 0.90875977,
      "num_input_tokens_seen": 20964250,
      "step": 993,
      "time_per_iteration": 2.7478911876678467
    },
    {
      "auxiliary_loss_clip": 0.01289526,
      "auxiliary_loss_mlp": 0.01034342,
      "balance_loss_clip": 1.06192303,
      "balance_loss_mlp": 1.02556229,
      "epoch": 0.11952143329525641,
      "flos": 13844533835520.0,
      "grad_norm": 2.792484801835597,
      "language_loss": 0.73033917,
      "learning_rate": 3.916614979300207e-06,
      "loss": 0.75357783,
      "num_input_tokens_seen": 20979095,
      "step": 994,
      "time_per_iteration": 2.659621238708496
    },
    {
      "auxiliary_loss_clip": 0.01435876,
      "auxiliary_loss_mlp": 0.01040258,
      "balance_loss_clip": 1.05965388,
      "balance_loss_mlp": 1.0308764,
      "epoch": 0.11964167618589551,
      "flos": 27015525792000.0,
      "grad_norm": 1.723592125533583,
      "language_loss": 0.78637284,
      "learning_rate": 3.9163922513802274e-06,
      "loss": 0.81113416,
      "num_input_tokens_seen": 21001430,
      "step": 995,
      "time_per_iteration": 2.8458261489868164
    },
    {
      "auxiliary_loss_clip": 0.01247724,
      "auxiliary_loss_mlp": 0.0103438,
      "balance_loss_clip": 1.06698012,
      "balance_loss_mlp": 1.02375245,
      "epoch": 0.1197619190765346,
      "flos": 12567622273920.0,
      "grad_norm": 3.0957854067896995,
      "language_loss": 0.83295041,
      "learning_rate": 3.916169232745067e-06,
      "loss": 0.85577142,
      "num_input_tokens_seen": 21019105,
      "step": 996,
      "time_per_iteration": 2.569119691848755
    },
    {
      "auxiliary_loss_clip": 0.0133529,
      "auxiliary_loss_mlp": 0.01037084,
      "balance_loss_clip": 1.06155181,
      "balance_loss_mlp": 1.02631319,
      "epoch": 0.11988216196717369,
      "flos": 16909437623040.0,
      "grad_norm": 2.657099925408867,
      "language_loss": 0.92236924,
      "learning_rate": 3.915945923428559e-06,
      "loss": 0.94609296,
      "num_input_tokens_seen": 21035630,
      "step": 997,
      "time_per_iteration": 2.7391390800476074
    },
    {
      "auxiliary_loss_clip": 0.01291913,
      "auxiliary_loss_mlp": 0.01036491,
      "balance_loss_clip": 1.0622673,
      "balance_loss_mlp": 1.02692449,
      "epoch": 0.12000240485781279,
      "flos": 16216577205120.0,
      "grad_norm": 2.5313805168709886,
      "language_loss": 0.83239079,
      "learning_rate": 3.915722323464577e-06,
      "loss": 0.8556748,
      "num_input_tokens_seen": 21054235,
      "step": 998,
      "time_per_iteration": 2.6309993267059326
    },
    {
      "auxiliary_loss_clip": 0.0129161,
      "auxiliary_loss_mlp": 0.01038423,
      "balance_loss_clip": 1.06459343,
      "balance_loss_mlp": 1.02897549,
      "epoch": 0.12012264774845187,
      "flos": 49344887525760.0,
      "grad_norm": 3.853325400645241,
      "language_loss": 0.70694077,
      "learning_rate": 3.91549843288704e-06,
      "loss": 0.73024106,
      "num_input_tokens_seen": 21077915,
      "step": 999,
      "time_per_iteration": 2.907522678375244
    },
    {
      "auxiliary_loss_clip": 0.01393864,
      "auxiliary_loss_mlp": 0.02582726,
      "balance_loss_clip": 1.05837131,
      "balance_loss_mlp": 1.00040722,
      "epoch": 0.12024289063909097,
      "flos": 26979435601920.0,
      "grad_norm": 2.886786809362589,
      "language_loss": 0.79168689,
      "learning_rate": 3.915274251729916e-06,
      "loss": 0.83145285,
      "num_input_tokens_seen": 21099205,
      "step": 1000,
      "time_per_iteration": 2.7574703693389893
    },
    {
      "auxiliary_loss_clip": 0.01398874,
      "auxiliary_loss_mlp": 0.01038683,
      "balance_loss_clip": 1.06489754,
      "balance_loss_mlp": 1.02866936,
      "epoch": 0.12036313352973005,
      "flos": 19537308633600.0,
      "grad_norm": 2.029983095546796,
      "language_loss": 0.90217364,
      "learning_rate": 3.91504978002721e-06,
      "loss": 0.9265492,
      "num_input_tokens_seen": 21118260,
      "step": 1001,
      "time_per_iteration": 2.913487672805786
    },
    {
      "auxiliary_loss_clip": 0.01319676,
      "auxiliary_loss_mlp": 0.02583087,
      "balance_loss_clip": 1.06075978,
      "balance_loss_mlp": 1.00035501,
      "epoch": 0.12048337642036915,
      "flos": 17268256535040.0,
      "grad_norm": 1.9447313997119695,
      "language_loss": 0.76220214,
      "learning_rate": 3.914825017812974e-06,
      "loss": 0.80122972,
      "num_input_tokens_seen": 21134910,
      "step": 1002,
      "time_per_iteration": 2.6729581356048584
    },
    {
      "auxiliary_loss_clip": 0.01317772,
      "auxiliary_loss_mlp": 0.01039911,
      "balance_loss_clip": 1.06616306,
      "balance_loss_mlp": 1.03001702,
      "epoch": 0.12060361931100824,
      "flos": 22856962654080.0,
      "grad_norm": 2.2885039956428392,
      "language_loss": 0.72545922,
      "learning_rate": 3.9145999651213065e-06,
      "loss": 0.74903607,
      "num_input_tokens_seen": 21154150,
      "step": 1003,
      "time_per_iteration": 3.735759973526001
    },
    {
      "auxiliary_loss_clip": 0.01293444,
      "auxiliary_loss_mlp": 0.01035621,
      "balance_loss_clip": 1.06507087,
      "balance_loss_mlp": 1.02579808,
      "epoch": 0.12072386220164733,
      "flos": 16726795943040.0,
      "grad_norm": 2.2614816570193716,
      "language_loss": 0.88570404,
      "learning_rate": 3.9143746219863465e-06,
      "loss": 0.90899467,
      "num_input_tokens_seen": 21171255,
      "step": 1004,
      "time_per_iteration": 2.672090768814087
    },
    {
      "auxiliary_loss_clip": 0.01278261,
      "auxiliary_loss_mlp": 0.0100214,
      "balance_loss_clip": 1.05397677,
      "balance_loss_mlp": 0.99982768,
      "epoch": 0.12084410509228642,
      "flos": 55144176105600.0,
      "grad_norm": 0.9189653375814852,
      "language_loss": 0.64760828,
      "learning_rate": 3.914148988442278e-06,
      "loss": 0.6704123,
      "num_input_tokens_seen": 21227045,
      "step": 1005,
      "time_per_iteration": 3.197863817214966
    },
    {
      "auxiliary_loss_clip": 0.01337729,
      "auxiliary_loss_mlp": 0.01038662,
      "balance_loss_clip": 1.06509173,
      "balance_loss_mlp": 1.0276711,
      "epoch": 0.1209643479829255,
      "flos": 26760236855040.0,
      "grad_norm": 2.640161586922222,
      "language_loss": 0.9527272,
      "learning_rate": 3.91392306452333e-06,
      "loss": 0.97649121,
      "num_input_tokens_seen": 21244120,
      "step": 1006,
      "time_per_iteration": 2.7459659576416016
    },
    {
      "auxiliary_loss_clip": 0.01251836,
      "auxiliary_loss_mlp": 0.0103907,
      "balance_loss_clip": 1.06962454,
      "balance_loss_mlp": 1.02856779,
      "epoch": 0.1210845908735646,
      "flos": 11035026725760.0,
      "grad_norm": 3.4409086786914687,
      "language_loss": 0.66485387,
      "learning_rate": 3.913696850263774e-06,
      "loss": 0.68776298,
      "num_input_tokens_seen": 21258485,
      "step": 1007,
      "time_per_iteration": 3.503448247909546
    },
    {
      "auxiliary_loss_clip": 0.01294444,
      "auxiliary_loss_mlp": 0.01036672,
      "balance_loss_clip": 1.06367779,
      "balance_loss_mlp": 1.02633619,
      "epoch": 0.1212048337642037,
      "flos": 20484631975680.0,
      "grad_norm": 2.5282719155222253,
      "language_loss": 0.79048109,
      "learning_rate": 3.913470345697929e-06,
      "loss": 0.81379229,
      "num_input_tokens_seen": 21277115,
      "step": 1008,
      "time_per_iteration": 3.547529935836792
    },
    {
      "auxiliary_loss_clip": 0.01419247,
      "auxiliary_loss_mlp": 0.01041107,
      "balance_loss_clip": 1.06251788,
      "balance_loss_mlp": 1.0303843,
      "epoch": 0.12132507665484278,
      "flos": 22346061557760.0,
      "grad_norm": 2.3257032300983194,
      "language_loss": 0.85947615,
      "learning_rate": 3.913243550860153e-06,
      "loss": 0.88407969,
      "num_input_tokens_seen": 21294880,
      "step": 1009,
      "time_per_iteration": 2.7789196968078613
    },
    {
      "auxiliary_loss_clip": 0.01295145,
      "auxiliary_loss_mlp": 0.0102929,
      "balance_loss_clip": 1.06855869,
      "balance_loss_mlp": 1.01896012,
      "epoch": 0.12144531954548188,
      "flos": 29314957818240.0,
      "grad_norm": 2.1232707265482045,
      "language_loss": 0.7597481,
      "learning_rate": 3.913016465784852e-06,
      "loss": 0.78299248,
      "num_input_tokens_seen": 21315555,
      "step": 1010,
      "time_per_iteration": 2.715496301651001
    },
    {
      "auxiliary_loss_clip": 0.0143896,
      "auxiliary_loss_mlp": 0.0103346,
      "balance_loss_clip": 1.05747843,
      "balance_loss_mlp": 1.02323818,
      "epoch": 0.12156556243612096,
      "flos": 20485242506880.0,
      "grad_norm": 2.3670575654094903,
      "language_loss": 0.7224142,
      "learning_rate": 3.912789090506474e-06,
      "loss": 0.74713838,
      "num_input_tokens_seen": 21334815,
      "step": 1011,
      "time_per_iteration": 3.6896867752075195
    },
    {
      "auxiliary_loss_clip": 0.01393311,
      "auxiliary_loss_mlp": 0.01035101,
      "balance_loss_clip": 1.05642164,
      "balance_loss_mlp": 1.02501631,
      "epoch": 0.12168580532676006,
      "flos": 16472009796480.0,
      "grad_norm": 2.215584114019043,
      "language_loss": 0.72113407,
      "learning_rate": 3.9125614250595114e-06,
      "loss": 0.74541819,
      "num_input_tokens_seen": 21351025,
      "step": 1012,
      "time_per_iteration": 2.8120834827423096
    },
    {
      "auxiliary_loss_clip": 0.01298827,
      "auxiliary_loss_mlp": 0.01037799,
      "balance_loss_clip": 1.06705296,
      "balance_loss_mlp": 1.02724862,
      "epoch": 0.12180604821739914,
      "flos": 15341290588800.0,
      "grad_norm": 2.5152550728034018,
      "language_loss": 0.88804054,
      "learning_rate": 3.912333469478502e-06,
      "loss": 0.91140687,
      "num_input_tokens_seen": 21368990,
      "step": 1013,
      "time_per_iteration": 2.6352038383483887
    },
    {
      "auxiliary_loss_clip": 0.0134446,
      "auxiliary_loss_mlp": 0.01040491,
      "balance_loss_clip": 1.06117713,
      "balance_loss_mlp": 1.03021526,
      "epoch": 0.12192629110803824,
      "flos": 19318038059520.0,
      "grad_norm": 2.598584918695924,
      "language_loss": 0.78111452,
      "learning_rate": 3.912105223798025e-06,
      "loss": 0.80496407,
      "num_input_tokens_seen": 21388410,
      "step": 1014,
      "time_per_iteration": 2.7253355979919434
    },
    {
      "auxiliary_loss_clip": 0.01319478,
      "auxiliary_loss_mlp": 0.01004546,
      "balance_loss_clip": 1.04120696,
      "balance_loss_mlp": 1.0022099,
      "epoch": 0.12204653399867733,
      "flos": 47725354085760.0,
      "grad_norm": 1.0006643638663886,
      "language_loss": 0.67604887,
      "learning_rate": 3.9118766880527065e-06,
      "loss": 0.69928914,
      "num_input_tokens_seen": 21442845,
      "step": 1015,
      "time_per_iteration": 3.2389609813690186
    },
    {
      "auxiliary_loss_clip": 0.01439881,
      "auxiliary_loss_mlp": 0.01041503,
      "balance_loss_clip": 1.06053257,
      "balance_loss_mlp": 1.03163886,
      "epoch": 0.12216677688931642,
      "flos": 18221936584320.0,
      "grad_norm": 1.716749675042218,
      "language_loss": 0.73764169,
      "learning_rate": 3.9116478622772145e-06,
      "loss": 0.76245558,
      "num_input_tokens_seen": 21461420,
      "step": 1016,
      "time_per_iteration": 2.8211042881011963
    },
    {
      "auxiliary_loss_clip": 0.01295985,
      "auxiliary_loss_mlp": 0.01042186,
      "balance_loss_clip": 1.06858826,
      "balance_loss_mlp": 1.03234506,
      "epoch": 0.12228701977995551,
      "flos": 27525636789120.0,
      "grad_norm": 1.723780567005092,
      "language_loss": 0.880144,
      "learning_rate": 3.911418746506261e-06,
      "loss": 0.90352571,
      "num_input_tokens_seen": 21481550,
      "step": 1017,
      "time_per_iteration": 2.732898473739624
    },
    {
      "auxiliary_loss_clip": 0.01299059,
      "auxiliary_loss_mlp": 0.01040168,
      "balance_loss_clip": 1.06894219,
      "balance_loss_mlp": 1.02965951,
      "epoch": 0.1224072626705946,
      "flos": 21798136517760.0,
      "grad_norm": 1.930967321459311,
      "language_loss": 0.78558606,
      "learning_rate": 3.911189340774604e-06,
      "loss": 0.80897838,
      "num_input_tokens_seen": 21501680,
      "step": 1018,
      "time_per_iteration": 2.631235361099243
    },
    {
      "auxiliary_loss_clip": 0.01353583,
      "auxiliary_loss_mlp": 0.01036554,
      "balance_loss_clip": 1.0645206,
      "balance_loss_mlp": 1.02711308,
      "epoch": 0.1225275055612337,
      "flos": 20703758895360.0,
      "grad_norm": 1.7230595805222502,
      "language_loss": 0.79995632,
      "learning_rate": 3.910959645117043e-06,
      "loss": 0.82385767,
      "num_input_tokens_seen": 21521015,
      "step": 1019,
      "time_per_iteration": 2.7133617401123047
    },
    {
      "auxiliary_loss_clip": 0.01260387,
      "auxiliary_loss_mlp": 0.02535217,
      "balance_loss_clip": 1.04663801,
      "balance_loss_mlp": 0.99988663,
      "epoch": 0.12264774845187278,
      "flos": 57745294462080.0,
      "grad_norm": 0.8181347583838138,
      "language_loss": 0.56718558,
      "learning_rate": 3.910729659568423e-06,
      "loss": 0.6051417,
      "num_input_tokens_seen": 21578200,
      "step": 1020,
      "time_per_iteration": 3.2254464626312256
    },
    {
      "auxiliary_loss_clip": 0.01343084,
      "auxiliary_loss_mlp": 0.0103881,
      "balance_loss_clip": 1.06438541,
      "balance_loss_mlp": 1.02861798,
      "epoch": 0.12276799134251187,
      "flos": 26396282298240.0,
      "grad_norm": 2.22206034318736,
      "language_loss": 0.8200112,
      "learning_rate": 3.9104993841636344e-06,
      "loss": 0.84383011,
      "num_input_tokens_seen": 21598770,
      "step": 1021,
      "time_per_iteration": 2.7387354373931885
    },
    {
      "auxiliary_loss_clip": 0.01340704,
      "auxiliary_loss_mlp": 0.0257778,
      "balance_loss_clip": 1.06590962,
      "balance_loss_mlp": 1.00028014,
      "epoch": 0.12288823423315097,
      "flos": 21064193919360.0,
      "grad_norm": 1.9110389560717111,
      "language_loss": 0.80865687,
      "learning_rate": 3.910268818937608e-06,
      "loss": 0.84784174,
      "num_input_tokens_seen": 21616925,
      "step": 1022,
      "time_per_iteration": 2.6528477668762207
    },
    {
      "auxiliary_loss_clip": 0.01406612,
      "auxiliary_loss_mlp": 0.01046553,
      "balance_loss_clip": 1.0599643,
      "balance_loss_mlp": 1.03695643,
      "epoch": 0.12300847712379005,
      "flos": 12312441077760.0,
      "grad_norm": 2.7178772481316575,
      "language_loss": 0.87827396,
      "learning_rate": 3.9100379639253196e-06,
      "loss": 0.90280557,
      "num_input_tokens_seen": 21633645,
      "step": 1023,
      "time_per_iteration": 2.7831780910491943
    },
    {
      "auxiliary_loss_clip": 0.01340408,
      "auxiliary_loss_mlp": 0.01035153,
      "balance_loss_clip": 1.05965173,
      "balance_loss_mlp": 1.02589619,
      "epoch": 0.12312872001442915,
      "flos": 16762239688320.0,
      "grad_norm": 2.990457049798616,
      "language_loss": 0.86351806,
      "learning_rate": 3.909806819161791e-06,
      "loss": 0.88727373,
      "num_input_tokens_seen": 21649120,
      "step": 1024,
      "time_per_iteration": 2.639815330505371
    },
    {
      "auxiliary_loss_clip": 0.01396582,
      "auxiliary_loss_mlp": 0.01036494,
      "balance_loss_clip": 1.06094265,
      "balance_loss_mlp": 1.02584219,
      "epoch": 0.12324896290506823,
      "flos": 18404937400320.0,
      "grad_norm": 1.9629491966316894,
      "language_loss": 0.86425287,
      "learning_rate": 3.909575384682086e-06,
      "loss": 0.8885836,
      "num_input_tokens_seen": 21668000,
      "step": 1025,
      "time_per_iteration": 2.740975856781006
    },
    {
      "auxiliary_loss_clip": 0.01296766,
      "auxiliary_loss_mlp": 0.01039074,
      "balance_loss_clip": 1.06305289,
      "balance_loss_mlp": 1.02964473,
      "epoch": 0.12336920579570733,
      "flos": 18915407533440.0,
      "grad_norm": 2.1534715785076934,
      "language_loss": 0.69360334,
      "learning_rate": 3.9093436605213144e-06,
      "loss": 0.71696168,
      "num_input_tokens_seen": 21688500,
      "step": 1026,
      "time_per_iteration": 2.7517013549804688
    },
    {
      "auxiliary_loss_clip": 0.01343613,
      "auxiliary_loss_mlp": 0.01038462,
      "balance_loss_clip": 1.06474471,
      "balance_loss_mlp": 1.02866864,
      "epoch": 0.12348944868634643,
      "flos": 23878369797120.0,
      "grad_norm": 2.176266823838163,
      "language_loss": 0.79669642,
      "learning_rate": 3.909111646714627e-06,
      "loss": 0.82051718,
      "num_input_tokens_seen": 21709345,
      "step": 1027,
      "time_per_iteration": 2.797309160232544
    },
    {
      "auxiliary_loss_clip": 0.01240356,
      "auxiliary_loss_mlp": 0.01035325,
      "balance_loss_clip": 1.06516373,
      "balance_loss_mlp": 1.02622962,
      "epoch": 0.12360969157698551,
      "flos": 19026084314880.0,
      "grad_norm": 2.910540673638394,
      "language_loss": 0.72376823,
      "learning_rate": 3.9088793432972206e-06,
      "loss": 0.74652505,
      "num_input_tokens_seen": 21728165,
      "step": 1028,
      "time_per_iteration": 2.705413579940796
    },
    {
      "auxiliary_loss_clip": 0.01440337,
      "auxiliary_loss_mlp": 0.01037581,
      "balance_loss_clip": 1.06041324,
      "balance_loss_mlp": 1.02833629,
      "epoch": 0.1237299344676246,
      "flos": 13224607983360.0,
      "grad_norm": 2.4391487258827564,
      "language_loss": 0.8223623,
      "learning_rate": 3.908646750304336e-06,
      "loss": 0.8471415,
      "num_input_tokens_seen": 21745850,
      "step": 1029,
      "time_per_iteration": 3.6682639122009277
    },
    {
      "auxiliary_loss_clip": 0.01345953,
      "auxiliary_loss_mlp": 0.0103923,
      "balance_loss_clip": 1.06654406,
      "balance_loss_mlp": 1.02977097,
      "epoch": 0.12385017735826369,
      "flos": 20485673470080.0,
      "grad_norm": 1.7917288627512735,
      "language_loss": 0.87332952,
      "learning_rate": 3.908413867771257e-06,
      "loss": 0.89718133,
      "num_input_tokens_seen": 21764760,
      "step": 1030,
      "time_per_iteration": 2.7442803382873535
    },
    {
      "auxiliary_loss_clip": 0.01294487,
      "auxiliary_loss_mlp": 0.01041184,
      "balance_loss_clip": 1.06720817,
      "balance_loss_mlp": 1.03126585,
      "epoch": 0.12397042024890279,
      "flos": 17347835116800.0,
      "grad_norm": 6.32031896179411,
      "language_loss": 0.80504519,
      "learning_rate": 3.908180695733311e-06,
      "loss": 0.82840192,
      "num_input_tokens_seen": 21784250,
      "step": 1031,
      "time_per_iteration": 2.6223134994506836
    },
    {
      "auxiliary_loss_clip": 0.01421523,
      "auxiliary_loss_mlp": 0.01048768,
      "balance_loss_clip": 1.05300879,
      "balance_loss_mlp": 1.03929675,
      "epoch": 0.12409066313954187,
      "flos": 20412343854720.0,
      "grad_norm": 2.2884537232718083,
      "language_loss": 0.83121616,
      "learning_rate": 3.907947234225871e-06,
      "loss": 0.85591906,
      "num_input_tokens_seen": 21803260,
      "step": 1032,
      "time_per_iteration": 2.75187349319458
    },
    {
      "auxiliary_loss_clip": 0.01482051,
      "auxiliary_loss_mlp": 0.01035867,
      "balance_loss_clip": 1.05372047,
      "balance_loss_mlp": 1.02591896,
      "epoch": 0.12421090603018096,
      "flos": 20736688688640.0,
      "grad_norm": 1.8535359309129105,
      "language_loss": 0.87115133,
      "learning_rate": 3.907713483284352e-06,
      "loss": 0.8963306,
      "num_input_tokens_seen": 21822735,
      "step": 1033,
      "time_per_iteration": 4.691546440124512
    },
    {
      "auxiliary_loss_clip": 0.01537814,
      "auxiliary_loss_mlp": 0.0103973,
      "balance_loss_clip": 1.05278254,
      "balance_loss_mlp": 1.03002071,
      "epoch": 0.12433114892082006,
      "flos": 24498834353280.0,
      "grad_norm": 2.500585422991512,
      "language_loss": 0.97483373,
      "learning_rate": 3.907479442944216e-06,
      "loss": 1.00060916,
      "num_input_tokens_seen": 21841140,
      "step": 1034,
      "time_per_iteration": 3.0827810764312744
    },
    {
      "auxiliary_loss_clip": 0.01290684,
      "auxiliary_loss_mlp": 0.01039132,
      "balance_loss_clip": 1.0669347,
      "balance_loss_mlp": 1.02969122,
      "epoch": 0.12445139181145914,
      "flos": 19682315838720.0,
      "grad_norm": 4.553894732239057,
      "language_loss": 0.92392284,
      "learning_rate": 3.907245113240963e-06,
      "loss": 0.94722092,
      "num_input_tokens_seen": 21859260,
      "step": 1035,
      "time_per_iteration": 3.0423970222473145
    },
    {
      "auxiliary_loss_clip": 0.01387064,
      "auxiliary_loss_mlp": 0.01035412,
      "balance_loss_clip": 1.0558157,
      "balance_loss_mlp": 1.02555382,
      "epoch": 0.12457163470209824,
      "flos": 46423087522560.0,
      "grad_norm": 2.3557287102610713,
      "language_loss": 0.7379967,
      "learning_rate": 3.907010494210144e-06,
      "loss": 0.76222152,
      "num_input_tokens_seen": 21881920,
      "step": 1036,
      "time_per_iteration": 3.0001912117004395
    },
    {
      "auxiliary_loss_clip": 0.01297196,
      "auxiliary_loss_mlp": 0.01037616,
      "balance_loss_clip": 1.06637919,
      "balance_loss_mlp": 1.02653003,
      "epoch": 0.12469187759273732,
      "flos": 20376289578240.0,
      "grad_norm": 2.3300754326467414,
      "language_loss": 0.92199647,
      "learning_rate": 3.9067755858873495e-06,
      "loss": 0.94534463,
      "num_input_tokens_seen": 21898720,
      "step": 1037,
      "time_per_iteration": 3.639268398284912
    },
    {
      "auxiliary_loss_clip": 0.01306794,
      "auxiliary_loss_mlp": 0.01005386,
      "balance_loss_clip": 1.03875816,
      "balance_loss_mlp": 1.00314522,
      "epoch": 0.12481212048337642,
      "flos": 69224641447680.0,
      "grad_norm": 0.863598851473667,
      "language_loss": 0.62823081,
      "learning_rate": 3.906540388308214e-06,
      "loss": 0.65135264,
      "num_input_tokens_seen": 21958305,
      "step": 1038,
      "time_per_iteration": 3.353687286376953
    },
    {
      "auxiliary_loss_clip": 0.01429277,
      "auxiliary_loss_mlp": 0.01040911,
      "balance_loss_clip": 1.05987155,
      "balance_loss_mlp": 1.03143406,
      "epoch": 0.12493236337401552,
      "flos": 18223696350720.0,
      "grad_norm": 3.5622364525151813,
      "language_loss": 0.81696403,
      "learning_rate": 3.906304901508417e-06,
      "loss": 0.84166598,
      "num_input_tokens_seen": 21977205,
      "step": 1039,
      "time_per_iteration": 2.8104772567749023
    },
    {
      "auxiliary_loss_clip": 0.01299577,
      "auxiliary_loss_mlp": 0.01035889,
      "balance_loss_clip": 1.07145238,
      "balance_loss_mlp": 1.02619112,
      "epoch": 0.12505260626465461,
      "flos": 30044375303040.0,
      "grad_norm": 2.951532629988132,
      "language_loss": 0.75706816,
      "learning_rate": 3.9060691255236835e-06,
      "loss": 0.78042287,
      "num_input_tokens_seen": 21997770,
      "step": 1040,
      "time_per_iteration": 2.770315170288086
    },
    {
      "auxiliary_loss_clip": 0.01289806,
      "auxiliary_loss_mlp": 0.01041617,
      "balance_loss_clip": 1.06177926,
      "balance_loss_mlp": 1.03169298,
      "epoch": 0.1251728491552937,
      "flos": 24433980347520.0,
      "grad_norm": 2.426636070807813,
      "language_loss": 0.80838907,
      "learning_rate": 3.905833060389778e-06,
      "loss": 0.83170331,
      "num_input_tokens_seen": 22021890,
      "step": 1041,
      "time_per_iteration": 2.883533239364624
    },
    {
      "auxiliary_loss_clip": 0.0124667,
      "auxiliary_loss_mlp": 0.02577704,
      "balance_loss_clip": 1.06891298,
      "balance_loss_mlp": 1.00033617,
      "epoch": 0.12529309204593278,
      "flos": 27119809952640.0,
      "grad_norm": 4.732655078779728,
      "language_loss": 0.78474116,
      "learning_rate": 3.905596706142513e-06,
      "loss": 0.82298487,
      "num_input_tokens_seen": 22043300,
      "step": 1042,
      "time_per_iteration": 2.696477174758911
    },
    {
      "auxiliary_loss_clip": 0.01383292,
      "auxiliary_loss_mlp": 0.01037714,
      "balance_loss_clip": 1.05595922,
      "balance_loss_mlp": 1.02753377,
      "epoch": 0.12541333493657186,
      "flos": 30774151923840.0,
      "grad_norm": 1.835497403340833,
      "language_loss": 0.86225051,
      "learning_rate": 3.9053600628177435e-06,
      "loss": 0.88646054,
      "num_input_tokens_seen": 22062910,
      "step": 1043,
      "time_per_iteration": 2.8718841075897217
    },
    {
      "auxiliary_loss_clip": 0.01242723,
      "auxiliary_loss_mlp": 0.01036548,
      "balance_loss_clip": 1.06495392,
      "balance_loss_mlp": 1.02730894,
      "epoch": 0.12553357782721097,
      "flos": 23659566099840.0,
      "grad_norm": 2.477610630537506,
      "language_loss": 0.85024589,
      "learning_rate": 3.905123130451367e-06,
      "loss": 0.87303859,
      "num_input_tokens_seen": 22084010,
      "step": 1044,
      "time_per_iteration": 2.6648738384246826
    },
    {
      "auxiliary_loss_clip": 0.01245588,
      "auxiliary_loss_mlp": 0.01039003,
      "balance_loss_clip": 1.06789231,
      "balance_loss_mlp": 1.02896523,
      "epoch": 0.12565382071785006,
      "flos": 24863758577280.0,
      "grad_norm": 2.2705900016151537,
      "language_loss": 0.79511189,
      "learning_rate": 3.904885909079326e-06,
      "loss": 0.81795776,
      "num_input_tokens_seen": 22102795,
      "step": 1045,
      "time_per_iteration": 2.675853967666626
    },
    {
      "auxiliary_loss_clip": 0.01291028,
      "auxiliary_loss_mlp": 0.01031709,
      "balance_loss_clip": 1.0649246,
      "balance_loss_mlp": 1.02214789,
      "epoch": 0.12577406360848914,
      "flos": 21360780518400.0,
      "grad_norm": 2.7702429910888524,
      "language_loss": 0.78026116,
      "learning_rate": 3.904648398737607e-06,
      "loss": 0.80348861,
      "num_input_tokens_seen": 22121360,
      "step": 1046,
      "time_per_iteration": 2.7044944763183594
    },
    {
      "auxiliary_loss_clip": 0.01243404,
      "auxiliary_loss_mlp": 0.01036418,
      "balance_loss_clip": 1.06525397,
      "balance_loss_mlp": 1.02743542,
      "epoch": 0.12589430649912825,
      "flos": 36138056774400.0,
      "grad_norm": 3.119109639960784,
      "language_loss": 0.781142,
      "learning_rate": 3.9044105994622406e-06,
      "loss": 0.80394018,
      "num_input_tokens_seen": 22142505,
      "step": 1047,
      "time_per_iteration": 2.6837165355682373
    },
    {
      "auxiliary_loss_clip": 0.01312788,
      "auxiliary_loss_mlp": 0.02583297,
      "balance_loss_clip": 1.06260026,
      "balance_loss_mlp": 1.00042391,
      "epoch": 0.12601454938976733,
      "flos": 25337671643520.0,
      "grad_norm": 4.36570031952575,
      "language_loss": 0.81893015,
      "learning_rate": 3.9041725112893005e-06,
      "loss": 0.85789096,
      "num_input_tokens_seen": 22163730,
      "step": 1048,
      "time_per_iteration": 2.7359445095062256
    },
    {
      "auxiliary_loss_clip": 0.01385725,
      "auxiliary_loss_mlp": 0.0103136,
      "balance_loss_clip": 1.06263566,
      "balance_loss_mlp": 1.02148938,
      "epoch": 0.12613479228040642,
      "flos": 15560094286080.0,
      "grad_norm": 6.287642404789657,
      "language_loss": 0.75046349,
      "learning_rate": 3.903934134254904e-06,
      "loss": 0.77463436,
      "num_input_tokens_seen": 22181520,
      "step": 1049,
      "time_per_iteration": 2.6382410526275635
    },
    {
      "auxiliary_loss_clip": 0.01301169,
      "auxiliary_loss_mlp": 0.01046265,
      "balance_loss_clip": 1.06569946,
      "balance_loss_mlp": 1.03574431,
      "epoch": 0.1262550351710455,
      "flos": 21470595373440.0,
      "grad_norm": 3.299026546041682,
      "language_loss": 0.84800333,
      "learning_rate": 3.903695468395213e-06,
      "loss": 0.8714776,
      "num_input_tokens_seen": 22199390,
      "step": 1050,
      "time_per_iteration": 2.672031879425049
    },
    {
      "auxiliary_loss_clip": 0.01348669,
      "auxiliary_loss_mlp": 0.01037689,
      "balance_loss_clip": 1.06135941,
      "balance_loss_mlp": 1.02748513,
      "epoch": 0.1263752780616846,
      "flos": 31576719456000.0,
      "grad_norm": 5.15939727483319,
      "language_loss": 0.55985272,
      "learning_rate": 3.903456513746434e-06,
      "loss": 0.58371627,
      "num_input_tokens_seen": 22220365,
      "step": 1051,
      "time_per_iteration": 2.756950855255127
    },
    {
      "auxiliary_loss_clip": 0.01242176,
      "auxiliary_loss_mlp": 0.01033624,
      "balance_loss_clip": 1.066082,
      "balance_loss_mlp": 1.02480888,
      "epoch": 0.1264955209523237,
      "flos": 28768217927040.0,
      "grad_norm": 1.85684324983257,
      "language_loss": 0.87579155,
      "learning_rate": 3.903217270344815e-06,
      "loss": 0.8985495,
      "num_input_tokens_seen": 22240615,
      "step": 1052,
      "time_per_iteration": 2.691006660461426
    },
    {
      "auxiliary_loss_clip": 0.01381825,
      "auxiliary_loss_mlp": 0.0103412,
      "balance_loss_clip": 1.05729461,
      "balance_loss_mlp": 1.02449942,
      "epoch": 0.12661576384296278,
      "flos": 29241125412480.0,
      "grad_norm": 2.0501026122530264,
      "language_loss": 0.82653546,
      "learning_rate": 3.902977738226648e-06,
      "loss": 0.85069495,
      "num_input_tokens_seen": 22261350,
      "step": 1053,
      "time_per_iteration": 2.772590398788452
    },
    {
      "auxiliary_loss_clip": 0.01295569,
      "auxiliary_loss_mlp": 0.01042874,
      "balance_loss_clip": 1.06772745,
      "balance_loss_mlp": 1.03218114,
      "epoch": 0.12673600673360189,
      "flos": 20850346298880.0,
      "grad_norm": 1.8253796299615879,
      "language_loss": 0.90757728,
      "learning_rate": 3.902737917428273e-06,
      "loss": 0.93096161,
      "num_input_tokens_seen": 22279515,
      "step": 1054,
      "time_per_iteration": 2.724503993988037
    },
    {
      "auxiliary_loss_clip": 0.01245023,
      "auxiliary_loss_mlp": 0.0103661,
      "balance_loss_clip": 1.06803191,
      "balance_loss_mlp": 1.02710342,
      "epoch": 0.12685624962424097,
      "flos": 25263695583360.0,
      "grad_norm": 1.833856767039498,
      "language_loss": 0.84042144,
      "learning_rate": 3.902497807986068e-06,
      "loss": 0.86323774,
      "num_input_tokens_seen": 22299535,
      "step": 1055,
      "time_per_iteration": 3.622091770172119
    },
    {
      "auxiliary_loss_clip": 0.01393145,
      "auxiliary_loss_mlp": 0.01034182,
      "balance_loss_clip": 1.05693316,
      "balance_loss_mlp": 1.0242517,
      "epoch": 0.12697649251488005,
      "flos": 27527109246720.0,
      "grad_norm": 1.6805193593399024,
      "language_loss": 0.84150159,
      "learning_rate": 3.902257409936458e-06,
      "loss": 0.86577487,
      "num_input_tokens_seen": 22320300,
      "step": 1056,
      "time_per_iteration": 2.7717368602752686
    },
    {
      "auxiliary_loss_clip": 0.01346749,
      "auxiliary_loss_mlp": 0.01037126,
      "balance_loss_clip": 1.06822801,
      "balance_loss_mlp": 1.02807844,
      "epoch": 0.12709673540551916,
      "flos": 21251863503360.0,
      "grad_norm": 3.7135101945624873,
      "language_loss": 0.83820093,
      "learning_rate": 3.902016723315912e-06,
      "loss": 0.86203969,
      "num_input_tokens_seen": 22338240,
      "step": 1057,
      "time_per_iteration": 2.7000856399536133
    },
    {
      "auxiliary_loss_clip": 0.01292485,
      "auxiliary_loss_mlp": 0.01040398,
      "balance_loss_clip": 1.06305468,
      "balance_loss_mlp": 1.03108215,
      "epoch": 0.12721697829615825,
      "flos": 25337707557120.0,
      "grad_norm": 4.111478008468384,
      "language_loss": 0.69559968,
      "learning_rate": 3.901775748160941e-06,
      "loss": 0.71892852,
      "num_input_tokens_seen": 22357420,
      "step": 1058,
      "time_per_iteration": 2.6608502864837646
    },
    {
      "auxiliary_loss_clip": 0.01247511,
      "auxiliary_loss_mlp": 0.01007918,
      "balance_loss_clip": 1.04084027,
      "balance_loss_mlp": 1.0056529,
      "epoch": 0.12733722118679733,
      "flos": 61943287754880.0,
      "grad_norm": 0.8025903075024308,
      "language_loss": 0.60786015,
      "learning_rate": 3.901534484508101e-06,
      "loss": 0.63041455,
      "num_input_tokens_seen": 22420095,
      "step": 1059,
      "time_per_iteration": 4.565524101257324
    },
    {
      "auxiliary_loss_clip": 0.01333275,
      "auxiliary_loss_mlp": 0.01036571,
      "balance_loss_clip": 1.05886221,
      "balance_loss_mlp": 1.02633119,
      "epoch": 0.1274574640774364,
      "flos": 26976742081920.0,
      "grad_norm": 1.9943701327762455,
      "language_loss": 0.74765861,
      "learning_rate": 3.901292932393991e-06,
      "loss": 0.77135706,
      "num_input_tokens_seen": 22438975,
      "step": 1060,
      "time_per_iteration": 2.795952081680298
    },
    {
      "auxiliary_loss_clip": 0.01249728,
      "auxiliary_loss_mlp": 0.01038241,
      "balance_loss_clip": 1.06968594,
      "balance_loss_mlp": 1.02874017,
      "epoch": 0.12757770696807552,
      "flos": 22236318529920.0,
      "grad_norm": 2.181677906686555,
      "language_loss": 0.8502875,
      "learning_rate": 3.9010510918552555e-06,
      "loss": 0.87316716,
      "num_input_tokens_seen": 22458050,
      "step": 1061,
      "time_per_iteration": 2.6524336338043213
    },
    {
      "auxiliary_loss_clip": 0.01344603,
      "auxiliary_loss_mlp": 0.01039221,
      "balance_loss_clip": 1.06109703,
      "balance_loss_mlp": 1.02865338,
      "epoch": 0.1276979498587146,
      "flos": 28547905858560.0,
      "grad_norm": 2.508868136373472,
      "language_loss": 0.74464768,
      "learning_rate": 3.900808962928581e-06,
      "loss": 0.7684859,
      "num_input_tokens_seen": 22475665,
      "step": 1062,
      "time_per_iteration": 2.764497995376587
    },
    {
      "auxiliary_loss_clip": 0.01244101,
      "auxiliary_loss_mlp": 0.01032045,
      "balance_loss_clip": 1.06895685,
      "balance_loss_mlp": 1.02283931,
      "epoch": 0.1278181927493537,
      "flos": 17420338719360.0,
      "grad_norm": 2.2708282896771976,
      "language_loss": 0.8938669,
      "learning_rate": 3.900566545650698e-06,
      "loss": 0.91662842,
      "num_input_tokens_seen": 22493335,
      "step": 1063,
      "time_per_iteration": 3.578157901763916
    },
    {
      "auxiliary_loss_clip": 0.01293241,
      "auxiliary_loss_mlp": 0.01044841,
      "balance_loss_clip": 1.06752348,
      "balance_loss_mlp": 1.03528619,
      "epoch": 0.1279384356399928,
      "flos": 21138636856320.0,
      "grad_norm": 2.341749103151016,
      "language_loss": 0.81419802,
      "learning_rate": 3.900323840058381e-06,
      "loss": 0.83757889,
      "num_input_tokens_seen": 22511045,
      "step": 1064,
      "time_per_iteration": 2.7035741806030273
    },
    {
      "auxiliary_loss_clip": 0.01291601,
      "auxiliary_loss_mlp": 0.01037516,
      "balance_loss_clip": 1.06175828,
      "balance_loss_mlp": 1.02855158,
      "epoch": 0.12805867853063188,
      "flos": 26576733248640.0,
      "grad_norm": 2.759902771815658,
      "language_loss": 0.81919968,
      "learning_rate": 3.900080846188449e-06,
      "loss": 0.84249091,
      "num_input_tokens_seen": 22529635,
      "step": 1065,
      "time_per_iteration": 2.765500068664551
    },
    {
      "auxiliary_loss_clip": 0.01245315,
      "auxiliary_loss_mlp": 0.01037051,
      "balance_loss_clip": 1.06710136,
      "balance_loss_mlp": 1.02746022,
      "epoch": 0.12817892142127096,
      "flos": 16436206915200.0,
      "grad_norm": 1.9865444586121985,
      "language_loss": 0.81153405,
      "learning_rate": 3.8998375640777625e-06,
      "loss": 0.83435774,
      "num_input_tokens_seen": 22547505,
      "step": 1066,
      "time_per_iteration": 2.6787545680999756
    },
    {
      "auxiliary_loss_clip": 0.01243247,
      "auxiliary_loss_mlp": 0.01016915,
      "balance_loss_clip": 1.03191364,
      "balance_loss_mlp": 1.01469827,
      "epoch": 0.12829916431191005,
      "flos": 60757049099520.0,
      "grad_norm": 0.706508092372491,
      "language_loss": 0.52649945,
      "learning_rate": 3.899593993763229e-06,
      "loss": 0.54910111,
      "num_input_tokens_seen": 22608465,
      "step": 1067,
      "time_per_iteration": 3.256310224533081
    },
    {
      "auxiliary_loss_clip": 0.01386423,
      "auxiliary_loss_mlp": 0.01038669,
      "balance_loss_clip": 1.06077981,
      "balance_loss_mlp": 1.02845263,
      "epoch": 0.12841940720254916,
      "flos": 29786895636480.0,
      "grad_norm": 2.264765518305721,
      "language_loss": 0.81282687,
      "learning_rate": 3.899350135281796e-06,
      "loss": 0.8370778,
      "num_input_tokens_seen": 22629465,
      "step": 1068,
      "time_per_iteration": 2.8394534587860107
    },
    {
      "auxiliary_loss_clip": 0.01390352,
      "auxiliary_loss_mlp": 0.01041419,
      "balance_loss_clip": 1.06016493,
      "balance_loss_mlp": 1.03188825,
      "epoch": 0.12853965009318824,
      "flos": 25951851319680.0,
      "grad_norm": 2.694704887766841,
      "language_loss": 0.79806817,
      "learning_rate": 3.8991059886704585e-06,
      "loss": 0.82238591,
      "num_input_tokens_seen": 22648970,
      "step": 1069,
      "time_per_iteration": 2.803236246109009
    },
    {
      "auxiliary_loss_clip": 0.01384311,
      "auxiliary_loss_mlp": 0.0104697,
      "balance_loss_clip": 1.0608021,
      "balance_loss_mlp": 1.03774309,
      "epoch": 0.12865989298382732,
      "flos": 30846871008000.0,
      "grad_norm": 2.0358647036535307,
      "language_loss": 0.83353686,
      "learning_rate": 3.898861553966252e-06,
      "loss": 0.85784972,
      "num_input_tokens_seen": 22668620,
      "step": 1070,
      "time_per_iteration": 2.8915727138519287
    },
    {
      "auxiliary_loss_clip": 0.01538432,
      "auxiliary_loss_mlp": 0.01041515,
      "balance_loss_clip": 1.05210805,
      "balance_loss_mlp": 1.03181744,
      "epoch": 0.12878013587446643,
      "flos": 25885776251520.0,
      "grad_norm": 2.3744056053138234,
      "language_loss": 0.88227868,
      "learning_rate": 3.898616831206257e-06,
      "loss": 0.90807819,
      "num_input_tokens_seen": 22689045,
      "step": 1071,
      "time_per_iteration": 3.0063436031341553
    },
    {
      "auxiliary_loss_clip": 0.01386603,
      "auxiliary_loss_mlp": 0.01035663,
      "balance_loss_clip": 1.05653954,
      "balance_loss_mlp": 1.02469587,
      "epoch": 0.12890037876510552,
      "flos": 23333138277120.0,
      "grad_norm": 2.226549465521729,
      "language_loss": 0.76570749,
      "learning_rate": 3.8983718204276e-06,
      "loss": 0.78993011,
      "num_input_tokens_seen": 22711265,
      "step": 1072,
      "time_per_iteration": 3.017740249633789
    },
    {
      "auxiliary_loss_clip": 0.01339223,
      "auxiliary_loss_mlp": 0.01032606,
      "balance_loss_clip": 1.06218982,
      "balance_loss_mlp": 1.02377903,
      "epoch": 0.1290206216557446,
      "flos": 23587242065280.0,
      "grad_norm": 1.7555074247924802,
      "language_loss": 0.82716227,
      "learning_rate": 3.898126521667446e-06,
      "loss": 0.8508805,
      "num_input_tokens_seen": 22731420,
      "step": 1073,
      "time_per_iteration": 2.7249915599823
    },
    {
      "auxiliary_loss_clip": 0.01290552,
      "auxiliary_loss_mlp": 0.01040039,
      "balance_loss_clip": 1.0618577,
      "balance_loss_mlp": 1.03010917,
      "epoch": 0.12914086454638368,
      "flos": 24170610850560.0,
      "grad_norm": 1.7604790071450669,
      "language_loss": 0.83473051,
      "learning_rate": 3.897880934963007e-06,
      "loss": 0.8580364,
      "num_input_tokens_seen": 22750970,
      "step": 1074,
      "time_per_iteration": 2.685483694076538
    },
    {
      "auxiliary_loss_clip": 0.01341828,
      "auxiliary_loss_mlp": 0.01037824,
      "balance_loss_clip": 1.06099582,
      "balance_loss_mlp": 1.02768552,
      "epoch": 0.1292611074370228,
      "flos": 20267157081600.0,
      "grad_norm": 2.491623084497036,
      "language_loss": 0.7849015,
      "learning_rate": 3.89763506035154e-06,
      "loss": 0.80869806,
      "num_input_tokens_seen": 22768820,
      "step": 1075,
      "time_per_iteration": 2.670273780822754
    },
    {
      "auxiliary_loss_clip": 0.01279084,
      "auxiliary_loss_mlp": 0.01041654,
      "balance_loss_clip": 1.05962682,
      "balance_loss_mlp": 1.03246915,
      "epoch": 0.12938135032766188,
      "flos": 27377684668800.0,
      "grad_norm": 1.7727017415757433,
      "language_loss": 0.8121444,
      "learning_rate": 3.897388897870343e-06,
      "loss": 0.83535182,
      "num_input_tokens_seen": 22789460,
      "step": 1076,
      "time_per_iteration": 2.716249465942383
    },
    {
      "auxiliary_loss_clip": 0.0135028,
      "auxiliary_loss_mlp": 0.01037792,
      "balance_loss_clip": 1.06110168,
      "balance_loss_mlp": 1.02727795,
      "epoch": 0.12950159321830096,
      "flos": 29277107861760.0,
      "grad_norm": 2.3028365028123305,
      "language_loss": 0.7471751,
      "learning_rate": 3.89714244755676e-06,
      "loss": 0.77105582,
      "num_input_tokens_seen": 22810820,
      "step": 1077,
      "time_per_iteration": 2.8057966232299805
    },
    {
      "auxiliary_loss_clip": 0.01397167,
      "auxiliary_loss_mlp": 0.01040059,
      "balance_loss_clip": 1.05386829,
      "balance_loss_mlp": 1.03000951,
      "epoch": 0.12962183610894007,
      "flos": 24534888629760.0,
      "grad_norm": 2.683895381081054,
      "language_loss": 0.86059237,
      "learning_rate": 3.896895709448175e-06,
      "loss": 0.88496464,
      "num_input_tokens_seen": 22830570,
      "step": 1078,
      "time_per_iteration": 2.7501142024993896
    },
    {
      "auxiliary_loss_clip": 0.01480562,
      "auxiliary_loss_mlp": 0.01040894,
      "balance_loss_clip": 1.05170631,
      "balance_loss_mlp": 1.03092206,
      "epoch": 0.12974207899957915,
      "flos": 11215944552960.0,
      "grad_norm": 3.63666254667062,
      "language_loss": 0.77384937,
      "learning_rate": 3.896648683582019e-06,
      "loss": 0.79906392,
      "num_input_tokens_seen": 22845905,
      "step": 1079,
      "time_per_iteration": 2.824110507965088
    },
    {
      "auxiliary_loss_clip": 0.0143582,
      "auxiliary_loss_mlp": 0.01034323,
      "balance_loss_clip": 1.06185532,
      "balance_loss_mlp": 1.02476835,
      "epoch": 0.12986232189021824,
      "flos": 24717889445760.0,
      "grad_norm": 2.018688304957884,
      "language_loss": 0.80809033,
      "learning_rate": 3.896401369995766e-06,
      "loss": 0.83279181,
      "num_input_tokens_seen": 22865710,
      "step": 1080,
      "time_per_iteration": 2.866117238998413
    },
    {
      "auxiliary_loss_clip": 0.01244394,
      "auxiliary_loss_mlp": 0.01030486,
      "balance_loss_clip": 1.06917,
      "balance_loss_mlp": 1.02111578,
      "epoch": 0.12998256478085732,
      "flos": 23915357827200.0,
      "grad_norm": 3.6439006008602894,
      "language_loss": 0.79551256,
      "learning_rate": 3.896153768726932e-06,
      "loss": 0.81826138,
      "num_input_tokens_seen": 22886020,
      "step": 1081,
      "time_per_iteration": 3.6568238735198975
    },
    {
      "auxiliary_loss_clip": 0.0129366,
      "auxiliary_loss_mlp": 0.01052043,
      "balance_loss_clip": 1.06803632,
      "balance_loss_mlp": 1.0422914,
      "epoch": 0.13010280767149643,
      "flos": 18624207974400.0,
      "grad_norm": 2.6892359351592074,
      "language_loss": 0.87502116,
      "learning_rate": 3.8959058798130806e-06,
      "loss": 0.89847815,
      "num_input_tokens_seen": 22903995,
      "step": 1082,
      "time_per_iteration": 2.648695230484009
    },
    {
      "auxiliary_loss_clip": 0.01344599,
      "auxiliary_loss_mlp": 0.02582366,
      "balance_loss_clip": 1.06214261,
      "balance_loss_mlp": 1.00052166,
      "epoch": 0.1302230505621355,
      "flos": 22783992174720.0,
      "grad_norm": 2.0663287598758355,
      "language_loss": 0.75120175,
      "learning_rate": 3.895657703291814e-06,
      "loss": 0.79047143,
      "num_input_tokens_seen": 22924100,
      "step": 1083,
      "time_per_iteration": 2.7483255863189697
    },
    {
      "auxiliary_loss_clip": 0.01349112,
      "auxiliary_loss_mlp": 0.01037536,
      "balance_loss_clip": 1.06229353,
      "balance_loss_mlp": 1.02684295,
      "epoch": 0.1303432934527746,
      "flos": 21323612920320.0,
      "grad_norm": 4.7239555414760135,
      "language_loss": 0.80034864,
      "learning_rate": 3.895409239200781e-06,
      "loss": 0.82421505,
      "num_input_tokens_seen": 22939985,
      "step": 1084,
      "time_per_iteration": 2.767587661743164
    },
    {
      "auxiliary_loss_clip": 0.012871,
      "auxiliary_loss_mlp": 0.0104056,
      "balance_loss_clip": 1.06101418,
      "balance_loss_mlp": 1.03051686,
      "epoch": 0.1304635363434137,
      "flos": 20922490765440.0,
      "grad_norm": 2.3488092712653548,
      "language_loss": 0.90904444,
      "learning_rate": 3.895160487577673e-06,
      "loss": 0.93232107,
      "num_input_tokens_seen": 22957555,
      "step": 1085,
      "time_per_iteration": 4.5021679401397705
    },
    {
      "auxiliary_loss_clip": 0.01197107,
      "auxiliary_loss_mlp": 0.01002446,
      "balance_loss_clip": 1.04086483,
      "balance_loss_mlp": 1.00032449,
      "epoch": 0.1305837792340528,
      "flos": 63245659080960.0,
      "grad_norm": 0.7868703211544962,
      "language_loss": 0.60932207,
      "learning_rate": 3.894911448460226e-06,
      "loss": 0.63131762,
      "num_input_tokens_seen": 23016870,
      "step": 1086,
      "time_per_iteration": 3.1641018390655518
    },
    {
      "auxiliary_loss_clip": 0.01517629,
      "auxiliary_loss_mlp": 0.01038907,
      "balance_loss_clip": 1.05172706,
      "balance_loss_mlp": 1.0284946,
      "epoch": 0.13070402212469187,
      "flos": 26428852955520.0,
      "grad_norm": 1.9519395581814911,
      "language_loss": 0.73089933,
      "learning_rate": 3.8946621218862195e-06,
      "loss": 0.75646472,
      "num_input_tokens_seen": 23037870,
      "step": 1087,
      "time_per_iteration": 3.0084848403930664
    },
    {
      "auxiliary_loss_clip": 0.01388439,
      "auxiliary_loss_mlp": 0.01032002,
      "balance_loss_clip": 1.06041622,
      "balance_loss_mlp": 1.02241194,
      "epoch": 0.13082426501533098,
      "flos": 27673409341440.0,
      "grad_norm": 2.2181223464362514,
      "language_loss": 0.89003354,
      "learning_rate": 3.894412507893475e-06,
      "loss": 0.91423792,
      "num_input_tokens_seen": 23058150,
      "step": 1088,
      "time_per_iteration": 2.9290642738342285
    },
    {
      "auxiliary_loss_clip": 0.01442545,
      "auxiliary_loss_mlp": 0.01042855,
      "balance_loss_clip": 1.05853152,
      "balance_loss_mlp": 1.03234053,
      "epoch": 0.13094450790597006,
      "flos": 24826770547200.0,
      "grad_norm": 2.0665356349500223,
      "language_loss": 0.72040093,
      "learning_rate": 3.894162606519859e-06,
      "loss": 0.74525487,
      "num_input_tokens_seen": 23077100,
      "step": 1089,
      "time_per_iteration": 3.6544559001922607
    },
    {
      "auxiliary_loss_clip": 0.01437543,
      "auxiliary_loss_mlp": 0.01046246,
      "balance_loss_clip": 1.06011796,
      "balance_loss_mlp": 1.0356065,
      "epoch": 0.13106475079660915,
      "flos": 19062605468160.0,
      "grad_norm": 3.078926725040933,
      "language_loss": 0.77237177,
      "learning_rate": 3.893912417803282e-06,
      "loss": 0.79720962,
      "num_input_tokens_seen": 23096815,
      "step": 1090,
      "time_per_iteration": 2.8977482318878174
    },
    {
      "auxiliary_loss_clip": 0.01407611,
      "auxiliary_loss_mlp": 0.01041664,
      "balance_loss_clip": 1.05550361,
      "balance_loss_mlp": 1.03141212,
      "epoch": 0.13118499368724823,
      "flos": 28913189218560.0,
      "grad_norm": 2.292341275730407,
      "language_loss": 0.77185667,
      "learning_rate": 3.8936619417816975e-06,
      "loss": 0.79634947,
      "num_input_tokens_seen": 23117145,
      "step": 1091,
      "time_per_iteration": 2.858304023742676
    },
    {
      "auxiliary_loss_clip": 0.01389388,
      "auxiliary_loss_mlp": 0.01039521,
      "balance_loss_clip": 1.06498075,
      "balance_loss_mlp": 1.02975202,
      "epoch": 0.13130523657788734,
      "flos": 14283398206080.0,
      "grad_norm": 1.8584823688509549,
      "language_loss": 0.7175324,
      "learning_rate": 3.8934111784931015e-06,
      "loss": 0.74182153,
      "num_input_tokens_seen": 23134595,
      "step": 1092,
      "time_per_iteration": 3.019113779067993
    },
    {
      "auxiliary_loss_clip": 0.01252166,
      "auxiliary_loss_mlp": 0.0100253,
      "balance_loss_clip": 1.04038727,
      "balance_loss_mlp": 1.00036037,
      "epoch": 0.13142547946852642,
      "flos": 70174155519360.0,
      "grad_norm": 1.0256689483183399,
      "language_loss": 0.59036577,
      "learning_rate": 3.893160127975535e-06,
      "loss": 0.61291277,
      "num_input_tokens_seen": 23195285,
      "step": 1093,
      "time_per_iteration": 3.4826316833496094
    },
    {
      "auxiliary_loss_clip": 0.01447332,
      "auxiliary_loss_mlp": 0.01034904,
      "balance_loss_clip": 1.06019521,
      "balance_loss_mlp": 1.02484894,
      "epoch": 0.1315457223591655,
      "flos": 45805998844800.0,
      "grad_norm": 2.9717537657156368,
      "language_loss": 0.8137148,
      "learning_rate": 3.8929087902670826e-06,
      "loss": 0.83853722,
      "num_input_tokens_seen": 23216915,
      "step": 1094,
      "time_per_iteration": 3.0307607650756836
    },
    {
      "auxiliary_loss_clip": 0.01200045,
      "auxiliary_loss_mlp": 0.01002839,
      "balance_loss_clip": 1.03841281,
      "balance_loss_mlp": 1.00075293,
      "epoch": 0.13166596524980462,
      "flos": 62881165820160.0,
      "grad_norm": 0.917975844055761,
      "language_loss": 0.60622758,
      "learning_rate": 3.8926571654058715e-06,
      "loss": 0.62825638,
      "num_input_tokens_seen": 23273560,
      "step": 1095,
      "time_per_iteration": 3.2036795616149902
    },
    {
      "auxiliary_loss_clip": 0.01387298,
      "auxiliary_loss_mlp": 0.0103693,
      "balance_loss_clip": 1.06277812,
      "balance_loss_mlp": 1.02709484,
      "epoch": 0.1317862081404437,
      "flos": 23586523793280.0,
      "grad_norm": 2.720720070806504,
      "language_loss": 0.77518898,
      "learning_rate": 3.892405253430074e-06,
      "loss": 0.79943132,
      "num_input_tokens_seen": 23291080,
      "step": 1096,
      "time_per_iteration": 2.7363550662994385
    },
    {
      "auxiliary_loss_clip": 0.01346619,
      "auxiliary_loss_mlp": 0.02581243,
      "balance_loss_clip": 1.06531835,
      "balance_loss_mlp": 1.00053859,
      "epoch": 0.13190645103108278,
      "flos": 20260764460800.0,
      "grad_norm": 1.9115761244534688,
      "language_loss": 0.82445014,
      "learning_rate": 3.892153054377904e-06,
      "loss": 0.86372876,
      "num_input_tokens_seen": 23308485,
      "step": 1097,
      "time_per_iteration": 2.73370623588562
    },
    {
      "auxiliary_loss_clip": 0.01395237,
      "auxiliary_loss_mlp": 0.01008192,
      "balance_loss_clip": 1.03317952,
      "balance_loss_mlp": 1.00578403,
      "epoch": 0.13202669392172187,
      "flos": 53455440136320.0,
      "grad_norm": 0.9306993470695897,
      "language_loss": 0.59371156,
      "learning_rate": 3.891900568287619e-06,
      "loss": 0.61774588,
      "num_input_tokens_seen": 23360870,
      "step": 1098,
      "time_per_iteration": 3.2817232608795166
    },
    {
      "auxiliary_loss_clip": 0.01362348,
      "auxiliary_loss_mlp": 0.01044328,
      "balance_loss_clip": 1.0637722,
      "balance_loss_mlp": 1.03246641,
      "epoch": 0.13214693681236098,
      "flos": 15851293845120.0,
      "grad_norm": 2.6858614546150945,
      "language_loss": 0.72370529,
      "learning_rate": 3.891647795197523e-06,
      "loss": 0.7477721,
      "num_input_tokens_seen": 23376910,
      "step": 1099,
      "time_per_iteration": 2.949690103530884
    },
    {
      "auxiliary_loss_clip": 0.01395689,
      "auxiliary_loss_mlp": 0.01039604,
      "balance_loss_clip": 1.05908823,
      "balance_loss_mlp": 1.02773058,
      "epoch": 0.13226717970300006,
      "flos": 19353840940800.0,
      "grad_norm": 2.558989441374826,
      "language_loss": 0.68553627,
      "learning_rate": 3.8913947351459605e-06,
      "loss": 0.70988917,
      "num_input_tokens_seen": 23394450,
      "step": 1100,
      "time_per_iteration": 2.802950143814087
    },
    {
      "auxiliary_loss_clip": 0.01241028,
      "auxiliary_loss_mlp": 0.01044203,
      "balance_loss_clip": 1.06605172,
      "balance_loss_mlp": 1.03517866,
      "epoch": 0.13238742259363914,
      "flos": 20698084546560.0,
      "grad_norm": 2.174380955770455,
      "language_loss": 0.67966318,
      "learning_rate": 3.89114138817132e-06,
      "loss": 0.70251554,
      "num_input_tokens_seen": 23411115,
      "step": 1101,
      "time_per_iteration": 2.695899486541748
    },
    {
      "auxiliary_loss_clip": 0.01290253,
      "auxiliary_loss_mlp": 0.01032284,
      "balance_loss_clip": 1.06748486,
      "balance_loss_mlp": 1.02367115,
      "epoch": 0.13250766548427825,
      "flos": 21032449274880.0,
      "grad_norm": 1.899023120367045,
      "language_loss": 0.84300721,
      "learning_rate": 3.890887754312035e-06,
      "loss": 0.86623257,
      "num_input_tokens_seen": 23429360,
      "step": 1102,
      "time_per_iteration": 2.718205690383911
    },
    {
      "auxiliary_loss_clip": 0.0133688,
      "auxiliary_loss_mlp": 0.01032485,
      "balance_loss_clip": 1.06145787,
      "balance_loss_mlp": 1.02266788,
      "epoch": 0.13262790837491734,
      "flos": 22637871648000.0,
      "grad_norm": 2.9351202053739063,
      "language_loss": 0.87825203,
      "learning_rate": 3.890633833606581e-06,
      "loss": 0.90194571,
      "num_input_tokens_seen": 23449050,
      "step": 1103,
      "time_per_iteration": 2.6689302921295166
    },
    {
      "auxiliary_loss_clip": 0.01292518,
      "auxiliary_loss_mlp": 0.01041183,
      "balance_loss_clip": 1.07025146,
      "balance_loss_mlp": 1.03108621,
      "epoch": 0.13274815126555642,
      "flos": 19683141851520.0,
      "grad_norm": 2.117535459900211,
      "language_loss": 0.69396079,
      "learning_rate": 3.890379626093477e-06,
      "loss": 0.71729779,
      "num_input_tokens_seen": 23468800,
      "step": 1104,
      "time_per_iteration": 2.6978399753570557
    },
    {
      "auxiliary_loss_clip": 0.01429205,
      "auxiliary_loss_mlp": 0.01038335,
      "balance_loss_clip": 1.05842233,
      "balance_loss_mlp": 1.02817857,
      "epoch": 0.1328683941561955,
      "flos": 21317687176320.0,
      "grad_norm": 2.0546260101673686,
      "language_loss": 0.92404813,
      "learning_rate": 3.890125131811287e-06,
      "loss": 0.94872355,
      "num_input_tokens_seen": 23486850,
      "step": 1105,
      "time_per_iteration": 2.7551114559173584
    },
    {
      "auxiliary_loss_clip": 0.01329829,
      "auxiliary_loss_mlp": 0.01037718,
      "balance_loss_clip": 1.05689156,
      "balance_loss_mlp": 1.02849102,
      "epoch": 0.1329886370468346,
      "flos": 13699131580800.0,
      "grad_norm": 3.321306453692712,
      "language_loss": 0.75449508,
      "learning_rate": 3.889870350798618e-06,
      "loss": 0.77817053,
      "num_input_tokens_seen": 23504195,
      "step": 1106,
      "time_per_iteration": 2.751676321029663
    },
    {
      "auxiliary_loss_clip": 0.01242234,
      "auxiliary_loss_mlp": 0.01036722,
      "balance_loss_clip": 1.06645429,
      "balance_loss_mlp": 1.0270009,
      "epoch": 0.1331088799374737,
      "flos": 21032413361280.0,
      "grad_norm": 2.4947137202395644,
      "language_loss": 0.7866652,
      "learning_rate": 3.889615283094119e-06,
      "loss": 0.80945474,
      "num_input_tokens_seen": 23523385,
      "step": 1107,
      "time_per_iteration": 3.630387306213379
    },
    {
      "auxiliary_loss_clip": 0.012478,
      "auxiliary_loss_mlp": 0.01040493,
      "balance_loss_clip": 1.06724739,
      "balance_loss_mlp": 1.02953792,
      "epoch": 0.13322912282811278,
      "flos": 18260432985600.0,
      "grad_norm": 2.726398752553281,
      "language_loss": 0.84623492,
      "learning_rate": 3.889359928736485e-06,
      "loss": 0.86911786,
      "num_input_tokens_seen": 23541330,
      "step": 1108,
      "time_per_iteration": 2.5981311798095703
    },
    {
      "auxiliary_loss_clip": 0.01338555,
      "auxiliary_loss_mlp": 0.02578688,
      "balance_loss_clip": 1.06523967,
      "balance_loss_mlp": 1.00053966,
      "epoch": 0.1333493657187519,
      "flos": 24460876656000.0,
      "grad_norm": 3.4998102597324343,
      "language_loss": 0.91094184,
      "learning_rate": 3.889104287764451e-06,
      "loss": 0.95011425,
      "num_input_tokens_seen": 23561705,
      "step": 1109,
      "time_per_iteration": 2.7944018840789795
    },
    {
      "auxiliary_loss_clip": 0.01334355,
      "auxiliary_loss_mlp": 0.0104451,
      "balance_loss_clip": 1.06362414,
      "balance_loss_mlp": 1.03456187,
      "epoch": 0.13346960860939097,
      "flos": 22158930677760.0,
      "grad_norm": 3.276376705547758,
      "language_loss": 0.90604055,
      "learning_rate": 3.888848360216798e-06,
      "loss": 0.92982924,
      "num_input_tokens_seen": 23579350,
      "step": 1110,
      "time_per_iteration": 3.7408578395843506
    },
    {
      "auxiliary_loss_clip": 0.01263262,
      "auxiliary_loss_mlp": 0.01005174,
      "balance_loss_clip": 1.04512239,
      "balance_loss_mlp": 1.00290871,
      "epoch": 0.13358985150003005,
      "flos": 67931212608000.0,
      "grad_norm": 0.7963980704956061,
      "language_loss": 0.56586409,
      "learning_rate": 3.888592146132351e-06,
      "loss": 0.58854842,
      "num_input_tokens_seen": 23640620,
      "step": 1111,
      "time_per_iteration": 4.333545207977295
    },
    {
      "auxiliary_loss_clip": 0.01294156,
      "auxiliary_loss_mlp": 0.01034811,
      "balance_loss_clip": 1.06857014,
      "balance_loss_mlp": 1.0257448,
      "epoch": 0.13371009439066917,
      "flos": 26834284742400.0,
      "grad_norm": 1.8954746084083058,
      "language_loss": 0.78733325,
      "learning_rate": 3.888335645549978e-06,
      "loss": 0.81062299,
      "num_input_tokens_seen": 23661040,
      "step": 1112,
      "time_per_iteration": 2.7247891426086426
    },
    {
      "auxiliary_loss_clip": 0.0124572,
      "auxiliary_loss_mlp": 0.01047801,
      "balance_loss_clip": 1.06932831,
      "balance_loss_mlp": 1.03893173,
      "epoch": 0.13383033728130825,
      "flos": 26322844942080.0,
      "grad_norm": 32.738588449575495,
      "language_loss": 0.81451762,
      "learning_rate": 3.888078858508588e-06,
      "loss": 0.83745283,
      "num_input_tokens_seen": 23680900,
      "step": 1113,
      "time_per_iteration": 2.640674114227295
    },
    {
      "auxiliary_loss_clip": 0.01340642,
      "auxiliary_loss_mlp": 0.01036466,
      "balance_loss_clip": 1.06740451,
      "balance_loss_mlp": 1.02620745,
      "epoch": 0.13395058017194733,
      "flos": 22563931501440.0,
      "grad_norm": 1.7774901554392817,
      "language_loss": 0.8419975,
      "learning_rate": 3.8878217850471365e-06,
      "loss": 0.86576855,
      "num_input_tokens_seen": 23700815,
      "step": 1114,
      "time_per_iteration": 2.7824339866638184
    },
    {
      "auxiliary_loss_clip": 0.01250378,
      "auxiliary_loss_mlp": 0.01040336,
      "balance_loss_clip": 1.07203531,
      "balance_loss_mlp": 1.03069162,
      "epoch": 0.13407082306258641,
      "flos": 25810938264960.0,
      "grad_norm": 2.7712476798183645,
      "language_loss": 0.74165183,
      "learning_rate": 3.887564425204621e-06,
      "loss": 0.76455897,
      "num_input_tokens_seen": 23722500,
      "step": 1115,
      "time_per_iteration": 3.6334645748138428
    },
    {
      "auxiliary_loss_clip": 0.01294442,
      "auxiliary_loss_mlp": 0.01002468,
      "balance_loss_clip": 1.03172398,
      "balance_loss_mlp": 1.00020313,
      "epoch": 0.13419106595322552,
      "flos": 68338365269760.0,
      "grad_norm": 0.8452824771640849,
      "language_loss": 0.54692215,
      "learning_rate": 3.887306779020083e-06,
      "loss": 0.56989115,
      "num_input_tokens_seen": 23777155,
      "step": 1116,
      "time_per_iteration": 3.2159583568573
    },
    {
      "auxiliary_loss_clip": 0.01300448,
      "auxiliary_loss_mlp": 0.01042108,
      "balance_loss_clip": 1.07105505,
      "balance_loss_mlp": 1.03230309,
      "epoch": 0.1343113088438646,
      "flos": 20449080489600.0,
      "grad_norm": 2.4310785462084366,
      "language_loss": 0.70732743,
      "learning_rate": 3.887048846532608e-06,
      "loss": 0.730753,
      "num_input_tokens_seen": 23794130,
      "step": 1117,
      "time_per_iteration": 2.6651527881622314
    },
    {
      "auxiliary_loss_clip": 0.01301163,
      "auxiliary_loss_mlp": 0.01005687,
      "balance_loss_clip": 1.03343248,
      "balance_loss_mlp": 1.00357652,
      "epoch": 0.1344315517345037,
      "flos": 67389784951680.0,
      "grad_norm": 0.7555369418193514,
      "language_loss": 0.58107632,
      "learning_rate": 3.8867906277813224e-06,
      "loss": 0.60414481,
      "num_input_tokens_seen": 23852285,
      "step": 1118,
      "time_per_iteration": 3.203640937805176
    },
    {
      "auxiliary_loss_clip": 0.01298724,
      "auxiliary_loss_mlp": 0.02578231,
      "balance_loss_clip": 1.06760991,
      "balance_loss_mlp": 1.00058818,
      "epoch": 0.1345517946251428,
      "flos": 40734442788480.0,
      "grad_norm": 2.2775733342100564,
      "language_loss": 0.74062473,
      "learning_rate": 3.886532122805399e-06,
      "loss": 0.77939427,
      "num_input_tokens_seen": 23874765,
      "step": 1119,
      "time_per_iteration": 2.8434252738952637
    },
    {
      "auxiliary_loss_clip": 0.01475784,
      "auxiliary_loss_mlp": 0.01036824,
      "balance_loss_clip": 1.0535419,
      "balance_loss_mlp": 1.0266192,
      "epoch": 0.13467203751578188,
      "flos": 22816850140800.0,
      "grad_norm": 2.0565895518162423,
      "language_loss": 0.89625829,
      "learning_rate": 3.886273331644053e-06,
      "loss": 0.92138433,
      "num_input_tokens_seen": 23893635,
      "step": 1120,
      "time_per_iteration": 2.9582810401916504
    },
    {
      "auxiliary_loss_clip": 0.0143476,
      "auxiliary_loss_mlp": 0.01038936,
      "balance_loss_clip": 1.06272984,
      "balance_loss_mlp": 1.02922606,
      "epoch": 0.13479228040642097,
      "flos": 17091576512640.0,
      "grad_norm": 8.289654547913319,
      "language_loss": 0.82399464,
      "learning_rate": 3.886014254336542e-06,
      "loss": 0.84873164,
      "num_input_tokens_seen": 23910110,
      "step": 1121,
      "time_per_iteration": 2.7270584106445312
    },
    {
      "auxiliary_loss_clip": 0.0129195,
      "auxiliary_loss_mlp": 0.0103932,
      "balance_loss_clip": 1.06513476,
      "balance_loss_mlp": 1.02961087,
      "epoch": 0.13491252329706005,
      "flos": 23730525417600.0,
      "grad_norm": 1.8137252214759514,
      "language_loss": 0.92645037,
      "learning_rate": 3.885754890922168e-06,
      "loss": 0.94976306,
      "num_input_tokens_seen": 23930440,
      "step": 1122,
      "time_per_iteration": 2.7240183353424072
    },
    {
      "auxiliary_loss_clip": 0.01530572,
      "auxiliary_loss_mlp": 0.01041192,
      "balance_loss_clip": 1.0564605,
      "balance_loss_mlp": 1.03132761,
      "epoch": 0.13503276618769916,
      "flos": 34127058960000.0,
      "grad_norm": 1.8820510642790582,
      "language_loss": 0.78514177,
      "learning_rate": 3.885495241440277e-06,
      "loss": 0.81085944,
      "num_input_tokens_seen": 23954535,
      "step": 1123,
      "time_per_iteration": 3.0027265548706055
    },
    {
      "auxiliary_loss_clip": 0.01244227,
      "auxiliary_loss_mlp": 0.01036593,
      "balance_loss_clip": 1.0683558,
      "balance_loss_mlp": 1.02727056,
      "epoch": 0.13515300907833824,
      "flos": 17712328377600.0,
      "grad_norm": 2.528916332846907,
      "language_loss": 0.74473965,
      "learning_rate": 3.885235305930257e-06,
      "loss": 0.76754791,
      "num_input_tokens_seen": 23972735,
      "step": 1124,
      "time_per_iteration": 2.8175761699676514
    },
    {
      "auxiliary_loss_clip": 0.01387828,
      "auxiliary_loss_mlp": 0.01044543,
      "balance_loss_clip": 1.06174064,
      "balance_loss_mlp": 1.03526831,
      "epoch": 0.13527325196897733,
      "flos": 20260872201600.0,
      "grad_norm": 2.393704098493691,
      "language_loss": 0.85246992,
      "learning_rate": 3.884975084431539e-06,
      "loss": 0.87679362,
      "num_input_tokens_seen": 23987685,
      "step": 1125,
      "time_per_iteration": 2.7183263301849365
    },
    {
      "auxiliary_loss_clip": 0.01290188,
      "auxiliary_loss_mlp": 0.02580475,
      "balance_loss_clip": 1.06737602,
      "balance_loss_mlp": 1.00065815,
      "epoch": 0.13539349485961644,
      "flos": 18186492839040.0,
      "grad_norm": 4.44622373560651,
      "language_loss": 0.91513228,
      "learning_rate": 3.8847145769836e-06,
      "loss": 0.95383888,
      "num_input_tokens_seen": 24004105,
      "step": 1126,
      "time_per_iteration": 2.811180830001831
    },
    {
      "auxiliary_loss_clip": 0.01246984,
      "auxiliary_loss_mlp": 0.01045395,
      "balance_loss_clip": 1.06983042,
      "balance_loss_mlp": 1.03609681,
      "epoch": 0.13551373775025552,
      "flos": 19317463441920.0,
      "grad_norm": 2.702149799799932,
      "language_loss": 0.66657043,
      "learning_rate": 3.884453783625959e-06,
      "loss": 0.68949425,
      "num_input_tokens_seen": 24021715,
      "step": 1127,
      "time_per_iteration": 2.609877109527588
    },
    {
      "auxiliary_loss_clip": 0.01336712,
      "auxiliary_loss_mlp": 0.01038214,
      "balance_loss_clip": 1.06534612,
      "balance_loss_mlp": 1.02898693,
      "epoch": 0.1356339806408946,
      "flos": 20850813175680.0,
      "grad_norm": 7.8254915271544165,
      "language_loss": 0.84967041,
      "learning_rate": 3.884192704398176e-06,
      "loss": 0.87341964,
      "num_input_tokens_seen": 24038915,
      "step": 1128,
      "time_per_iteration": 2.665424346923828
    },
    {
      "auxiliary_loss_clip": 0.01259249,
      "auxiliary_loss_mlp": 0.01036644,
      "balance_loss_clip": 1.06604838,
      "balance_loss_mlp": 1.02666056,
      "epoch": 0.13575422353153369,
      "flos": 50476037696640.0,
      "grad_norm": 3.168482599447299,
      "language_loss": 0.74991333,
      "learning_rate": 3.883931339339858e-06,
      "loss": 0.77287233,
      "num_input_tokens_seen": 24063300,
      "step": 1129,
      "time_per_iteration": 2.907115936279297
    },
    {
      "auxiliary_loss_clip": 0.0129769,
      "auxiliary_loss_mlp": 0.01032827,
      "balance_loss_clip": 1.06683099,
      "balance_loss_mlp": 1.02283132,
      "epoch": 0.1358744664221728,
      "flos": 18150797698560.0,
      "grad_norm": 1.8512742124744874,
      "language_loss": 0.78848314,
      "learning_rate": 3.883669688490654e-06,
      "loss": 0.81178832,
      "num_input_tokens_seen": 24081070,
      "step": 1130,
      "time_per_iteration": 2.621971607208252
    },
    {
      "auxiliary_loss_clip": 0.01329653,
      "auxiliary_loss_mlp": 0.02580181,
      "balance_loss_clip": 1.06033778,
      "balance_loss_mlp": 1.00064003,
      "epoch": 0.13599470931281188,
      "flos": 18442966924800.0,
      "grad_norm": 2.881299195851193,
      "language_loss": 0.85314608,
      "learning_rate": 3.883407751890256e-06,
      "loss": 0.89224434,
      "num_input_tokens_seen": 24099675,
      "step": 1131,
      "time_per_iteration": 2.658878803253174
    },
    {
      "auxiliary_loss_clip": 0.01352713,
      "auxiliary_loss_mlp": 0.01042585,
      "balance_loss_clip": 1.05960345,
      "balance_loss_mlp": 1.03220224,
      "epoch": 0.13611495220345096,
      "flos": 26680766014080.0,
      "grad_norm": 1.6040664008497794,
      "language_loss": 0.85595262,
      "learning_rate": 3.8831455295783994e-06,
      "loss": 0.87990558,
      "num_input_tokens_seen": 24118925,
      "step": 1132,
      "time_per_iteration": 2.739625930786133
    },
    {
      "auxiliary_loss_clip": 0.01335061,
      "auxiliary_loss_mlp": 0.01039131,
      "balance_loss_clip": 1.06263649,
      "balance_loss_mlp": 1.02980864,
      "epoch": 0.13623519509409007,
      "flos": 21686238673920.0,
      "grad_norm": 1.6711946218511649,
      "language_loss": 0.74401051,
      "learning_rate": 3.882883021594864e-06,
      "loss": 0.76775241,
      "num_input_tokens_seen": 24137065,
      "step": 1133,
      "time_per_iteration": 3.591701030731201
    },
    {
      "auxiliary_loss_clip": 0.01385784,
      "auxiliary_loss_mlp": 0.01039081,
      "balance_loss_clip": 1.0633471,
      "balance_loss_mlp": 1.02975917,
      "epoch": 0.13635543798472916,
      "flos": 14830389492480.0,
      "grad_norm": 2.027740822976571,
      "language_loss": 0.87268341,
      "learning_rate": 3.8826202279794705e-06,
      "loss": 0.89693201,
      "num_input_tokens_seen": 24154125,
      "step": 1134,
      "time_per_iteration": 2.738779306411743
    },
    {
      "auxiliary_loss_clip": 0.01246832,
      "auxiliary_loss_mlp": 0.01046328,
      "balance_loss_clip": 1.06996536,
      "balance_loss_mlp": 1.03548598,
      "epoch": 0.13647568087536824,
      "flos": 22890323410560.0,
      "grad_norm": 3.02277750481882,
      "language_loss": 0.70237637,
      "learning_rate": 3.882357148772085e-06,
      "loss": 0.725308,
      "num_input_tokens_seen": 24171550,
      "step": 1135,
      "time_per_iteration": 2.636211395263672
    },
    {
      "auxiliary_loss_clip": 0.01379104,
      "auxiliary_loss_mlp": 0.01039354,
      "balance_loss_clip": 1.06166828,
      "balance_loss_mlp": 1.02944744,
      "epoch": 0.13659592376600732,
      "flos": 19937927998080.0,
      "grad_norm": 2.7748438390763632,
      "language_loss": 0.84489805,
      "learning_rate": 3.882093784012617e-06,
      "loss": 0.86908257,
      "num_input_tokens_seen": 24190190,
      "step": 1136,
      "time_per_iteration": 3.5408265590667725
    },
    {
      "auxiliary_loss_clip": 0.01341196,
      "auxiliary_loss_mlp": 0.01037447,
      "balance_loss_clip": 1.0654825,
      "balance_loss_mlp": 1.02824426,
      "epoch": 0.13671616665664643,
      "flos": 21428579439360.0,
      "grad_norm": 2.0493509896375954,
      "language_loss": 0.84032804,
      "learning_rate": 3.881830133741019e-06,
      "loss": 0.86411446,
      "num_input_tokens_seen": 24209055,
      "step": 1137,
      "time_per_iteration": 3.60079288482666
    },
    {
      "auxiliary_loss_clip": 0.01396176,
      "auxiliary_loss_mlp": 0.01038188,
      "balance_loss_clip": 1.06800008,
      "balance_loss_mlp": 1.02826369,
      "epoch": 0.13683640954728551,
      "flos": 22778138257920.0,
      "grad_norm": 2.0032663930666934,
      "language_loss": 0.76268166,
      "learning_rate": 3.881566197997285e-06,
      "loss": 0.78702533,
      "num_input_tokens_seen": 24225490,
      "step": 1138,
      "time_per_iteration": 2.82951021194458
    },
    {
      "auxiliary_loss_clip": 0.01335771,
      "auxiliary_loss_mlp": 0.01040205,
      "balance_loss_clip": 1.06614339,
      "balance_loss_mlp": 1.03161037,
      "epoch": 0.1369566524379246,
      "flos": 21725884310400.0,
      "grad_norm": 2.5884105700516966,
      "language_loss": 0.74954677,
      "learning_rate": 3.881301976821456e-06,
      "loss": 0.77330649,
      "num_input_tokens_seen": 24245520,
      "step": 1139,
      "time_per_iteration": 2.7211484909057617
    },
    {
      "auxiliary_loss_clip": 0.012875,
      "auxiliary_loss_mlp": 0.01033169,
      "balance_loss_clip": 1.06480312,
      "balance_loss_mlp": 1.02394176,
      "epoch": 0.1370768953285637,
      "flos": 18624459369600.0,
      "grad_norm": 2.362134967045639,
      "language_loss": 0.90862304,
      "learning_rate": 3.881037470253612e-06,
      "loss": 0.93182969,
      "num_input_tokens_seen": 24265035,
      "step": 1140,
      "time_per_iteration": 3.5152883529663086
    },
    {
      "auxiliary_loss_clip": 0.01436222,
      "auxiliary_loss_mlp": 0.01047304,
      "balance_loss_clip": 1.05953455,
      "balance_loss_mlp": 1.03738558,
      "epoch": 0.1371971382192028,
      "flos": 14939521989120.0,
      "grad_norm": 2.736708997979135,
      "language_loss": 0.79619569,
      "learning_rate": 3.88077267833388e-06,
      "loss": 0.82103097,
      "num_input_tokens_seen": 24281550,
      "step": 1141,
      "time_per_iteration": 2.78802752494812
    },
    {
      "auxiliary_loss_clip": 0.0143371,
      "auxiliary_loss_mlp": 0.01037254,
      "balance_loss_clip": 1.05917835,
      "balance_loss_mlp": 1.02729368,
      "epoch": 0.13731738110984187,
      "flos": 19023785844480.0,
      "grad_norm": 2.096980003319845,
      "language_loss": 0.84009826,
      "learning_rate": 3.880507601102427e-06,
      "loss": 0.86480784,
      "num_input_tokens_seen": 24299485,
      "step": 1142,
      "time_per_iteration": 2.7342684268951416
    },
    {
      "auxiliary_loss_clip": 0.0124504,
      "auxiliary_loss_mlp": 0.01037369,
      "balance_loss_clip": 1.07255149,
      "balance_loss_mlp": 1.02808857,
      "epoch": 0.13743762400048098,
      "flos": 18187462506240.0,
      "grad_norm": 2.351620964382535,
      "language_loss": 0.81967211,
      "learning_rate": 3.880242238599467e-06,
      "loss": 0.84249622,
      "num_input_tokens_seen": 24316010,
      "step": 1143,
      "time_per_iteration": 2.6631782054901123
    },
    {
      "auxiliary_loss_clip": 0.01242106,
      "auxiliary_loss_mlp": 0.01034842,
      "balance_loss_clip": 1.06793737,
      "balance_loss_mlp": 1.02451801,
      "epoch": 0.13755786689112007,
      "flos": 21031982398080.0,
      "grad_norm": 1.663897023011702,
      "language_loss": 0.83078283,
      "learning_rate": 3.879976590865254e-06,
      "loss": 0.85355234,
      "num_input_tokens_seen": 24335465,
      "step": 1144,
      "time_per_iteration": 2.661515712738037
    },
    {
      "auxiliary_loss_clip": 0.01337964,
      "auxiliary_loss_mlp": 0.01039966,
      "balance_loss_clip": 1.06714511,
      "balance_loss_mlp": 1.02972579,
      "epoch": 0.13767810978175915,
      "flos": 21360636864000.0,
      "grad_norm": 2.4542638780236175,
      "language_loss": 0.8701216,
      "learning_rate": 3.879710657940087e-06,
      "loss": 0.89390087,
      "num_input_tokens_seen": 24354415,
      "step": 1145,
      "time_per_iteration": 2.793872833251953
    },
    {
      "auxiliary_loss_clip": 0.01287181,
      "auxiliary_loss_mlp": 0.01032683,
      "balance_loss_clip": 1.06588161,
      "balance_loss_mlp": 1.02316403,
      "epoch": 0.13779835267239823,
      "flos": 30592084861440.0,
      "grad_norm": 2.0911328907835656,
      "language_loss": 0.70162308,
      "learning_rate": 3.879444439864308e-06,
      "loss": 0.72482175,
      "num_input_tokens_seen": 24373990,
      "step": 1146,
      "time_per_iteration": 2.7725989818573
    },
    {
      "auxiliary_loss_clip": 0.01288982,
      "auxiliary_loss_mlp": 0.0258227,
      "balance_loss_clip": 1.06322682,
      "balance_loss_mlp": 1.00084102,
      "epoch": 0.13791859556303734,
      "flos": 22669867687680.0,
      "grad_norm": 2.12524975082109,
      "language_loss": 0.85832077,
      "learning_rate": 3.879177936678301e-06,
      "loss": 0.89703327,
      "num_input_tokens_seen": 24392995,
      "step": 1147,
      "time_per_iteration": 2.654008626937866
    },
    {
      "auxiliary_loss_clip": 0.01340593,
      "auxiliary_loss_mlp": 0.01034247,
      "balance_loss_clip": 1.06325948,
      "balance_loss_mlp": 1.02479386,
      "epoch": 0.13803883845367643,
      "flos": 35224166016000.0,
      "grad_norm": 1.9719479432006661,
      "language_loss": 0.77208126,
      "learning_rate": 3.878911148422496e-06,
      "loss": 0.79582959,
      "num_input_tokens_seen": 24414470,
      "step": 1148,
      "time_per_iteration": 2.8433077335357666
    },
    {
      "auxiliary_loss_clip": 0.01295613,
      "auxiliary_loss_mlp": 0.0104426,
      "balance_loss_clip": 1.06974959,
      "balance_loss_mlp": 1.03485417,
      "epoch": 0.1381590813443155,
      "flos": 32014542332160.0,
      "grad_norm": 3.2570887080985433,
      "language_loss": 0.70508271,
      "learning_rate": 3.878644075137364e-06,
      "loss": 0.72848141,
      "num_input_tokens_seen": 24435120,
      "step": 1149,
      "time_per_iteration": 2.7706148624420166
    },
    {
      "auxiliary_loss_clip": 0.01370702,
      "auxiliary_loss_mlp": 0.01040581,
      "balance_loss_clip": 1.05762637,
      "balance_loss_mlp": 1.03106785,
      "epoch": 0.13827932423495462,
      "flos": 17821855923840.0,
      "grad_norm": 2.759840580936066,
      "language_loss": 0.79550618,
      "learning_rate": 3.878376716863418e-06,
      "loss": 0.819619,
      "num_input_tokens_seen": 24451420,
      "step": 1150,
      "time_per_iteration": 2.7194623947143555
    },
    {
      "auxiliary_loss_clip": 0.01335412,
      "auxiliary_loss_mlp": 0.01037742,
      "balance_loss_clip": 1.06176054,
      "balance_loss_mlp": 1.02768052,
      "epoch": 0.1383995671255937,
      "flos": 19427098728960.0,
      "grad_norm": 2.1997708888708853,
      "language_loss": 0.71952426,
      "learning_rate": 3.878109073641219e-06,
      "loss": 0.74325579,
      "num_input_tokens_seen": 24470450,
      "step": 1151,
      "time_per_iteration": 2.6973438262939453
    },
    {
      "auxiliary_loss_clip": 0.01436015,
      "auxiliary_loss_mlp": 0.01044445,
      "balance_loss_clip": 1.06308901,
      "balance_loss_mlp": 1.03545022,
      "epoch": 0.13851981001623279,
      "flos": 28296603331200.0,
      "grad_norm": 1.6537559566663962,
      "language_loss": 0.81019962,
      "learning_rate": 3.877841145511366e-06,
      "loss": 0.83500421,
      "num_input_tokens_seen": 24493190,
      "step": 1152,
      "time_per_iteration": 2.8671274185180664
    },
    {
      "auxiliary_loss_clip": 0.01254527,
      "auxiliary_loss_mlp": 0.01044144,
      "balance_loss_clip": 1.06618059,
      "balance_loss_mlp": 1.03416646,
      "epoch": 0.13864005290687187,
      "flos": 21213079793280.0,
      "grad_norm": 2.01112195580312,
      "language_loss": 0.82961023,
      "learning_rate": 3.8775729325145035e-06,
      "loss": 0.852597,
      "num_input_tokens_seen": 24512425,
      "step": 1153,
      "time_per_iteration": 2.6836555004119873
    },
    {
      "auxiliary_loss_clip": 0.01286731,
      "auxiliary_loss_mlp": 0.01002439,
      "balance_loss_clip": 1.03682041,
      "balance_loss_mlp": 0.99997121,
      "epoch": 0.13876029579751098,
      "flos": 71653389413760.0,
      "grad_norm": 0.7999059977067157,
      "language_loss": 0.64701462,
      "learning_rate": 3.877304434691321e-06,
      "loss": 0.66990626,
      "num_input_tokens_seen": 24579275,
      "step": 1154,
      "time_per_iteration": 3.422255277633667
    },
    {
      "auxiliary_loss_clip": 0.01391284,
      "auxiliary_loss_mlp": 0.01037007,
      "balance_loss_clip": 1.06381285,
      "balance_loss_mlp": 1.02792358,
      "epoch": 0.13888053868815006,
      "flos": 21941348042880.0,
      "grad_norm": 1.778548295453032,
      "language_loss": 0.7955125,
      "learning_rate": 3.877035652082548e-06,
      "loss": 0.81979543,
      "num_input_tokens_seen": 24598720,
      "step": 1155,
      "time_per_iteration": 2.8107852935791016
    },
    {
      "auxiliary_loss_clip": 0.0133376,
      "auxiliary_loss_mlp": 0.0105045,
      "balance_loss_clip": 1.06445801,
      "balance_loss_mlp": 1.04034066,
      "epoch": 0.13900078157878915,
      "flos": 19608627087360.0,
      "grad_norm": 1.7463407663893655,
      "language_loss": 0.85311484,
      "learning_rate": 3.87676658472896e-06,
      "loss": 0.87695694,
      "num_input_tokens_seen": 24617530,
      "step": 1156,
      "time_per_iteration": 2.6930294036865234
    },
    {
      "auxiliary_loss_clip": 0.01288544,
      "auxiliary_loss_mlp": 0.01042347,
      "balance_loss_clip": 1.0626229,
      "balance_loss_mlp": 1.03214836,
      "epoch": 0.13912102446942826,
      "flos": 22638051216000.0,
      "grad_norm": 1.993002499743212,
      "language_loss": 0.85373634,
      "learning_rate": 3.876497232671372e-06,
      "loss": 0.87704527,
      "num_input_tokens_seen": 24637485,
      "step": 1157,
      "time_per_iteration": 2.6358494758605957
    },
    {
      "auxiliary_loss_clip": 0.01438611,
      "auxiliary_loss_mlp": 0.01037087,
      "balance_loss_clip": 1.05801392,
      "balance_loss_mlp": 1.02762747,
      "epoch": 0.13924126736006734,
      "flos": 29643324975360.0,
      "grad_norm": 2.3670781864294397,
      "language_loss": 0.83453226,
      "learning_rate": 3.876227595950647e-06,
      "loss": 0.85928917,
      "num_input_tokens_seen": 24656915,
      "step": 1158,
      "time_per_iteration": 3.7373085021972656
    },
    {
      "auxiliary_loss_clip": 0.0123785,
      "auxiliary_loss_mlp": 0.01039352,
      "balance_loss_clip": 1.06601238,
      "balance_loss_mlp": 1.02989876,
      "epoch": 0.13936151025070642,
      "flos": 27417653527680.0,
      "grad_norm": 1.934696429293343,
      "language_loss": 0.78893954,
      "learning_rate": 3.875957674607686e-06,
      "loss": 0.81171155,
      "num_input_tokens_seen": 24679190,
      "step": 1159,
      "time_per_iteration": 2.6655707359313965
    },
    {
      "auxiliary_loss_clip": 0.01282803,
      "auxiliary_loss_mlp": 0.0258444,
      "balance_loss_clip": 1.06079459,
      "balance_loss_mlp": 1.00089645,
      "epoch": 0.1394817531413455,
      "flos": 16399326625920.0,
      "grad_norm": 2.304027128232615,
      "language_loss": 0.88142115,
      "learning_rate": 3.8756874686834386e-06,
      "loss": 0.9200936,
      "num_input_tokens_seen": 24697405,
      "step": 1160,
      "time_per_iteration": 2.6889147758483887
    },
    {
      "auxiliary_loss_clip": 0.01289091,
      "auxiliary_loss_mlp": 0.02583655,
      "balance_loss_clip": 1.06243479,
      "balance_loss_mlp": 1.00085592,
      "epoch": 0.13960199603198462,
      "flos": 30922319525760.0,
      "grad_norm": 1.7938193879242268,
      "language_loss": 0.80449772,
      "learning_rate": 3.875416978218893e-06,
      "loss": 0.84322518,
      "num_input_tokens_seen": 24720600,
      "step": 1161,
      "time_per_iteration": 2.7121808528900146
    },
    {
      "auxiliary_loss_clip": 0.01357067,
      "auxiliary_loss_mlp": 0.01039016,
      "balance_loss_clip": 1.05940795,
      "balance_loss_mlp": 1.02931213,
      "epoch": 0.1397222389226237,
      "flos": 18113773754880.0,
      "grad_norm": 2.334614924009087,
      "language_loss": 0.82525563,
      "learning_rate": 3.8751462032550835e-06,
      "loss": 0.8492164,
      "num_input_tokens_seen": 24737605,
      "step": 1162,
      "time_per_iteration": 2.7338407039642334
    },
    {
      "auxiliary_loss_clip": 0.01330737,
      "auxiliary_loss_mlp": 0.01046115,
      "balance_loss_clip": 1.06478524,
      "balance_loss_mlp": 1.03644109,
      "epoch": 0.13984248181326278,
      "flos": 16872772815360.0,
      "grad_norm": 2.113999324237088,
      "language_loss": 0.83327878,
      "learning_rate": 3.874875143833085e-06,
      "loss": 0.8570472,
      "num_input_tokens_seen": 24755845,
      "step": 1163,
      "time_per_iteration": 4.455169439315796
    },
    {
      "auxiliary_loss_clip": 0.01290528,
      "auxiliary_loss_mlp": 0.01034141,
      "balance_loss_clip": 1.06657314,
      "balance_loss_mlp": 1.02421141,
      "epoch": 0.1399627247039019,
      "flos": 54121401267840.0,
      "grad_norm": 1.7945904792970173,
      "language_loss": 0.68931532,
      "learning_rate": 3.874603799994019e-06,
      "loss": 0.71256196,
      "num_input_tokens_seen": 24779380,
      "step": 1164,
      "time_per_iteration": 3.0180506706237793
    },
    {
      "auxiliary_loss_clip": 0.01379173,
      "auxiliary_loss_mlp": 0.01039319,
      "balance_loss_clip": 1.05703187,
      "balance_loss_mlp": 1.03050971,
      "epoch": 0.14008296759454097,
      "flos": 11765521618560.0,
      "grad_norm": 5.239547867787885,
      "language_loss": 0.86828625,
      "learning_rate": 3.874332171779046e-06,
      "loss": 0.89247113,
      "num_input_tokens_seen": 24794260,
      "step": 1165,
      "time_per_iteration": 2.8194854259490967
    },
    {
      "auxiliary_loss_clip": 0.01382823,
      "auxiliary_loss_mlp": 0.01034814,
      "balance_loss_clip": 1.05824637,
      "balance_loss_mlp": 1.02583742,
      "epoch": 0.14020321048518006,
      "flos": 22017514832640.0,
      "grad_norm": 2.464453077999299,
      "language_loss": 0.75727749,
      "learning_rate": 3.874060259229373e-06,
      "loss": 0.78145385,
      "num_input_tokens_seen": 24815835,
      "step": 1166,
      "time_per_iteration": 3.653068780899048
    },
    {
      "auxiliary_loss_clip": 0.01290366,
      "auxiliary_loss_mlp": 0.01042018,
      "balance_loss_clip": 1.06762862,
      "balance_loss_mlp": 1.03233767,
      "epoch": 0.14032345337581917,
      "flos": 23404313076480.0,
      "grad_norm": 2.2608457006764167,
      "language_loss": 0.93589455,
      "learning_rate": 3.873788062386249e-06,
      "loss": 0.95921838,
      "num_input_tokens_seen": 24834095,
      "step": 1167,
      "time_per_iteration": 2.7514114379882812
    },
    {
      "auxiliary_loss_clip": 0.01392689,
      "auxiliary_loss_mlp": 0.0104125,
      "balance_loss_clip": 1.06525838,
      "balance_loss_mlp": 1.03193998,
      "epoch": 0.14044369626645825,
      "flos": 29645767100160.0,
      "grad_norm": 3.0531584789909028,
      "language_loss": 0.82086897,
      "learning_rate": 3.873515581290965e-06,
      "loss": 0.84520835,
      "num_input_tokens_seen": 24858900,
      "step": 1168,
      "time_per_iteration": 2.8313028812408447
    },
    {
      "auxiliary_loss_clip": 0.01386521,
      "auxiliary_loss_mlp": 0.01039813,
      "balance_loss_clip": 1.0647521,
      "balance_loss_mlp": 1.03089666,
      "epoch": 0.14056393915709733,
      "flos": 18332972501760.0,
      "grad_norm": 2.1869151629319292,
      "language_loss": 0.7503643,
      "learning_rate": 3.8732428159848575e-06,
      "loss": 0.77462763,
      "num_input_tokens_seen": 24877875,
      "step": 1169,
      "time_per_iteration": 2.686420440673828
    },
    {
      "auxiliary_loss_clip": 0.0129065,
      "auxiliary_loss_mlp": 0.01040543,
      "balance_loss_clip": 1.06954193,
      "balance_loss_mlp": 1.03059506,
      "epoch": 0.14068418204773642,
      "flos": 26687517770880.0,
      "grad_norm": 1.92410383940294,
      "language_loss": 0.78333652,
      "learning_rate": 3.872969766509304e-06,
      "loss": 0.80664843,
      "num_input_tokens_seen": 24898430,
      "step": 1170,
      "time_per_iteration": 2.7926411628723145
    },
    {
      "auxiliary_loss_clip": 0.01287467,
      "auxiliary_loss_mlp": 0.01002406,
      "balance_loss_clip": 1.03307402,
      "balance_loss_mlp": 1.00034356,
      "epoch": 0.14080442493837553,
      "flos": 65259314501760.0,
      "grad_norm": 0.7654609407625333,
      "language_loss": 0.55628324,
      "learning_rate": 3.872696432905726e-06,
      "loss": 0.57918197,
      "num_input_tokens_seen": 24959250,
      "step": 1171,
      "time_per_iteration": 3.2620954513549805
    },
    {
      "auxiliary_loss_clip": 0.01290963,
      "auxiliary_loss_mlp": 0.01039397,
      "balance_loss_clip": 1.06282949,
      "balance_loss_mlp": 1.02927065,
      "epoch": 0.1409246678290146,
      "flos": 25776715582080.0,
      "grad_norm": 3.5929861316836407,
      "language_loss": 0.7155661,
      "learning_rate": 3.872422815215589e-06,
      "loss": 0.73886967,
      "num_input_tokens_seen": 24978330,
      "step": 1172,
      "time_per_iteration": 2.7150886058807373
    },
    {
      "auxiliary_loss_clip": 0.0127894,
      "auxiliary_loss_mlp": 0.01036096,
      "balance_loss_clip": 1.05651462,
      "balance_loss_mlp": 1.02580249,
      "epoch": 0.1410449107196537,
      "flos": 21868521217920.0,
      "grad_norm": 2.0028469374832722,
      "language_loss": 0.74479181,
      "learning_rate": 3.8721489134803994e-06,
      "loss": 0.76794219,
      "num_input_tokens_seen": 24997120,
      "step": 1173,
      "time_per_iteration": 2.6243033409118652
    },
    {
      "auxiliary_loss_clip": 0.01286288,
      "auxiliary_loss_mlp": 0.01036533,
      "balance_loss_clip": 1.06552982,
      "balance_loss_mlp": 1.02700782,
      "epoch": 0.1411651536102928,
      "flos": 16684133564160.0,
      "grad_norm": 2.123088185214369,
      "language_loss": 0.72481573,
      "learning_rate": 3.871874727741707e-06,
      "loss": 0.74804395,
      "num_input_tokens_seen": 25014350,
      "step": 1174,
      "time_per_iteration": 2.6778087615966797
    },
    {
      "auxiliary_loss_clip": 0.0128473,
      "auxiliary_loss_mlp": 0.01033832,
      "balance_loss_clip": 1.06749964,
      "balance_loss_mlp": 1.02511191,
      "epoch": 0.1412853965009319,
      "flos": 20992264934400.0,
      "grad_norm": 1.7678846339249012,
      "language_loss": 0.96685755,
      "learning_rate": 3.871600258041108e-06,
      "loss": 0.99004316,
      "num_input_tokens_seen": 25033875,
      "step": 1175,
      "time_per_iteration": 2.7081143856048584
    },
    {
      "auxiliary_loss_clip": 0.01334787,
      "auxiliary_loss_mlp": 0.01039373,
      "balance_loss_clip": 1.06158495,
      "balance_loss_mlp": 1.03000891,
      "epoch": 0.14140563939157097,
      "flos": 20335279224960.0,
      "grad_norm": 3.7198824037387426,
      "language_loss": 0.86381757,
      "learning_rate": 3.871325504420238e-06,
      "loss": 0.88755918,
      "num_input_tokens_seen": 25052865,
      "step": 1176,
      "time_per_iteration": 2.701585292816162
    },
    {
      "auxiliary_loss_clip": 0.01237713,
      "auxiliary_loss_mlp": 0.01032708,
      "balance_loss_clip": 1.0676465,
      "balance_loss_mlp": 1.02349329,
      "epoch": 0.14152588228221005,
      "flos": 21068826773760.0,
      "grad_norm": 2.1896544135266907,
      "language_loss": 0.82061362,
      "learning_rate": 3.871050466920776e-06,
      "loss": 0.84331781,
      "num_input_tokens_seen": 25072770,
      "step": 1177,
      "time_per_iteration": 2.795872449874878
    },
    {
      "auxiliary_loss_clip": 0.01379444,
      "auxiliary_loss_mlp": 0.01044283,
      "balance_loss_clip": 1.057868,
      "balance_loss_mlp": 1.03518176,
      "epoch": 0.14164612517284916,
      "flos": 18223157646720.0,
      "grad_norm": 1.8838415088974632,
      "language_loss": 0.79517615,
      "learning_rate": 3.870775145584447e-06,
      "loss": 0.81941342,
      "num_input_tokens_seen": 25090550,
      "step": 1178,
      "time_per_iteration": 2.684641122817993
    },
    {
      "auxiliary_loss_clip": 0.01343159,
      "auxiliary_loss_mlp": 0.01040195,
      "balance_loss_clip": 1.06396604,
      "balance_loss_mlp": 1.03021681,
      "epoch": 0.14176636806348825,
      "flos": 22744454279040.0,
      "grad_norm": 3.699326328707347,
      "language_loss": 0.64871299,
      "learning_rate": 3.8704995404530145e-06,
      "loss": 0.67254657,
      "num_input_tokens_seen": 25106175,
      "step": 1179,
      "time_per_iteration": 2.7411813735961914
    },
    {
      "auxiliary_loss_clip": 0.01235081,
      "auxiliary_loss_mlp": 0.01035946,
      "balance_loss_clip": 1.06747746,
      "balance_loss_mlp": 1.02782178,
      "epoch": 0.14188661095412733,
      "flos": 22091095843200.0,
      "grad_norm": 2.071450114916175,
      "language_loss": 0.84899968,
      "learning_rate": 3.87022365156829e-06,
      "loss": 0.87170994,
      "num_input_tokens_seen": 25126890,
      "step": 1180,
      "time_per_iteration": 2.701616048812866
    },
    {
      "auxiliary_loss_clip": 0.01485697,
      "auxiliary_loss_mlp": 0.01033921,
      "balance_loss_clip": 1.05690098,
      "balance_loss_mlp": 1.02513504,
      "epoch": 0.14200685384476644,
      "flos": 24352390604160.0,
      "grad_norm": 2.441751801008685,
      "language_loss": 0.81036997,
      "learning_rate": 3.869947478972123e-06,
      "loss": 0.83556616,
      "num_input_tokens_seen": 25147915,
      "step": 1181,
      "time_per_iteration": 2.921440601348877
    },
    {
      "auxiliary_loss_clip": 0.01284749,
      "auxiliary_loss_mlp": 0.01034407,
      "balance_loss_clip": 1.06326127,
      "balance_loss_mlp": 1.02522779,
      "epoch": 0.14212709673540552,
      "flos": 24022048199040.0,
      "grad_norm": 2.1189825072708097,
      "language_loss": 0.823874,
      "learning_rate": 3.869671022706412e-06,
      "loss": 0.84706557,
      "num_input_tokens_seen": 25166645,
      "step": 1182,
      "time_per_iteration": 2.7932496070861816
    },
    {
      "auxiliary_loss_clip": 0.01420674,
      "auxiliary_loss_mlp": 0.01035352,
      "balance_loss_clip": 1.05251217,
      "balance_loss_mlp": 1.02600586,
      "epoch": 0.1422473396260446,
      "flos": 26431797870720.0,
      "grad_norm": 2.0395695182890727,
      "language_loss": 0.65423143,
      "learning_rate": 3.869394282813092e-06,
      "loss": 0.6787917,
      "num_input_tokens_seen": 25185845,
      "step": 1183,
      "time_per_iteration": 2.797970771789551
    },
    {
      "auxiliary_loss_clip": 0.01392359,
      "auxiliary_loss_mlp": 0.01035943,
      "balance_loss_clip": 1.05871284,
      "balance_loss_mlp": 1.0263468,
      "epoch": 0.1423675825166837,
      "flos": 17055306754560.0,
      "grad_norm": 2.72778973616459,
      "language_loss": 0.89143813,
      "learning_rate": 3.869117259334147e-06,
      "loss": 0.91572118,
      "num_input_tokens_seen": 25203770,
      "step": 1184,
      "time_per_iteration": 2.6758124828338623
    },
    {
      "auxiliary_loss_clip": 0.01283126,
      "auxiliary_loss_mlp": 0.01035984,
      "balance_loss_clip": 1.06228673,
      "balance_loss_mlp": 1.02685261,
      "epoch": 0.1424878254073228,
      "flos": 17929480049280.0,
      "grad_norm": 1.7474066776101012,
      "language_loss": 0.82094824,
      "learning_rate": 3.868839952311599e-06,
      "loss": 0.8441394,
      "num_input_tokens_seen": 25221725,
      "step": 1185,
      "time_per_iteration": 3.6009488105773926
    },
    {
      "auxiliary_loss_clip": 0.01331146,
      "auxiliary_loss_mlp": 0.01036365,
      "balance_loss_clip": 1.06433296,
      "balance_loss_mlp": 1.02755547,
      "epoch": 0.14260806829796188,
      "flos": 20303606407680.0,
      "grad_norm": 2.255238709545601,
      "language_loss": 0.80629849,
      "learning_rate": 3.868562361787516e-06,
      "loss": 0.82997364,
      "num_input_tokens_seen": 25240855,
      "step": 1186,
      "time_per_iteration": 2.715620279312134
    },
    {
      "auxiliary_loss_clip": 0.0148524,
      "auxiliary_loss_mlp": 0.01032002,
      "balance_loss_clip": 1.05165255,
      "balance_loss_mlp": 1.02273393,
      "epoch": 0.14272831118860096,
      "flos": 23185724860800.0,
      "grad_norm": 3.40921173515509,
      "language_loss": 0.69124246,
      "learning_rate": 3.868284487804009e-06,
      "loss": 0.71641487,
      "num_input_tokens_seen": 25260085,
      "step": 1187,
      "time_per_iteration": 2.86678409576416
    },
    {
      "auxiliary_loss_clip": 0.01340745,
      "auxiliary_loss_mlp": 0.01037187,
      "balance_loss_clip": 1.06055748,
      "balance_loss_mlp": 1.02844858,
      "epoch": 0.14284855407924008,
      "flos": 27232210586880.0,
      "grad_norm": 1.8487722618513007,
      "language_loss": 0.78104866,
      "learning_rate": 3.86800633040323e-06,
      "loss": 0.80482799,
      "num_input_tokens_seen": 25280675,
      "step": 1188,
      "time_per_iteration": 4.1762495040893555
    },
    {
      "auxiliary_loss_clip": 0.01334226,
      "auxiliary_loss_mlp": 0.02582126,
      "balance_loss_clip": 1.06450117,
      "balance_loss_mlp": 1.00067997,
      "epoch": 0.14296879696987916,
      "flos": 28184202696960.0,
      "grad_norm": 2.2095262040210706,
      "language_loss": 0.78005183,
      "learning_rate": 3.867727889627376e-06,
      "loss": 0.81921536,
      "num_input_tokens_seen": 25300290,
      "step": 1189,
      "time_per_iteration": 3.595963716506958
    },
    {
      "auxiliary_loss_clip": 0.01373432,
      "auxiliary_loss_mlp": 0.01037122,
      "balance_loss_clip": 1.0601095,
      "balance_loss_mlp": 1.02755582,
      "epoch": 0.14308903986051824,
      "flos": 19390290266880.0,
      "grad_norm": 2.2178316519889654,
      "language_loss": 0.78478098,
      "learning_rate": 3.867449165518687e-06,
      "loss": 0.80888653,
      "num_input_tokens_seen": 25316760,
      "step": 1190,
      "time_per_iteration": 2.6924540996551514
    },
    {
      "auxiliary_loss_clip": 0.01240928,
      "auxiliary_loss_mlp": 0.02584965,
      "balance_loss_clip": 1.06707132,
      "balance_loss_mlp": 1.00068355,
      "epoch": 0.14320928275115732,
      "flos": 17457506317440.0,
      "grad_norm": 1.8692882727744065,
      "language_loss": 0.71139455,
      "learning_rate": 3.867170158119444e-06,
      "loss": 0.74965346,
      "num_input_tokens_seen": 25335760,
      "step": 1191,
      "time_per_iteration": 2.648721933364868
    },
    {
      "auxiliary_loss_clip": 0.01241237,
      "auxiliary_loss_mlp": 0.01033708,
      "balance_loss_clip": 1.06700253,
      "balance_loss_mlp": 1.02449346,
      "epoch": 0.14332952564179643,
      "flos": 21466070259840.0,
      "grad_norm": 2.019109682857975,
      "language_loss": 0.76028681,
      "learning_rate": 3.866890867471972e-06,
      "loss": 0.78303623,
      "num_input_tokens_seen": 25354230,
      "step": 1192,
      "time_per_iteration": 3.5189919471740723
    },
    {
      "auxiliary_loss_clip": 0.01288866,
      "auxiliary_loss_mlp": 0.0103715,
      "balance_loss_clip": 1.05779338,
      "balance_loss_mlp": 1.02711248,
      "epoch": 0.14344976853243552,
      "flos": 16396992241920.0,
      "grad_norm": 2.6527082825223007,
      "language_loss": 0.89793551,
      "learning_rate": 3.86661129361864e-06,
      "loss": 0.92119563,
      "num_input_tokens_seen": 25368720,
      "step": 1193,
      "time_per_iteration": 2.671086311340332
    },
    {
      "auxiliary_loss_clip": 0.01336608,
      "auxiliary_loss_mlp": 0.01042944,
      "balance_loss_clip": 1.06439483,
      "balance_loss_mlp": 1.03294206,
      "epoch": 0.1435700114230746,
      "flos": 18916736336640.0,
      "grad_norm": 3.026880728578261,
      "language_loss": 0.86078882,
      "learning_rate": 3.866331436601859e-06,
      "loss": 0.88458431,
      "num_input_tokens_seen": 25386715,
      "step": 1194,
      "time_per_iteration": 2.637600898742676
    },
    {
      "auxiliary_loss_clip": 0.01241476,
      "auxiliary_loss_mlp": 0.01035615,
      "balance_loss_clip": 1.06974196,
      "balance_loss_mlp": 1.02612615,
      "epoch": 0.1436902543137137,
      "flos": 19755394058880.0,
      "grad_norm": 2.4480150113749892,
      "language_loss": 0.73761648,
      "learning_rate": 3.866051296464083e-06,
      "loss": 0.76038742,
      "num_input_tokens_seen": 25405550,
      "step": 1195,
      "time_per_iteration": 2.59543776512146
    },
    {
      "auxiliary_loss_clip": 0.01238812,
      "auxiliary_loss_mlp": 0.02584345,
      "balance_loss_clip": 1.06471753,
      "balance_loss_mlp": 1.0006423,
      "epoch": 0.1438104972043528,
      "flos": 14684807669760.0,
      "grad_norm": 3.962081774742827,
      "language_loss": 0.85376573,
      "learning_rate": 3.86577087324781e-06,
      "loss": 0.89199728,
      "num_input_tokens_seen": 25422040,
      "step": 1196,
      "time_per_iteration": 2.6189918518066406
    },
    {
      "auxiliary_loss_clip": 0.01283554,
      "auxiliary_loss_mlp": 0.01038931,
      "balance_loss_clip": 1.06562877,
      "balance_loss_mlp": 1.0294714,
      "epoch": 0.14393074009499188,
      "flos": 17092330698240.0,
      "grad_norm": 2.404079574651244,
      "language_loss": 0.77677053,
      "learning_rate": 3.865490166995578e-06,
      "loss": 0.79999542,
      "num_input_tokens_seen": 25440270,
      "step": 1197,
      "time_per_iteration": 2.6092305183410645
    },
    {
      "auxiliary_loss_clip": 0.01291806,
      "auxiliary_loss_mlp": 0.01039918,
      "balance_loss_clip": 1.06795406,
      "balance_loss_mlp": 1.03132296,
      "epoch": 0.144050982985631,
      "flos": 30476200608000.0,
      "grad_norm": 2.172372413655442,
      "language_loss": 0.84431541,
      "learning_rate": 3.86520917774997e-06,
      "loss": 0.86763275,
      "num_input_tokens_seen": 25459705,
      "step": 1198,
      "time_per_iteration": 2.774808406829834
    },
    {
      "auxiliary_loss_clip": 0.01283228,
      "auxiliary_loss_mlp": 0.01043019,
      "balance_loss_clip": 1.06383193,
      "balance_loss_mlp": 1.03331578,
      "epoch": 0.14417122587627007,
      "flos": 17858484817920.0,
      "grad_norm": 2.0715497942054673,
      "language_loss": 0.75300509,
      "learning_rate": 3.864927905553614e-06,
      "loss": 0.77626753,
      "num_input_tokens_seen": 25477615,
      "step": 1199,
      "time_per_iteration": 2.6093714237213135
    },
    {
      "auxiliary_loss_clip": 0.01379511,
      "auxiliary_loss_mlp": 0.01034295,
      "balance_loss_clip": 1.05815601,
      "balance_loss_mlp": 1.0253365,
      "epoch": 0.14429146876690915,
      "flos": 21613914639360.0,
      "grad_norm": 2.0717603965476314,
      "language_loss": 0.89040101,
      "learning_rate": 3.8646463504491765e-06,
      "loss": 0.9145391,
      "num_input_tokens_seen": 25497750,
      "step": 1200,
      "time_per_iteration": 2.7505252361297607
    },
    {
      "auxiliary_loss_clip": 0.01289627,
      "auxiliary_loss_mlp": 0.01039482,
      "balance_loss_clip": 1.06812155,
      "balance_loss_mlp": 1.02977204,
      "epoch": 0.14441171165754824,
      "flos": 23258120722560.0,
      "grad_norm": 1.9460681098553418,
      "language_loss": 0.83176601,
      "learning_rate": 3.8643645124793705e-06,
      "loss": 0.85505712,
      "num_input_tokens_seen": 25516650,
      "step": 1201,
      "time_per_iteration": 2.7777130603790283
    },
    {
      "auxiliary_loss_clip": 0.0128356,
      "auxiliary_loss_mlp": 0.01042942,
      "balance_loss_clip": 1.0636878,
      "balance_loss_mlp": 1.0327673,
      "epoch": 0.14453195454818735,
      "flos": 42854213963520.0,
      "grad_norm": 1.9836350940892595,
      "language_loss": 0.74993527,
      "learning_rate": 3.8640823916869515e-06,
      "loss": 0.77320033,
      "num_input_tokens_seen": 25540960,
      "step": 1202,
      "time_per_iteration": 2.906296491622925
    },
    {
      "auxiliary_loss_clip": 0.01237574,
      "auxiliary_loss_mlp": 0.01038281,
      "balance_loss_clip": 1.06519413,
      "balance_loss_mlp": 1.02864313,
      "epoch": 0.14465219743882643,
      "flos": 27235873774080.0,
      "grad_norm": 1.557451218160784,
      "language_loss": 0.7850765,
      "learning_rate": 3.863799988114714e-06,
      "loss": 0.8078351,
      "num_input_tokens_seen": 25562990,
      "step": 1203,
      "time_per_iteration": 2.7187600135803223
    },
    {
      "auxiliary_loss_clip": 0.01240407,
      "auxiliary_loss_mlp": 0.01037358,
      "balance_loss_clip": 1.06620598,
      "balance_loss_mlp": 1.02755904,
      "epoch": 0.1447724403294655,
      "flos": 16690705752960.0,
      "grad_norm": 2.561181110356706,
      "language_loss": 0.71076787,
      "learning_rate": 3.863517301805502e-06,
      "loss": 0.73354554,
      "num_input_tokens_seen": 25581380,
      "step": 1204,
      "time_per_iteration": 2.5897533893585205
    },
    {
      "auxiliary_loss_clip": 0.013899,
      "auxiliary_loss_mlp": 0.01045692,
      "balance_loss_clip": 1.06531847,
      "balance_loss_mlp": 1.03548789,
      "epoch": 0.14489268322010462,
      "flos": 20073741321600.0,
      "grad_norm": 10.09589567685479,
      "language_loss": 0.9699474,
      "learning_rate": 3.863234332802196e-06,
      "loss": 0.99430335,
      "num_input_tokens_seen": 25593585,
      "step": 1205,
      "time_per_iteration": 2.7155702114105225
    },
    {
      "auxiliary_loss_clip": 0.01333188,
      "auxiliary_loss_mlp": 0.01040054,
      "balance_loss_clip": 1.06071663,
      "balance_loss_mlp": 1.03090513,
      "epoch": 0.1450129261107437,
      "flos": 27125627955840.0,
      "grad_norm": 2.0610439801734213,
      "language_loss": 0.74060857,
      "learning_rate": 3.862951081147723e-06,
      "loss": 0.764341,
      "num_input_tokens_seen": 25613750,
      "step": 1206,
      "time_per_iteration": 2.71290922164917
    },
    {
      "auxiliary_loss_clip": 0.01289154,
      "auxiliary_loss_mlp": 0.01035815,
      "balance_loss_clip": 1.06605053,
      "balance_loss_mlp": 1.02662373,
      "epoch": 0.1451331690013828,
      "flos": 25702344472320.0,
      "grad_norm": 2.3162500647257684,
      "language_loss": 0.78373778,
      "learning_rate": 3.862667546885053e-06,
      "loss": 0.80698746,
      "num_input_tokens_seen": 25632300,
      "step": 1207,
      "time_per_iteration": 2.7533113956451416
    },
    {
      "auxiliary_loss_clip": 0.01294747,
      "auxiliary_loss_mlp": 0.01036086,
      "balance_loss_clip": 1.06095052,
      "balance_loss_mlp": 1.02632904,
      "epoch": 0.14525341189202187,
      "flos": 25737393168000.0,
      "grad_norm": 1.9553351165239061,
      "language_loss": 0.73854578,
      "learning_rate": 3.8623837300571965e-06,
      "loss": 0.76185411,
      "num_input_tokens_seen": 25651285,
      "step": 1208,
      "time_per_iteration": 2.6811482906341553
    },
    {
      "auxiliary_loss_clip": 0.01237654,
      "auxiliary_loss_mlp": 0.01039142,
      "balance_loss_clip": 1.06615734,
      "balance_loss_mlp": 1.02914584,
      "epoch": 0.14537365478266098,
      "flos": 23073898844160.0,
      "grad_norm": 2.642302963726878,
      "language_loss": 0.84119207,
      "learning_rate": 3.8620996307072085e-06,
      "loss": 0.86396003,
      "num_input_tokens_seen": 25671990,
      "step": 1209,
      "time_per_iteration": 2.6845502853393555
    },
    {
      "auxiliary_loss_clip": 0.01390198,
      "auxiliary_loss_mlp": 0.01035919,
      "balance_loss_clip": 1.05860829,
      "balance_loss_mlp": 1.02662098,
      "epoch": 0.14549389767330007,
      "flos": 20595021448320.0,
      "grad_norm": 1.7730608522890507,
      "language_loss": 0.64036536,
      "learning_rate": 3.861815248878188e-06,
      "loss": 0.66462648,
      "num_input_tokens_seen": 25689475,
      "step": 1210,
      "time_per_iteration": 3.741718053817749
    },
    {
      "auxiliary_loss_clip": 0.01331009,
      "auxiliary_loss_mlp": 0.01036279,
      "balance_loss_clip": 1.06417418,
      "balance_loss_mlp": 1.02659333,
      "epoch": 0.14561414056393915,
      "flos": 15121804533120.0,
      "grad_norm": 3.5015605283598275,
      "language_loss": 0.80084765,
      "learning_rate": 3.861530584613274e-06,
      "loss": 0.82452053,
      "num_input_tokens_seen": 25707475,
      "step": 1211,
      "time_per_iteration": 2.6540687084198
    },
    {
      "auxiliary_loss_clip": 0.01290143,
      "auxiliary_loss_mlp": 0.02585434,
      "balance_loss_clip": 1.06601119,
      "balance_loss_mlp": 1.00063109,
      "epoch": 0.14573438345457826,
      "flos": 19427493778560.0,
      "grad_norm": 4.807352151556901,
      "language_loss": 0.82139975,
      "learning_rate": 3.86124563795565e-06,
      "loss": 0.86015546,
      "num_input_tokens_seen": 25726290,
      "step": 1212,
      "time_per_iteration": 2.7357795238494873
    },
    {
      "auxiliary_loss_clip": 0.012402,
      "auxiliary_loss_mlp": 0.01040501,
      "balance_loss_clip": 1.06979012,
      "balance_loss_mlp": 1.0320785,
      "epoch": 0.14585462634521734,
      "flos": 24828422572800.0,
      "grad_norm": 1.6737852191696254,
      "language_loss": 0.70229781,
      "learning_rate": 3.860960408948543e-06,
      "loss": 0.72510481,
      "num_input_tokens_seen": 25748040,
      "step": 1213,
      "time_per_iteration": 2.7664380073547363
    },
    {
      "auxiliary_loss_clip": 0.01283924,
      "auxiliary_loss_mlp": 0.01031818,
      "balance_loss_clip": 1.06750286,
      "balance_loss_mlp": 1.02298427,
      "epoch": 0.14597486923585642,
      "flos": 15448627405440.0,
      "grad_norm": 3.7078280720642085,
      "language_loss": 0.8970589,
      "learning_rate": 3.860674897635222e-06,
      "loss": 0.92021632,
      "num_input_tokens_seen": 25764525,
      "step": 1214,
      "time_per_iteration": 2.635326385498047
    },
    {
      "auxiliary_loss_clip": 0.01286425,
      "auxiliary_loss_mlp": 0.01035565,
      "balance_loss_clip": 1.06633711,
      "balance_loss_mlp": 1.02594435,
      "epoch": 0.1460951121264955,
      "flos": 16655154266880.0,
      "grad_norm": 2.5095445956774887,
      "language_loss": 0.8354702,
      "learning_rate": 3.860389104058998e-06,
      "loss": 0.85869008,
      "num_input_tokens_seen": 25782755,
      "step": 1215,
      "time_per_iteration": 4.399419546127319
    },
    {
      "auxiliary_loss_clip": 0.01337964,
      "auxiliary_loss_mlp": 0.01038921,
      "balance_loss_clip": 1.06399441,
      "balance_loss_mlp": 1.02915156,
      "epoch": 0.14621535501713462,
      "flos": 24863291700480.0,
      "grad_norm": 2.1435437670291035,
      "language_loss": 0.73102176,
      "learning_rate": 3.860103028263227e-06,
      "loss": 0.7547906,
      "num_input_tokens_seen": 25805860,
      "step": 1216,
      "time_per_iteration": 2.699784994125366
    },
    {
      "auxiliary_loss_clip": 0.01431964,
      "auxiliary_loss_mlp": 0.01040365,
      "balance_loss_clip": 1.05524695,
      "balance_loss_mlp": 1.03151965,
      "epoch": 0.1463355979077737,
      "flos": 25228000442880.0,
      "grad_norm": 2.1355929116300127,
      "language_loss": 0.69877476,
      "learning_rate": 3.859816670291304e-06,
      "loss": 0.72349805,
      "num_input_tokens_seen": 25824955,
      "step": 1217,
      "time_per_iteration": 2.819456100463867
    },
    {
      "auxiliary_loss_clip": 0.01473884,
      "auxiliary_loss_mlp": 0.01052378,
      "balance_loss_clip": 1.05504346,
      "balance_loss_mlp": 1.04328883,
      "epoch": 0.14645584079841278,
      "flos": 22054143726720.0,
      "grad_norm": 2.2332065926758875,
      "language_loss": 0.90002853,
      "learning_rate": 3.859530030186672e-06,
      "loss": 0.92529118,
      "num_input_tokens_seen": 25841965,
      "step": 1218,
      "time_per_iteration": 3.7252860069274902
    },
    {
      "auxiliary_loss_clip": 0.01336165,
      "auxiliary_loss_mlp": 0.01036646,
      "balance_loss_clip": 1.06451321,
      "balance_loss_mlp": 1.02765751,
      "epoch": 0.1465760836890519,
      "flos": 23623870959360.0,
      "grad_norm": 2.390659511522882,
      "language_loss": 0.83282697,
      "learning_rate": 3.859243107992813e-06,
      "loss": 0.85655504,
      "num_input_tokens_seen": 25860770,
      "step": 1219,
      "time_per_iteration": 2.7622504234313965
    },
    {
      "auxiliary_loss_clip": 0.01387028,
      "auxiliary_loss_mlp": 0.01040941,
      "balance_loss_clip": 1.05645728,
      "balance_loss_mlp": 1.03146958,
      "epoch": 0.14669632657969098,
      "flos": 37407893356800.0,
      "grad_norm": 3.7430065036172753,
      "language_loss": 0.78426373,
      "learning_rate": 3.858955903753252e-06,
      "loss": 0.80854344,
      "num_input_tokens_seen": 25879410,
      "step": 1220,
      "time_per_iteration": 2.8809471130371094
    },
    {
      "auxiliary_loss_clip": 0.01289125,
      "auxiliary_loss_mlp": 0.01042682,
      "balance_loss_clip": 1.06560457,
      "balance_loss_mlp": 1.03398573,
      "epoch": 0.14681656947033006,
      "flos": 28365910623360.0,
      "grad_norm": 1.6013364309516696,
      "language_loss": 0.83576238,
      "learning_rate": 3.858668417511559e-06,
      "loss": 0.85908043,
      "num_input_tokens_seen": 25902160,
      "step": 1221,
      "time_per_iteration": 2.716440200805664
    },
    {
      "auxiliary_loss_clip": 0.01291286,
      "auxiliary_loss_mlp": 0.01041402,
      "balance_loss_clip": 1.06487775,
      "balance_loss_mlp": 1.03263974,
      "epoch": 0.14693681236096917,
      "flos": 18479488078080.0,
      "grad_norm": 4.524548881011971,
      "language_loss": 0.76439846,
      "learning_rate": 3.8583806493113445e-06,
      "loss": 0.78772527,
      "num_input_tokens_seen": 25920505,
      "step": 1222,
      "time_per_iteration": 2.7597532272338867
    },
    {
      "auxiliary_loss_clip": 0.01285186,
      "auxiliary_loss_mlp": 0.01036044,
      "balance_loss_clip": 1.06541753,
      "balance_loss_mlp": 1.02581024,
      "epoch": 0.14705705525160825,
      "flos": 20777806782720.0,
      "grad_norm": 2.9250873667435044,
      "language_loss": 0.82193816,
      "learning_rate": 3.858092599196263e-06,
      "loss": 0.84515047,
      "num_input_tokens_seen": 25938460,
      "step": 1223,
      "time_per_iteration": 2.6675233840942383
    },
    {
      "auxiliary_loss_clip": 0.01284022,
      "auxiliary_loss_mlp": 0.0103671,
      "balance_loss_clip": 1.06514502,
      "balance_loss_mlp": 1.0274241,
      "epoch": 0.14717729814224734,
      "flos": 29932944336000.0,
      "grad_norm": 2.2231358249103628,
      "language_loss": 0.82467389,
      "learning_rate": 3.857804267210012e-06,
      "loss": 0.8478812,
      "num_input_tokens_seen": 25957760,
      "step": 1224,
      "time_per_iteration": 2.7411625385284424
    },
    {
      "auxiliary_loss_clip": 0.01375078,
      "auxiliary_loss_mlp": 0.01036827,
      "balance_loss_clip": 1.05695152,
      "balance_loss_mlp": 1.02770734,
      "epoch": 0.14729754103288642,
      "flos": 20047491457920.0,
      "grad_norm": 1.9654467995624836,
      "language_loss": 0.88021928,
      "learning_rate": 3.857515653396331e-06,
      "loss": 0.9043383,
      "num_input_tokens_seen": 25974970,
      "step": 1225,
      "time_per_iteration": 2.8067467212677
    },
    {
      "auxiliary_loss_clip": 0.01432606,
      "auxiliary_loss_mlp": 0.01035694,
      "balance_loss_clip": 1.0593617,
      "balance_loss_mlp": 1.0266943,
      "epoch": 0.14741778392352553,
      "flos": 19281516906240.0,
      "grad_norm": 3.0240811004013146,
      "language_loss": 0.8740201,
      "learning_rate": 3.857226757799002e-06,
      "loss": 0.8987031,
      "num_input_tokens_seen": 25992525,
      "step": 1226,
      "time_per_iteration": 2.808821439743042
    },
    {
      "auxiliary_loss_clip": 0.01331745,
      "auxiliary_loss_mlp": 0.0103507,
      "balance_loss_clip": 1.06107366,
      "balance_loss_mlp": 1.02593255,
      "epoch": 0.1475380268141646,
      "flos": 25411108999680.0,
      "grad_norm": 2.5126777809336196,
      "language_loss": 0.74403095,
      "learning_rate": 3.85693758046185e-06,
      "loss": 0.76769912,
      "num_input_tokens_seen": 26010815,
      "step": 1227,
      "time_per_iteration": 2.684375524520874
    },
    {
      "auxiliary_loss_clip": 0.01240143,
      "auxiliary_loss_mlp": 0.01035665,
      "balance_loss_clip": 1.07155538,
      "balance_loss_mlp": 1.02684903,
      "epoch": 0.1476582697048037,
      "flos": 20847652778880.0,
      "grad_norm": 1.8123919454370871,
      "language_loss": 0.82718837,
      "learning_rate": 3.8566481214287435e-06,
      "loss": 0.8499465,
      "num_input_tokens_seen": 26028935,
      "step": 1228,
      "time_per_iteration": 2.6630842685699463
    },
    {
      "auxiliary_loss_clip": 0.01378927,
      "auxiliary_loss_mlp": 0.01040086,
      "balance_loss_clip": 1.05604076,
      "balance_loss_mlp": 1.03160405,
      "epoch": 0.1477785125954428,
      "flos": 14028109269120.0,
      "grad_norm": 2.131197169845581,
      "language_loss": 0.90740615,
      "learning_rate": 3.8563583807435935e-06,
      "loss": 0.93159628,
      "num_input_tokens_seen": 26045080,
      "step": 1229,
      "time_per_iteration": 2.6914825439453125
    },
    {
      "auxiliary_loss_clip": 0.01284137,
      "auxiliary_loss_mlp": 0.025789,
      "balance_loss_clip": 1.0639894,
      "balance_loss_mlp": 1.00055838,
      "epoch": 0.1478987554860819,
      "flos": 20516699842560.0,
      "grad_norm": 2.136072441899214,
      "language_loss": 0.77680862,
      "learning_rate": 3.856068358450353e-06,
      "loss": 0.81543899,
      "num_input_tokens_seen": 26065030,
      "step": 1230,
      "time_per_iteration": 2.6888210773468018
    },
    {
      "auxiliary_loss_clip": 0.01331095,
      "auxiliary_loss_mlp": 0.0103843,
      "balance_loss_clip": 1.06599247,
      "balance_loss_mlp": 1.02911973,
      "epoch": 0.14801899837672097,
      "flos": 17857012360320.0,
      "grad_norm": 2.478797027916171,
      "language_loss": 0.86279243,
      "learning_rate": 3.8557780545930186e-06,
      "loss": 0.88648766,
      "num_input_tokens_seen": 26083445,
      "step": 1231,
      "time_per_iteration": 2.8432765007019043
    },
    {
      "auxiliary_loss_clip": 0.01328873,
      "auxiliary_loss_mlp": 0.01034752,
      "balance_loss_clip": 1.06472468,
      "balance_loss_mlp": 1.02556145,
      "epoch": 0.14813924126736006,
      "flos": 20881408584960.0,
      "grad_norm": 2.1068793273522757,
      "language_loss": 0.79667693,
      "learning_rate": 3.855487469215628e-06,
      "loss": 0.82031322,
      "num_input_tokens_seen": 26102375,
      "step": 1232,
      "time_per_iteration": 2.71262526512146
    },
    {
      "auxiliary_loss_clip": 0.01382615,
      "auxiliary_loss_mlp": 0.010288,
      "balance_loss_clip": 1.06217873,
      "balance_loss_mlp": 1.02003813,
      "epoch": 0.14825948415799917,
      "flos": 37414070496000.0,
      "grad_norm": 2.374938820733192,
      "language_loss": 0.72836989,
      "learning_rate": 3.855196602362264e-06,
      "loss": 0.75248408,
      "num_input_tokens_seen": 26125295,
      "step": 1233,
      "time_per_iteration": 2.9065310955047607
    },
    {
      "auxiliary_loss_clip": 0.01286021,
      "auxiliary_loss_mlp": 0.01033669,
      "balance_loss_clip": 1.06385767,
      "balance_loss_mlp": 1.0247283,
      "epoch": 0.14837972704863825,
      "flos": 22014641744640.0,
      "grad_norm": 2.2281952228649615,
      "language_loss": 0.94359267,
      "learning_rate": 3.854905454077051e-06,
      "loss": 0.96678954,
      "num_input_tokens_seen": 26142905,
      "step": 1234,
      "time_per_iteration": 2.662846565246582
    },
    {
      "auxiliary_loss_clip": 0.01531749,
      "auxiliary_loss_mlp": 0.01031093,
      "balance_loss_clip": 1.05383658,
      "balance_loss_mlp": 1.02128839,
      "epoch": 0.14849996993927733,
      "flos": 20996323171200.0,
      "grad_norm": 1.8989005867453117,
      "language_loss": 0.88522178,
      "learning_rate": 3.854614024404155e-06,
      "loss": 0.91085017,
      "num_input_tokens_seen": 26161215,
      "step": 1235,
      "time_per_iteration": 2.954106092453003
    },
    {
      "auxiliary_loss_clip": 0.01320923,
      "auxiliary_loss_mlp": 0.010311,
      "balance_loss_clip": 1.05892706,
      "balance_loss_mlp": 1.02184939,
      "epoch": 0.14862021282991644,
      "flos": 20047994248320.0,
      "grad_norm": 1.9356969094457368,
      "language_loss": 0.89508677,
      "learning_rate": 3.8543223133877865e-06,
      "loss": 0.918607,
      "num_input_tokens_seen": 26179810,
      "step": 1236,
      "time_per_iteration": 3.817535161972046
    },
    {
      "auxiliary_loss_clip": 0.01321774,
      "auxiliary_loss_mlp": 0.01035339,
      "balance_loss_clip": 1.05684519,
      "balance_loss_mlp": 1.02602863,
      "epoch": 0.14874045572055553,
      "flos": 22712027276160.0,
      "grad_norm": 1.9616459328957105,
      "language_loss": 0.88229436,
      "learning_rate": 3.854030321072198e-06,
      "loss": 0.90586555,
      "num_input_tokens_seen": 26199715,
      "step": 1237,
      "time_per_iteration": 2.6981470584869385
    },
    {
      "auxiliary_loss_clip": 0.01386768,
      "auxiliary_loss_mlp": 0.01041115,
      "balance_loss_clip": 1.05939555,
      "balance_loss_mlp": 1.03241873,
      "epoch": 0.1488606986111946,
      "flos": 25411288567680.0,
      "grad_norm": 1.9467577296585745,
      "language_loss": 0.73856282,
      "learning_rate": 3.853738047501682e-06,
      "loss": 0.76284158,
      "num_input_tokens_seen": 26220275,
      "step": 1238,
      "time_per_iteration": 2.797975540161133
    },
    {
      "auxiliary_loss_clip": 0.01285931,
      "auxiliary_loss_mlp": 0.01035936,
      "balance_loss_clip": 1.06816077,
      "balance_loss_mlp": 1.02632177,
      "epoch": 0.1489809415018337,
      "flos": 17018749687680.0,
      "grad_norm": 3.1553173087265423,
      "language_loss": 0.77528411,
      "learning_rate": 3.85344549272058e-06,
      "loss": 0.7985028,
      "num_input_tokens_seen": 26238255,
      "step": 1239,
      "time_per_iteration": 2.6182901859283447
    },
    {
      "auxiliary_loss_clip": 0.01282754,
      "auxiliary_loss_mlp": 0.01040567,
      "balance_loss_clip": 1.06411147,
      "balance_loss_mlp": 1.03111982,
      "epoch": 0.1491011843924728,
      "flos": 33659394860160.0,
      "grad_norm": 1.8313018169484425,
      "language_loss": 0.82387269,
      "learning_rate": 3.853152656773269e-06,
      "loss": 0.84710586,
      "num_input_tokens_seen": 26259690,
      "step": 1240,
      "time_per_iteration": 3.6577577590942383
    },
    {
      "auxiliary_loss_clip": 0.01328274,
      "auxiliary_loss_mlp": 0.01037213,
      "balance_loss_clip": 1.0633142,
      "balance_loss_mlp": 1.02846956,
      "epoch": 0.14922142728311188,
      "flos": 21179000764800.0,
      "grad_norm": 2.1634207404515347,
      "language_loss": 0.85119659,
      "learning_rate": 3.852859539704174e-06,
      "loss": 0.87485141,
      "num_input_tokens_seen": 26278990,
      "step": 1241,
      "time_per_iteration": 3.788844108581543
    },
    {
      "auxiliary_loss_clip": 0.01425485,
      "auxiliary_loss_mlp": 0.01036267,
      "balance_loss_clip": 1.05336213,
      "balance_loss_mlp": 1.02637863,
      "epoch": 0.14934167017375097,
      "flos": 29860548474240.0,
      "grad_norm": 2.077542323553634,
      "language_loss": 0.76443899,
      "learning_rate": 3.85256614155776e-06,
      "loss": 0.78905654,
      "num_input_tokens_seen": 26299120,
      "step": 1242,
      "time_per_iteration": 2.8875086307525635
    },
    {
      "auxiliary_loss_clip": 0.01281766,
      "auxiliary_loss_mlp": 0.01030474,
      "balance_loss_clip": 1.06191826,
      "balance_loss_mlp": 1.02115774,
      "epoch": 0.14946191306439008,
      "flos": 17019216564480.0,
      "grad_norm": 2.463049957009014,
      "language_loss": 0.74444985,
      "learning_rate": 3.852272462378535e-06,
      "loss": 0.76757222,
      "num_input_tokens_seen": 26316995,
      "step": 1243,
      "time_per_iteration": 2.736884593963623
    },
    {
      "auxiliary_loss_clip": 0.01330488,
      "auxiliary_loss_mlp": 0.01037703,
      "balance_loss_clip": 1.06280923,
      "balance_loss_mlp": 1.02846384,
      "epoch": 0.14958215595502916,
      "flos": 15669047214720.0,
      "grad_norm": 2.0269881645736425,
      "language_loss": 0.77646244,
      "learning_rate": 3.85197850221105e-06,
      "loss": 0.80014443,
      "num_input_tokens_seen": 26333295,
      "step": 1244,
      "time_per_iteration": 3.6059176921844482
    },
    {
      "auxiliary_loss_clip": 0.01282137,
      "auxiliary_loss_mlp": 0.01041973,
      "balance_loss_clip": 1.06547904,
      "balance_loss_mlp": 1.03369963,
      "epoch": 0.14970239884566824,
      "flos": 33108560818560.0,
      "grad_norm": 1.8513158643483836,
      "language_loss": 0.75996572,
      "learning_rate": 3.851684261099899e-06,
      "loss": 0.78320682,
      "num_input_tokens_seen": 26355035,
      "step": 1245,
      "time_per_iteration": 2.7528088092803955
    },
    {
      "auxiliary_loss_clip": 0.01284781,
      "auxiliary_loss_mlp": 0.01032778,
      "balance_loss_clip": 1.06150281,
      "balance_loss_mlp": 1.02361703,
      "epoch": 0.14982264173630733,
      "flos": 17821245392640.0,
      "grad_norm": 1.8973559913669202,
      "language_loss": 0.8675645,
      "learning_rate": 3.851389739089718e-06,
      "loss": 0.89074004,
      "num_input_tokens_seen": 26371655,
      "step": 1246,
      "time_per_iteration": 2.8658058643341064
    },
    {
      "auxiliary_loss_clip": 0.01289471,
      "auxiliary_loss_mlp": 0.0103602,
      "balance_loss_clip": 1.06908345,
      "balance_loss_mlp": 1.02660811,
      "epoch": 0.14994288462694644,
      "flos": 32409559175040.0,
      "grad_norm": 3.290388447774354,
      "language_loss": 0.80188715,
      "learning_rate": 3.851094936225186e-06,
      "loss": 0.82514203,
      "num_input_tokens_seen": 26392540,
      "step": 1247,
      "time_per_iteration": 2.7324016094207764
    },
    {
      "auxiliary_loss_clip": 0.01327916,
      "auxiliary_loss_mlp": 0.01031666,
      "balance_loss_clip": 1.0635041,
      "balance_loss_mlp": 1.02302909,
      "epoch": 0.15006312751758552,
      "flos": 31794661226880.0,
      "grad_norm": 2.8517386515997267,
      "language_loss": 0.76600474,
      "learning_rate": 3.850799852551024e-06,
      "loss": 0.78960061,
      "num_input_tokens_seen": 26414960,
      "step": 1248,
      "time_per_iteration": 2.8486580848693848
    },
    {
      "auxiliary_loss_clip": 0.01277687,
      "auxiliary_loss_mlp": 0.0103293,
      "balance_loss_clip": 1.0611825,
      "balance_loss_mlp": 1.02412629,
      "epoch": 0.1501833704082246,
      "flos": 16618022582400.0,
      "grad_norm": 2.3131962230661958,
      "language_loss": 0.86290193,
      "learning_rate": 3.850504488111995e-06,
      "loss": 0.88600808,
      "num_input_tokens_seen": 26431635,
      "step": 1249,
      "time_per_iteration": 2.6721627712249756
    },
    {
      "auxiliary_loss_clip": 0.01327197,
      "auxiliary_loss_mlp": 0.01037162,
      "balance_loss_clip": 1.06218314,
      "balance_loss_mlp": 1.02846539,
      "epoch": 0.15030361329886371,
      "flos": 23471178243840.0,
      "grad_norm": 2.3470908732082436,
      "language_loss": 0.82672793,
      "learning_rate": 3.850208842952907e-06,
      "loss": 0.85037148,
      "num_input_tokens_seen": 26450440,
      "step": 1250,
      "time_per_iteration": 2.7788538932800293
    },
    {
      "auxiliary_loss_clip": 0.01442084,
      "auxiliary_loss_mlp": 0.01031407,
      "balance_loss_clip": 1.05891192,
      "balance_loss_mlp": 1.02289534,
      "epoch": 0.1504238561895028,
      "flos": 25629409906560.0,
      "grad_norm": 2.2925063004402197,
      "language_loss": 0.79170775,
      "learning_rate": 3.849912917118608e-06,
      "loss": 0.81644261,
      "num_input_tokens_seen": 26471480,
      "step": 1251,
      "time_per_iteration": 2.8455350399017334
    },
    {
      "auxiliary_loss_clip": 0.01205944,
      "auxiliary_loss_mlp": 0.01002563,
      "balance_loss_clip": 1.04749799,
      "balance_loss_mlp": 1.00059569,
      "epoch": 0.15054409908014188,
      "flos": 52095146129280.0,
      "grad_norm": 0.8693981341527335,
      "language_loss": 0.5925225,
      "learning_rate": 3.849616710653992e-06,
      "loss": 0.61460757,
      "num_input_tokens_seen": 26532950,
      "step": 1252,
      "time_per_iteration": 3.25661039352417
    },
    {
      "auxiliary_loss_clip": 0.01278964,
      "auxiliary_loss_mlp": 0.01032178,
      "balance_loss_clip": 1.0639286,
      "balance_loss_mlp": 1.02318358,
      "epoch": 0.150664341970781,
      "flos": 18880251096960.0,
      "grad_norm": 1.7104668879663563,
      "language_loss": 0.74536121,
      "learning_rate": 3.84932022360399e-06,
      "loss": 0.76847267,
      "num_input_tokens_seen": 26551615,
      "step": 1253,
      "time_per_iteration": 2.7308239936828613
    },
    {
      "auxiliary_loss_clip": 0.01331268,
      "auxiliary_loss_mlp": 0.01034415,
      "balance_loss_clip": 1.06563735,
      "balance_loss_mlp": 1.02553415,
      "epoch": 0.15078458486142007,
      "flos": 22163240309760.0,
      "grad_norm": 3.394743881819202,
      "language_loss": 0.85095757,
      "learning_rate": 3.849023456013581e-06,
      "loss": 0.87461436,
      "num_input_tokens_seen": 26569175,
      "step": 1254,
      "time_per_iteration": 2.7142319679260254
    },
    {
      "auxiliary_loss_clip": 0.01290243,
      "auxiliary_loss_mlp": 0.01033224,
      "balance_loss_clip": 1.06519735,
      "balance_loss_mlp": 1.02459288,
      "epoch": 0.15090482775205916,
      "flos": 26651894457600.0,
      "grad_norm": 4.932024067200915,
      "language_loss": 0.62132788,
      "learning_rate": 3.848726407927784e-06,
      "loss": 0.6445626,
      "num_input_tokens_seen": 26589560,
      "step": 1255,
      "time_per_iteration": 2.739807367324829
    },
    {
      "auxiliary_loss_clip": 0.01332685,
      "auxiliary_loss_mlp": 0.01037767,
      "balance_loss_clip": 1.06440568,
      "balance_loss_mlp": 1.02904654,
      "epoch": 0.15102507064269824,
      "flos": 21798998444160.0,
      "grad_norm": 2.3848949943682616,
      "language_loss": 0.8636992,
      "learning_rate": 3.84842907939166e-06,
      "loss": 0.88740373,
      "num_input_tokens_seen": 26608785,
      "step": 1256,
      "time_per_iteration": 2.7361133098602295
    },
    {
      "auxiliary_loss_clip": 0.0137436,
      "auxiliary_loss_mlp": 0.01029942,
      "balance_loss_clip": 1.05844975,
      "balance_loss_mlp": 1.02124608,
      "epoch": 0.15114531353333735,
      "flos": 22820908377600.0,
      "grad_norm": 2.9445732153360287,
      "language_loss": 0.71107441,
      "learning_rate": 3.8481314704503146e-06,
      "loss": 0.73511744,
      "num_input_tokens_seen": 26628615,
      "step": 1257,
      "time_per_iteration": 2.838121175765991
    },
    {
      "auxiliary_loss_clip": 0.01281888,
      "auxiliary_loss_mlp": 0.01038625,
      "balance_loss_clip": 1.0687418,
      "balance_loss_mlp": 1.02962422,
      "epoch": 0.15126555642397643,
      "flos": 19682674974720.0,
      "grad_norm": 2.2248988210543548,
      "language_loss": 0.87907481,
      "learning_rate": 3.847833581148895e-06,
      "loss": 0.90227991,
      "num_input_tokens_seen": 26647525,
      "step": 1258,
      "time_per_iteration": 2.712095260620117
    },
    {
      "auxiliary_loss_clip": 0.01232326,
      "auxiliary_loss_mlp": 0.01035707,
      "balance_loss_clip": 1.06383932,
      "balance_loss_mlp": 1.02629519,
      "epoch": 0.15138579931461552,
      "flos": 28726022424960.0,
      "grad_norm": 2.2883283198862077,
      "language_loss": 0.81124437,
      "learning_rate": 3.84753541153259e-06,
      "loss": 0.83392471,
      "num_input_tokens_seen": 26667095,
      "step": 1259,
      "time_per_iteration": 2.6916379928588867
    },
    {
      "auxiliary_loss_clip": 0.01287838,
      "auxiliary_loss_mlp": 0.0103804,
      "balance_loss_clip": 1.06967163,
      "balance_loss_mlp": 1.02834833,
      "epoch": 0.15150604220525463,
      "flos": 22127006465280.0,
      "grad_norm": 1.8718624313807875,
      "language_loss": 0.8314001,
      "learning_rate": 3.847236961646633e-06,
      "loss": 0.85465896,
      "num_input_tokens_seen": 26686075,
      "step": 1260,
      "time_per_iteration": 2.683565855026245
    },
    {
      "auxiliary_loss_clip": 0.01329639,
      "auxiliary_loss_mlp": 0.01035362,
      "balance_loss_clip": 1.06103563,
      "balance_loss_mlp": 1.02615321,
      "epoch": 0.1516262850958937,
      "flos": 12968708515200.0,
      "grad_norm": 3.4190190394145135,
      "language_loss": 0.7810303,
      "learning_rate": 3.846938231536296e-06,
      "loss": 0.80468029,
      "num_input_tokens_seen": 26701695,
      "step": 1261,
      "time_per_iteration": 2.6184017658233643
    },
    {
      "auxiliary_loss_clip": 0.01290661,
      "auxiliary_loss_mlp": 0.01047674,
      "balance_loss_clip": 1.07206845,
      "balance_loss_mlp": 1.03863204,
      "epoch": 0.1517465279865328,
      "flos": 21797130936960.0,
      "grad_norm": 2.387667107899127,
      "language_loss": 0.80802053,
      "learning_rate": 3.8466392212468995e-06,
      "loss": 0.83140385,
      "num_input_tokens_seen": 26721885,
      "step": 1262,
      "time_per_iteration": 2.6544251441955566
    },
    {
      "auxiliary_loss_clip": 0.01251422,
      "auxiliary_loss_mlp": 0.01002828,
      "balance_loss_clip": 1.04331076,
      "balance_loss_mlp": 1.00070596,
      "epoch": 0.15186677087717187,
      "flos": 58174569901440.0,
      "grad_norm": 0.8193800226589002,
      "language_loss": 0.61911356,
      "learning_rate": 3.8463399308238e-06,
      "loss": 0.64165604,
      "num_input_tokens_seen": 26780990,
      "step": 1263,
      "time_per_iteration": 4.148198366165161
    },
    {
      "auxiliary_loss_clip": 0.01285216,
      "auxiliary_loss_mlp": 0.01040149,
      "balance_loss_clip": 1.06724072,
      "balance_loss_mlp": 1.0310595,
      "epoch": 0.15198701376781099,
      "flos": 32669696448000.0,
      "grad_norm": 4.934203572011821,
      "language_loss": 0.64282298,
      "learning_rate": 3.846040360312402e-06,
      "loss": 0.6660766,
      "num_input_tokens_seen": 26804250,
      "step": 1264,
      "time_per_iteration": 2.7288947105407715
    },
    {
      "auxiliary_loss_clip": 0.01233709,
      "auxiliary_loss_mlp": 0.01039396,
      "balance_loss_clip": 1.06581473,
      "balance_loss_mlp": 1.03021121,
      "epoch": 0.15210725665845007,
      "flos": 28402575431040.0,
      "grad_norm": 2.632003086834704,
      "language_loss": 0.81144047,
      "learning_rate": 3.8457405097581485e-06,
      "loss": 0.83417153,
      "num_input_tokens_seen": 26823240,
      "step": 1265,
      "time_per_iteration": 2.637934923171997
    },
    {
      "auxiliary_loss_clip": 0.0138376,
      "auxiliary_loss_mlp": 0.0103866,
      "balance_loss_clip": 1.05629253,
      "balance_loss_mlp": 1.02924848,
      "epoch": 0.15222749954908915,
      "flos": 19938179393280.0,
      "grad_norm": 1.820237649603486,
      "language_loss": 0.7818346,
      "learning_rate": 3.8454403792065275e-06,
      "loss": 0.80605888,
      "num_input_tokens_seen": 26842060,
      "step": 1266,
      "time_per_iteration": 2.796902894973755
    },
    {
      "auxiliary_loss_clip": 0.01373436,
      "auxiliary_loss_mlp": 0.01035947,
      "balance_loss_clip": 1.06015229,
      "balance_loss_mlp": 1.02698231,
      "epoch": 0.15234774243972826,
      "flos": 21324223451520.0,
      "grad_norm": 2.4353702546048805,
      "language_loss": 0.85318828,
      "learning_rate": 3.845139968703068e-06,
      "loss": 0.87728208,
      "num_input_tokens_seen": 26859580,
      "step": 1267,
      "time_per_iteration": 4.460983991622925
    },
    {
      "auxiliary_loss_clip": 0.01429193,
      "auxiliary_loss_mlp": 0.01036547,
      "balance_loss_clip": 1.05777586,
      "balance_loss_mlp": 1.02734411,
      "epoch": 0.15246798533036734,
      "flos": 25957812977280.0,
      "grad_norm": 2.3088722542902422,
      "language_loss": 0.8301543,
      "learning_rate": 3.844839278293342e-06,
      "loss": 0.85481167,
      "num_input_tokens_seen": 26880430,
      "step": 1268,
      "time_per_iteration": 2.7997350692749023
    },
    {
      "auxiliary_loss_clip": 0.01235774,
      "auxiliary_loss_mlp": 0.01038632,
      "balance_loss_clip": 1.06774616,
      "balance_loss_mlp": 1.02927959,
      "epoch": 0.15258822822100643,
      "flos": 25811907932160.0,
      "grad_norm": 2.300275896209008,
      "language_loss": 0.76704943,
      "learning_rate": 3.8445383080229654e-06,
      "loss": 0.78979349,
      "num_input_tokens_seen": 26896445,
      "step": 1269,
      "time_per_iteration": 2.6947295665740967
    },
    {
      "auxiliary_loss_clip": 0.01323038,
      "auxiliary_loss_mlp": 0.01030913,
      "balance_loss_clip": 1.05876672,
      "balance_loss_mlp": 1.02171612,
      "epoch": 0.1527084711116455,
      "flos": 25265455349760.0,
      "grad_norm": 2.192400280543031,
      "language_loss": 0.73582244,
      "learning_rate": 3.844237057937593e-06,
      "loss": 0.75936198,
      "num_input_tokens_seen": 26915450,
      "step": 1270,
      "time_per_iteration": 3.7945406436920166
    },
    {
      "auxiliary_loss_clip": 0.01287719,
      "auxiliary_loss_mlp": 0.01032413,
      "balance_loss_clip": 1.06391239,
      "balance_loss_mlp": 1.0231086,
      "epoch": 0.15282871400228462,
      "flos": 29240227572480.0,
      "grad_norm": 3.6322052938273752,
      "language_loss": 0.77863926,
      "learning_rate": 3.843935528082926e-06,
      "loss": 0.8018406,
      "num_input_tokens_seen": 26936475,
      "step": 1271,
      "time_per_iteration": 2.722851037979126
    },
    {
      "auxiliary_loss_clip": 0.01286608,
      "auxiliary_loss_mlp": 0.01031483,
      "balance_loss_clip": 1.06618452,
      "balance_loss_mlp": 1.02207696,
      "epoch": 0.1529489568929237,
      "flos": 20882952869760.0,
      "grad_norm": 2.1454606462957972,
      "language_loss": 0.85204029,
      "learning_rate": 3.843633718504704e-06,
      "loss": 0.87522119,
      "num_input_tokens_seen": 26954920,
      "step": 1272,
      "time_per_iteration": 2.7190608978271484
    },
    {
      "auxiliary_loss_clip": 0.01385459,
      "auxiliary_loss_mlp": 0.01034021,
      "balance_loss_clip": 1.06245196,
      "balance_loss_mlp": 1.02550375,
      "epoch": 0.1530691997835628,
      "flos": 20083833043200.0,
      "grad_norm": 2.828490769257473,
      "language_loss": 0.9030388,
      "learning_rate": 3.843331629248715e-06,
      "loss": 0.92723358,
      "num_input_tokens_seen": 26972520,
      "step": 1273,
      "time_per_iteration": 2.7166967391967773
    },
    {
      "auxiliary_loss_clip": 0.01235482,
      "auxiliary_loss_mlp": 0.01029961,
      "balance_loss_clip": 1.0691092,
      "balance_loss_mlp": 1.02097893,
      "epoch": 0.1531894426742019,
      "flos": 28759814144640.0,
      "grad_norm": 10.010637470633105,
      "language_loss": 0.76398003,
      "learning_rate": 3.843029260360782e-06,
      "loss": 0.78663445,
      "num_input_tokens_seen": 26990890,
      "step": 1274,
      "time_per_iteration": 2.755166530609131
    },
    {
      "auxiliary_loss_clip": 0.01281336,
      "auxiliary_loss_mlp": 0.01037585,
      "balance_loss_clip": 1.06740057,
      "balance_loss_mlp": 1.02878165,
      "epoch": 0.15330968556484098,
      "flos": 22236282616320.0,
      "grad_norm": 1.9089467747231852,
      "language_loss": 0.78815025,
      "learning_rate": 3.8427266118867755e-06,
      "loss": 0.8113395,
      "num_input_tokens_seen": 27010640,
      "step": 1275,
      "time_per_iteration": 2.6692371368408203
    },
    {
      "auxiliary_loss_clip": 0.01331817,
      "auxiliary_loss_mlp": 0.01035359,
      "balance_loss_clip": 1.06432247,
      "balance_loss_mlp": 1.02583456,
      "epoch": 0.15342992845548006,
      "flos": 27527504296320.0,
      "grad_norm": 2.0310424635959947,
      "language_loss": 0.82980818,
      "learning_rate": 3.842423683872608e-06,
      "loss": 0.85347992,
      "num_input_tokens_seen": 27031215,
      "step": 1276,
      "time_per_iteration": 2.771913528442383
    },
    {
      "auxiliary_loss_clip": 0.01283903,
      "auxiliary_loss_mlp": 0.01037528,
      "balance_loss_clip": 1.06307197,
      "balance_loss_mlp": 1.02809858,
      "epoch": 0.15355017134611917,
      "flos": 19609596754560.0,
      "grad_norm": 2.465832112917977,
      "language_loss": 0.77809888,
      "learning_rate": 3.842120476364232e-06,
      "loss": 0.80131316,
      "num_input_tokens_seen": 27049665,
      "step": 1277,
      "time_per_iteration": 2.683453321456909
    },
    {
      "auxiliary_loss_clip": 0.01290452,
      "auxiliary_loss_mlp": 0.01034119,
      "balance_loss_clip": 1.06536198,
      "balance_loss_mlp": 1.02476704,
      "epoch": 0.15367041423675826,
      "flos": 18478590238080.0,
      "grad_norm": 2.064162225671212,
      "language_loss": 0.84058762,
      "learning_rate": 3.841816989407644e-06,
      "loss": 0.86383337,
      "num_input_tokens_seen": 27065155,
      "step": 1278,
      "time_per_iteration": 2.622837543487549
    },
    {
      "auxiliary_loss_clip": 0.01378981,
      "auxiliary_loss_mlp": 0.01034266,
      "balance_loss_clip": 1.05927801,
      "balance_loss_mlp": 1.02505708,
      "epoch": 0.15379065712739734,
      "flos": 41427662342400.0,
      "grad_norm": 2.6658999650565134,
      "language_loss": 0.76535511,
      "learning_rate": 3.841513223048884e-06,
      "loss": 0.7894876,
      "num_input_tokens_seen": 27085840,
      "step": 1279,
      "time_per_iteration": 2.9073240756988525
    },
    {
      "auxiliary_loss_clip": 0.01378031,
      "auxiliary_loss_mlp": 0.01031041,
      "balance_loss_clip": 1.05942321,
      "balance_loss_mlp": 1.02187932,
      "epoch": 0.15391090001803642,
      "flos": 22054215553920.0,
      "grad_norm": 2.2025093539821516,
      "language_loss": 0.78705734,
      "learning_rate": 3.841209177334031e-06,
      "loss": 0.81114805,
      "num_input_tokens_seen": 27104200,
      "step": 1280,
      "time_per_iteration": 2.760091781616211
    },
    {
      "auxiliary_loss_clip": 0.01276602,
      "auxiliary_loss_mlp": 0.01032171,
      "balance_loss_clip": 1.06195199,
      "balance_loss_mlp": 1.02324235,
      "epoch": 0.15403114290867553,
      "flos": 15450351258240.0,
      "grad_norm": 1.809465540091898,
      "language_loss": 0.75136304,
      "learning_rate": 3.84090485230921e-06,
      "loss": 0.77445078,
      "num_input_tokens_seen": 27122440,
      "step": 1281,
      "time_per_iteration": 2.6565163135528564
    },
    {
      "auxiliary_loss_clip": 0.01232905,
      "auxiliary_loss_mlp": 0.01036367,
      "balance_loss_clip": 1.06640267,
      "balance_loss_mlp": 1.02697909,
      "epoch": 0.15415138579931462,
      "flos": 17929156826880.0,
      "grad_norm": 2.637601227383634,
      "language_loss": 0.76425481,
      "learning_rate": 3.840600248020588e-06,
      "loss": 0.78694749,
      "num_input_tokens_seen": 27139380,
      "step": 1282,
      "time_per_iteration": 2.617185592651367
    },
    {
      "auxiliary_loss_clip": 0.01284536,
      "auxiliary_loss_mlp": 0.01032458,
      "balance_loss_clip": 1.05996239,
      "balance_loss_mlp": 1.02308238,
      "epoch": 0.1542716286899537,
      "flos": 11429325296640.0,
      "grad_norm": 2.291131101403316,
      "language_loss": 0.7985394,
      "learning_rate": 3.840295364514371e-06,
      "loss": 0.82170933,
      "num_input_tokens_seen": 27156760,
      "step": 1283,
      "time_per_iteration": 2.6870031356811523
    },
    {
      "auxiliary_loss_clip": 0.01329524,
      "auxiliary_loss_mlp": 0.01030738,
      "balance_loss_clip": 1.06125283,
      "balance_loss_mlp": 1.02216721,
      "epoch": 0.1543918715805928,
      "flos": 17420338719360.0,
      "grad_norm": 4.77614452864955,
      "language_loss": 0.78475344,
      "learning_rate": 3.83999020183681e-06,
      "loss": 0.80835605,
      "num_input_tokens_seen": 27175455,
      "step": 1284,
      "time_per_iteration": 2.624479293823242
    },
    {
      "auxiliary_loss_clip": 0.01471883,
      "auxiliary_loss_mlp": 0.01039055,
      "balance_loss_clip": 1.05558324,
      "balance_loss_mlp": 1.0299654,
      "epoch": 0.1545121144712319,
      "flos": 17786376264960.0,
      "grad_norm": 2.1370821521120957,
      "language_loss": 0.7893225,
      "learning_rate": 3.839684760034199e-06,
      "loss": 0.81443185,
      "num_input_tokens_seen": 27193660,
      "step": 1285,
      "time_per_iteration": 2.9282948970794678
    },
    {
      "auxiliary_loss_clip": 0.0137665,
      "auxiliary_loss_mlp": 0.01038149,
      "balance_loss_clip": 1.05820179,
      "balance_loss_mlp": 1.02848744,
      "epoch": 0.15463235736187098,
      "flos": 28220185146240.0,
      "grad_norm": 3.3642662922835522,
      "language_loss": 0.65255916,
      "learning_rate": 3.8393790391528716e-06,
      "loss": 0.67670715,
      "num_input_tokens_seen": 27214355,
      "step": 1286,
      "time_per_iteration": 2.8821663856506348
    },
    {
      "auxiliary_loss_clip": 0.01324243,
      "auxiliary_loss_mlp": 0.01040224,
      "balance_loss_clip": 1.05603313,
      "balance_loss_mlp": 1.03059161,
      "epoch": 0.15475260025251006,
      "flos": 22856890826880.0,
      "grad_norm": 2.3068168745835407,
      "language_loss": 0.89176273,
      "learning_rate": 3.8390730392392075e-06,
      "loss": 0.91540742,
      "num_input_tokens_seen": 27234335,
      "step": 1287,
      "time_per_iteration": 2.7800090312957764
    },
    {
      "auxiliary_loss_clip": 0.01231388,
      "auxiliary_loss_mlp": 0.01039611,
      "balance_loss_clip": 1.06329501,
      "balance_loss_mlp": 1.02977586,
      "epoch": 0.15487284314314917,
      "flos": 17602872658560.0,
      "grad_norm": 3.2399917745098046,
      "language_loss": 0.79293954,
      "learning_rate": 3.838766760339626e-06,
      "loss": 0.81564951,
      "num_input_tokens_seen": 27252860,
      "step": 1288,
      "time_per_iteration": 3.467919111251831
    },
    {
      "auxiliary_loss_clip": 0.01365407,
      "auxiliary_loss_mlp": 0.01035084,
      "balance_loss_clip": 1.05487871,
      "balance_loss_mlp": 1.02610207,
      "epoch": 0.15499308603378825,
      "flos": 20082037363200.0,
      "grad_norm": 3.437782493172087,
      "language_loss": 0.79120666,
      "learning_rate": 3.838460202500587e-06,
      "loss": 0.81521153,
      "num_input_tokens_seen": 27268650,
      "step": 1289,
      "time_per_iteration": 2.711481809616089
    },
    {
      "auxiliary_loss_clip": 0.01378218,
      "auxiliary_loss_mlp": 0.01035625,
      "balance_loss_clip": 1.06444907,
      "balance_loss_mlp": 1.02608228,
      "epoch": 0.15511332892442733,
      "flos": 15918051271680.0,
      "grad_norm": 2.3198769579902834,
      "language_loss": 0.74456012,
      "learning_rate": 3.838153365768599e-06,
      "loss": 0.76869857,
      "num_input_tokens_seen": 27285160,
      "step": 1290,
      "time_per_iteration": 2.7140684127807617
    },
    {
      "auxiliary_loss_clip": 0.01375851,
      "auxiliary_loss_mlp": 0.01040282,
      "balance_loss_clip": 1.06320298,
      "balance_loss_mlp": 1.03087687,
      "epoch": 0.15523357181506645,
      "flos": 41282475569280.0,
      "grad_norm": 2.9721809637243304,
      "language_loss": 0.75019288,
      "learning_rate": 3.837846250190206e-06,
      "loss": 0.77435416,
      "num_input_tokens_seen": 27308025,
      "step": 1291,
      "time_per_iteration": 2.8715617656707764
    },
    {
      "auxiliary_loss_clip": 0.01377837,
      "auxiliary_loss_mlp": 0.0257811,
      "balance_loss_clip": 1.05885744,
      "balance_loss_mlp": 1.0004431,
      "epoch": 0.15535381470570553,
      "flos": 18478769806080.0,
      "grad_norm": 2.433205534423664,
      "language_loss": 0.76702583,
      "learning_rate": 3.837538855811998e-06,
      "loss": 0.80658531,
      "num_input_tokens_seen": 27326200,
      "step": 1292,
      "time_per_iteration": 2.8650379180908203
    },
    {
      "auxiliary_loss_clip": 0.01282065,
      "auxiliary_loss_mlp": 0.01038453,
      "balance_loss_clip": 1.0639776,
      "balance_loss_mlp": 1.02939939,
      "epoch": 0.1554740575963446,
      "flos": 13918150759680.0,
      "grad_norm": 2.191543632034166,
      "language_loss": 0.71115327,
      "learning_rate": 3.837231182680606e-06,
      "loss": 0.73435843,
      "num_input_tokens_seen": 27344165,
      "step": 1293,
      "time_per_iteration": 4.477279901504517
    },
    {
      "auxiliary_loss_clip": 0.01285266,
      "auxiliary_loss_mlp": 0.01032428,
      "balance_loss_clip": 1.06308329,
      "balance_loss_mlp": 1.0228858,
      "epoch": 0.1555943004869837,
      "flos": 20847078161280.0,
      "grad_norm": 1.6301760033462263,
      "language_loss": 0.76124537,
      "learning_rate": 3.836923230842706e-06,
      "loss": 0.78442234,
      "num_input_tokens_seen": 27363280,
      "step": 1294,
      "time_per_iteration": 2.6569464206695557
    },
    {
      "auxiliary_loss_clip": 0.01428574,
      "auxiliary_loss_mlp": 0.01029691,
      "balance_loss_clip": 1.05233049,
      "balance_loss_mlp": 1.02089918,
      "epoch": 0.1557145433776228,
      "flos": 22085888371200.0,
      "grad_norm": 2.2314971780355277,
      "language_loss": 0.80597359,
      "learning_rate": 3.836615000345011e-06,
      "loss": 0.83055627,
      "num_input_tokens_seen": 27381460,
      "step": 1295,
      "time_per_iteration": 2.801683187484741
    },
    {
      "auxiliary_loss_clip": 0.01233676,
      "auxiliary_loss_mlp": 0.01029989,
      "balance_loss_clip": 1.06662083,
      "balance_loss_mlp": 1.02149594,
      "epoch": 0.1558347862682619,
      "flos": 19791987039360.0,
      "grad_norm": 2.3540270533789887,
      "language_loss": 0.77829301,
      "learning_rate": 3.836306491234282e-06,
      "loss": 0.80092967,
      "num_input_tokens_seen": 27399310,
      "step": 1296,
      "time_per_iteration": 3.6349635124206543
    },
    {
      "auxiliary_loss_clip": 0.01330871,
      "auxiliary_loss_mlp": 0.01038972,
      "balance_loss_clip": 1.06803143,
      "balance_loss_mlp": 1.0299716,
      "epoch": 0.15595502915890097,
      "flos": 17237086508160.0,
      "grad_norm": 2.400441496495865,
      "language_loss": 0.75380945,
      "learning_rate": 3.835997703557317e-06,
      "loss": 0.7775079,
      "num_input_tokens_seen": 27416050,
      "step": 1297,
      "time_per_iteration": 2.6636579036712646
    },
    {
      "auxiliary_loss_clip": 0.01431565,
      "auxiliary_loss_mlp": 0.01039764,
      "balance_loss_clip": 1.0528115,
      "balance_loss_mlp": 1.03038239,
      "epoch": 0.15607527204954008,
      "flos": 19719519350400.0,
      "grad_norm": 1.9004147349007385,
      "language_loss": 0.79842657,
      "learning_rate": 3.83568863736096e-06,
      "loss": 0.82313991,
      "num_input_tokens_seen": 27434920,
      "step": 1298,
      "time_per_iteration": 2.8169009685516357
    },
    {
      "auxiliary_loss_clip": 0.01378401,
      "auxiliary_loss_mlp": 0.01036066,
      "balance_loss_clip": 1.05424786,
      "balance_loss_mlp": 1.02665997,
      "epoch": 0.15619551494017916,
      "flos": 18515650095360.0,
      "grad_norm": 3.4242557774599702,
      "language_loss": 0.89311433,
      "learning_rate": 3.8353792926920975e-06,
      "loss": 0.91725904,
      "num_input_tokens_seen": 27453570,
      "step": 1299,
      "time_per_iteration": 2.730025291442871
    },
    {
      "auxiliary_loss_clip": 0.0128577,
      "auxiliary_loss_mlp": 0.01040378,
      "balance_loss_clip": 1.06470609,
      "balance_loss_mlp": 1.03104448,
      "epoch": 0.15631575783081825,
      "flos": 19902125116800.0,
      "grad_norm": 2.588427768614712,
      "language_loss": 0.81978399,
      "learning_rate": 3.835069669597655e-06,
      "loss": 0.84304547,
      "num_input_tokens_seen": 27471960,
      "step": 1300,
      "time_per_iteration": 2.7109203338623047
    },
    {
      "auxiliary_loss_clip": 0.01286047,
      "auxiliary_loss_mlp": 0.02583776,
      "balance_loss_clip": 1.06326497,
      "balance_loss_mlp": 1.00048923,
      "epoch": 0.15643600072145733,
      "flos": 20777663128320.0,
      "grad_norm": 2.193896963856943,
      "language_loss": 0.80007756,
      "learning_rate": 3.834759768124603e-06,
      "loss": 0.83877581,
      "num_input_tokens_seen": 27490835,
      "step": 1301,
      "time_per_iteration": 2.692371129989624
    },
    {
      "auxiliary_loss_clip": 0.01330644,
      "auxiliary_loss_mlp": 0.01042164,
      "balance_loss_clip": 1.06353426,
      "balance_loss_mlp": 1.03250861,
      "epoch": 0.15655624361209644,
      "flos": 18546389159040.0,
      "grad_norm": 2.8769111755542345,
      "language_loss": 0.7627638,
      "learning_rate": 3.834449588319953e-06,
      "loss": 0.78649187,
      "num_input_tokens_seen": 27508870,
      "step": 1302,
      "time_per_iteration": 2.716305732727051
    },
    {
      "auxiliary_loss_clip": 0.01279761,
      "auxiliary_loss_mlp": 0.01044207,
      "balance_loss_clip": 1.06582379,
      "balance_loss_mlp": 1.03441405,
      "epoch": 0.15667648650273552,
      "flos": 25229544727680.0,
      "grad_norm": 2.641452399583895,
      "language_loss": 0.85210168,
      "learning_rate": 3.834139130230758e-06,
      "loss": 0.87534136,
      "num_input_tokens_seen": 27528175,
      "step": 1303,
      "time_per_iteration": 2.7111051082611084
    },
    {
      "auxiliary_loss_clip": 0.01330579,
      "auxiliary_loss_mlp": 0.01032484,
      "balance_loss_clip": 1.05793762,
      "balance_loss_mlp": 1.02297115,
      "epoch": 0.1567967293933746,
      "flos": 24827093769600.0,
      "grad_norm": 1.7101404893171706,
      "language_loss": 0.80882078,
      "learning_rate": 3.833828393904117e-06,
      "loss": 0.8324514,
      "num_input_tokens_seen": 27548455,
      "step": 1304,
      "time_per_iteration": 2.7442586421966553
    },
    {
      "auxiliary_loss_clip": 0.01370672,
      "auxiliary_loss_mlp": 0.01035117,
      "balance_loss_clip": 1.05352974,
      "balance_loss_mlp": 1.02538908,
      "epoch": 0.15691697228401372,
      "flos": 19164555244800.0,
      "grad_norm": 2.4676504496153657,
      "language_loss": 0.77581191,
      "learning_rate": 3.833517379387165e-06,
      "loss": 0.79986978,
      "num_input_tokens_seen": 27564910,
      "step": 1305,
      "time_per_iteration": 2.764780282974243
    },
    {
      "auxiliary_loss_clip": 0.01285458,
      "auxiliary_loss_mlp": 0.01033272,
      "balance_loss_clip": 1.06300914,
      "balance_loss_mlp": 1.02392578,
      "epoch": 0.1570372151746528,
      "flos": 24790931752320.0,
      "grad_norm": 2.012425762480874,
      "language_loss": 0.88897502,
      "learning_rate": 3.833206086727085e-06,
      "loss": 0.9121623,
      "num_input_tokens_seen": 27584260,
      "step": 1306,
      "time_per_iteration": 2.69722318649292
    },
    {
      "auxiliary_loss_clip": 0.01380577,
      "auxiliary_loss_mlp": 0.01032653,
      "balance_loss_clip": 1.0545516,
      "balance_loss_mlp": 1.02340877,
      "epoch": 0.15715745806529188,
      "flos": 24863650836480.0,
      "grad_norm": 2.9263861552076897,
      "language_loss": 0.70528698,
      "learning_rate": 3.8328945159710994e-06,
      "loss": 0.72941935,
      "num_input_tokens_seen": 27604440,
      "step": 1307,
      "time_per_iteration": 2.7881438732147217
    },
    {
      "auxiliary_loss_clip": 0.01291257,
      "auxiliary_loss_mlp": 0.02576995,
      "balance_loss_clip": 1.06774449,
      "balance_loss_mlp": 1.00047493,
      "epoch": 0.157277700955931,
      "flos": 21872148491520.0,
      "grad_norm": 3.1076971473298705,
      "language_loss": 0.88859296,
      "learning_rate": 3.832582667166473e-06,
      "loss": 0.92727548,
      "num_input_tokens_seen": 27624250,
      "step": 1308,
      "time_per_iteration": 2.673759698867798
    },
    {
      "auxiliary_loss_clip": 0.01333181,
      "auxiliary_loss_mlp": 0.01032512,
      "balance_loss_clip": 1.05963063,
      "balance_loss_mlp": 1.02355933,
      "epoch": 0.15739794384657008,
      "flos": 24533344344960.0,
      "grad_norm": 2.066051630357901,
      "language_loss": 0.81504214,
      "learning_rate": 3.8322705403605125e-06,
      "loss": 0.83869898,
      "num_input_tokens_seen": 27644595,
      "step": 1309,
      "time_per_iteration": 2.768216133117676
    },
    {
      "auxiliary_loss_clip": 0.01321548,
      "auxiliary_loss_mlp": 0.01028344,
      "balance_loss_clip": 1.05857456,
      "balance_loss_mlp": 1.01973712,
      "epoch": 0.15751818673720916,
      "flos": 17745329998080.0,
      "grad_norm": 8.743601966666455,
      "language_loss": 0.81456852,
      "learning_rate": 3.831958135600568e-06,
      "loss": 0.83806741,
      "num_input_tokens_seen": 27662145,
      "step": 1310,
      "time_per_iteration": 2.6298184394836426
    },
    {
      "auxiliary_loss_clip": 0.01284794,
      "auxiliary_loss_mlp": 0.01034947,
      "balance_loss_clip": 1.06404781,
      "balance_loss_mlp": 1.02619696,
      "epoch": 0.15763842962784824,
      "flos": 17858520731520.0,
      "grad_norm": 2.2638046756072807,
      "language_loss": 0.79516357,
      "learning_rate": 3.831645452934032e-06,
      "loss": 0.81836092,
      "num_input_tokens_seen": 27680575,
      "step": 1311,
      "time_per_iteration": 2.6748552322387695
    },
    {
      "auxiliary_loss_clip": 0.0123524,
      "auxiliary_loss_mlp": 0.01032081,
      "balance_loss_clip": 1.0685606,
      "balance_loss_mlp": 1.02309227,
      "epoch": 0.15775867251848735,
      "flos": 26980908059520.0,
      "grad_norm": 2.1443480424563712,
      "language_loss": 0.80262864,
      "learning_rate": 3.831332492408336e-06,
      "loss": 0.82530177,
      "num_input_tokens_seen": 27701985,
      "step": 1312,
      "time_per_iteration": 2.6757545471191406
    },
    {
      "auxiliary_loss_clip": 0.01325107,
      "auxiliary_loss_mlp": 0.01031397,
      "balance_loss_clip": 1.05749011,
      "balance_loss_mlp": 1.02225924,
      "epoch": 0.15787891540912644,
      "flos": 19240398812160.0,
      "grad_norm": 2.4916802772552997,
      "language_loss": 0.69355345,
      "learning_rate": 3.831019254070957e-06,
      "loss": 0.71711844,
      "num_input_tokens_seen": 27719770,
      "step": 1313,
      "time_per_iteration": 2.7027721405029297
    },
    {
      "auxiliary_loss_clip": 0.0142848,
      "auxiliary_loss_mlp": 0.01036322,
      "balance_loss_clip": 1.05631256,
      "balance_loss_mlp": 1.02692795,
      "epoch": 0.15799915829976552,
      "flos": 27271102037760.0,
      "grad_norm": 8.7029703845329,
      "language_loss": 0.95156813,
      "learning_rate": 3.8307057379694135e-06,
      "loss": 0.9762162,
      "num_input_tokens_seen": 27739105,
      "step": 1314,
      "time_per_iteration": 3.7809805870056152
    },
    {
      "auxiliary_loss_clip": 0.01233608,
      "auxiliary_loss_mlp": 0.01035256,
      "balance_loss_clip": 1.06346524,
      "balance_loss_mlp": 1.02626133,
      "epoch": 0.15811940119040463,
      "flos": 20405520270720.0,
      "grad_norm": 2.191699177632541,
      "language_loss": 0.8265574,
      "learning_rate": 3.830391944151264e-06,
      "loss": 0.84924603,
      "num_input_tokens_seen": 27754985,
      "step": 1315,
      "time_per_iteration": 2.592554807662964
    },
    {
      "auxiliary_loss_clip": 0.0132956,
      "auxiliary_loss_mlp": 0.0103389,
      "balance_loss_clip": 1.0574857,
      "balance_loss_mlp": 1.02500927,
      "epoch": 0.1582396440810437,
      "flos": 32599347661440.0,
      "grad_norm": 1.8851320641489395,
      "language_loss": 0.67523563,
      "learning_rate": 3.830077872664114e-06,
      "loss": 0.69887018,
      "num_input_tokens_seen": 27776110,
      "step": 1316,
      "time_per_iteration": 2.829266309738159
    },
    {
      "auxiliary_loss_clip": 0.01482384,
      "auxiliary_loss_mlp": 0.01031921,
      "balance_loss_clip": 1.05430198,
      "balance_loss_mlp": 1.02275431,
      "epoch": 0.1583598869716828,
      "flos": 33800559310080.0,
      "grad_norm": 3.6867462552200396,
      "language_loss": 0.72891682,
      "learning_rate": 3.829763523555604e-06,
      "loss": 0.75405991,
      "num_input_tokens_seen": 27796510,
      "step": 1317,
      "time_per_iteration": 2.8823046684265137
    },
    {
      "auxiliary_loss_clip": 0.01281644,
      "auxiliary_loss_mlp": 0.01037702,
      "balance_loss_clip": 1.06758046,
      "balance_loss_mlp": 1.02914262,
      "epoch": 0.15848012986232188,
      "flos": 24681332378880.0,
      "grad_norm": 3.3872992549735907,
      "language_loss": 0.7811588,
      "learning_rate": 3.829448896873423e-06,
      "loss": 0.80435222,
      "num_input_tokens_seen": 27815610,
      "step": 1318,
      "time_per_iteration": 2.7282354831695557
    },
    {
      "auxiliary_loss_clip": 0.01422146,
      "auxiliary_loss_mlp": 0.02582895,
      "balance_loss_clip": 1.05675626,
      "balance_loss_mlp": 1.00039256,
      "epoch": 0.158600372752961,
      "flos": 22602068766720.0,
      "grad_norm": 8.471460245681248,
      "language_loss": 0.79153746,
      "learning_rate": 3.829133992665299e-06,
      "loss": 0.83158785,
      "num_input_tokens_seen": 27834735,
      "step": 1319,
      "time_per_iteration": 3.7018277645111084
    },
    {
      "auxiliary_loss_clip": 0.01273733,
      "auxiliary_loss_mlp": 0.01033919,
      "balance_loss_clip": 1.06145644,
      "balance_loss_mlp": 1.02481782,
      "epoch": 0.15872061564360007,
      "flos": 27927944092800.0,
      "grad_norm": 4.905872243925333,
      "language_loss": 0.88928574,
      "learning_rate": 3.828818810979002e-06,
      "loss": 0.91236234,
      "num_input_tokens_seen": 27853065,
      "step": 1320,
      "time_per_iteration": 2.653956413269043
    },
    {
      "auxiliary_loss_clip": 0.01233672,
      "auxiliary_loss_mlp": 0.01029245,
      "balance_loss_clip": 1.06754696,
      "balance_loss_mlp": 1.02020872,
      "epoch": 0.15884085853423915,
      "flos": 23696805525120.0,
      "grad_norm": 2.799270267978755,
      "language_loss": 0.80452502,
      "learning_rate": 3.8285033518623454e-06,
      "loss": 0.82715416,
      "num_input_tokens_seen": 27873315,
      "step": 1321,
      "time_per_iteration": 2.657195568084717
    },
    {
      "auxiliary_loss_clip": 0.01285611,
      "auxiliary_loss_mlp": 0.01034838,
      "balance_loss_clip": 1.06374288,
      "balance_loss_mlp": 1.02444303,
      "epoch": 0.15896110142487826,
      "flos": 23112359331840.0,
      "grad_norm": 3.2742172720887703,
      "language_loss": 0.81316459,
      "learning_rate": 3.8281876153631845e-06,
      "loss": 0.8363691,
      "num_input_tokens_seen": 27890070,
      "step": 1322,
      "time_per_iteration": 3.5065791606903076
    },
    {
      "auxiliary_loss_clip": 0.01430983,
      "auxiliary_loss_mlp": 0.01036167,
      "balance_loss_clip": 1.05630386,
      "balance_loss_mlp": 1.02680945,
      "epoch": 0.15908134431551735,
      "flos": 14685238632960.0,
      "grad_norm": 2.1720105810767274,
      "language_loss": 0.64689624,
      "learning_rate": 3.827871601529416e-06,
      "loss": 0.67156768,
      "num_input_tokens_seen": 27908590,
      "step": 1323,
      "time_per_iteration": 2.757573366165161
    },
    {
      "auxiliary_loss_clip": 0.01322457,
      "auxiliary_loss_mlp": 0.01031855,
      "balance_loss_clip": 1.05934525,
      "balance_loss_mlp": 1.02311122,
      "epoch": 0.15920158720615643,
      "flos": 20193611984640.0,
      "grad_norm": 2.945869454663328,
      "language_loss": 0.80504751,
      "learning_rate": 3.827555310408979e-06,
      "loss": 0.82859063,
      "num_input_tokens_seen": 27927985,
      "step": 1324,
      "time_per_iteration": 2.7074549198150635
    },
    {
      "auxiliary_loss_clip": 0.01382594,
      "auxiliary_loss_mlp": 0.01037817,
      "balance_loss_clip": 1.06506586,
      "balance_loss_mlp": 1.02848887,
      "epoch": 0.1593218300967955,
      "flos": 24826626892800.0,
      "grad_norm": 1.8510701401337846,
      "language_loss": 0.83186847,
      "learning_rate": 3.827238742049854e-06,
      "loss": 0.85607255,
      "num_input_tokens_seen": 27948280,
      "step": 1325,
      "time_per_iteration": 2.7707784175872803
    },
    {
      "auxiliary_loss_clip": 0.01230111,
      "auxiliary_loss_mlp": 0.01035096,
      "balance_loss_clip": 1.0629375,
      "balance_loss_mlp": 1.02619743,
      "epoch": 0.15944207298743462,
      "flos": 28328707111680.0,
      "grad_norm": 1.8039251262164828,
      "language_loss": 0.51599193,
      "learning_rate": 3.826921896500066e-06,
      "loss": 0.53864402,
      "num_input_tokens_seen": 27969565,
      "step": 1326,
      "time_per_iteration": 2.70924973487854
    },
    {
      "auxiliary_loss_clip": 0.01384886,
      "auxiliary_loss_mlp": 0.01032668,
      "balance_loss_clip": 1.05969238,
      "balance_loss_mlp": 1.0235728,
      "epoch": 0.1595623158780737,
      "flos": 22964838174720.0,
      "grad_norm": 9.32373691443938,
      "language_loss": 0.78048062,
      "learning_rate": 3.826604773807678e-06,
      "loss": 0.80465615,
      "num_input_tokens_seen": 27987540,
      "step": 1327,
      "time_per_iteration": 2.732954263687134
    },
    {
      "auxiliary_loss_clip": 0.01329517,
      "auxiliary_loss_mlp": 0.01031047,
      "balance_loss_clip": 1.05680537,
      "balance_loss_mlp": 1.02148604,
      "epoch": 0.1596825587687128,
      "flos": 19710540950400.0,
      "grad_norm": 2.9162553108940767,
      "language_loss": 0.74556386,
      "learning_rate": 3.826287374020798e-06,
      "loss": 0.76916951,
      "num_input_tokens_seen": 28002345,
      "step": 1328,
      "time_per_iteration": 2.6842048168182373
    },
    {
      "auxiliary_loss_clip": 0.01233291,
      "auxiliary_loss_mlp": 0.01042677,
      "balance_loss_clip": 1.06747687,
      "balance_loss_mlp": 1.03303885,
      "epoch": 0.1598028016593519,
      "flos": 22637727993600.0,
      "grad_norm": 1.9630859155810778,
      "language_loss": 0.8246367,
      "learning_rate": 3.825969697187575e-06,
      "loss": 0.84739631,
      "num_input_tokens_seen": 28021675,
      "step": 1329,
      "time_per_iteration": 2.638132095336914
    },
    {
      "auxiliary_loss_clip": 0.01380289,
      "auxiliary_loss_mlp": 0.01029814,
      "balance_loss_clip": 1.05624676,
      "balance_loss_mlp": 1.02040219,
      "epoch": 0.15992304454999098,
      "flos": 20482908122880.0,
      "grad_norm": 4.49563349981601,
      "language_loss": 0.69360447,
      "learning_rate": 3.8256517433562015e-06,
      "loss": 0.71770549,
      "num_input_tokens_seen": 28039615,
      "step": 1330,
      "time_per_iteration": 2.7775821685791016
    },
    {
      "auxiliary_loss_clip": 0.01229613,
      "auxiliary_loss_mlp": 0.01038841,
      "balance_loss_clip": 1.06391239,
      "balance_loss_mlp": 1.02928007,
      "epoch": 0.16004328744063007,
      "flos": 17676094533120.0,
      "grad_norm": 2.3032069003246276,
      "language_loss": 0.92024672,
      "learning_rate": 3.82533351257491e-06,
      "loss": 0.94293118,
      "num_input_tokens_seen": 28057565,
      "step": 1331,
      "time_per_iteration": 2.6200575828552246
    },
    {
      "auxiliary_loss_clip": 0.0128074,
      "auxiliary_loss_mlp": 0.01035266,
      "balance_loss_clip": 1.06636178,
      "balance_loss_mlp": 1.02685618,
      "epoch": 0.16016353033126918,
      "flos": 24098717779200.0,
      "grad_norm": 1.8563699719828266,
      "language_loss": 0.88798207,
      "learning_rate": 3.825015004891975e-06,
      "loss": 0.91114211,
      "num_input_tokens_seen": 28076305,
      "step": 1332,
      "time_per_iteration": 2.6794424057006836
    },
    {
      "auxiliary_loss_clip": 0.01277258,
      "auxiliary_loss_mlp": 0.01033486,
      "balance_loss_clip": 1.06096172,
      "balance_loss_mlp": 1.02519464,
      "epoch": 0.16028377322190826,
      "flos": 27634841112960.0,
      "grad_norm": 1.7247970479428871,
      "language_loss": 0.75883764,
      "learning_rate": 3.824696220355716e-06,
      "loss": 0.78194505,
      "num_input_tokens_seen": 28097895,
      "step": 1333,
      "time_per_iteration": 2.666212797164917
    },
    {
      "auxiliary_loss_clip": 0.01325778,
      "auxiliary_loss_mlp": 0.01031723,
      "balance_loss_clip": 1.06196439,
      "balance_loss_mlp": 1.02335417,
      "epoch": 0.16040401611254734,
      "flos": 20961202648320.0,
      "grad_norm": 1.5470809826311183,
      "language_loss": 0.78951478,
      "learning_rate": 3.824377159014491e-06,
      "loss": 0.81308985,
      "num_input_tokens_seen": 28118790,
      "step": 1334,
      "time_per_iteration": 2.7269270420074463
    },
    {
      "auxiliary_loss_clip": 0.01280596,
      "auxiliary_loss_mlp": 0.01029063,
      "balance_loss_clip": 1.06420064,
      "balance_loss_mlp": 1.02055132,
      "epoch": 0.16052425900318643,
      "flos": 21247051080960.0,
      "grad_norm": 2.388011250340376,
      "language_loss": 0.85179365,
      "learning_rate": 3.824057820916702e-06,
      "loss": 0.87489027,
      "num_input_tokens_seen": 28135995,
      "step": 1335,
      "time_per_iteration": 2.613114833831787
    },
    {
      "auxiliary_loss_clip": 0.01332182,
      "auxiliary_loss_mlp": 0.01032302,
      "balance_loss_clip": 1.0617857,
      "balance_loss_mlp": 1.02268159,
      "epoch": 0.16064450189382554,
      "flos": 15524004096000.0,
      "grad_norm": 3.524332487290133,
      "language_loss": 0.71980077,
      "learning_rate": 3.8237382061107904e-06,
      "loss": 0.74344563,
      "num_input_tokens_seen": 28152715,
      "step": 1336,
      "time_per_iteration": 2.6544723510742188
    },
    {
      "auxiliary_loss_clip": 0.01517902,
      "auxiliary_loss_mlp": 0.01035999,
      "balance_loss_clip": 1.05059099,
      "balance_loss_mlp": 1.02672482,
      "epoch": 0.16076474478446462,
      "flos": 21178497974400.0,
      "grad_norm": 1.981289981892937,
      "language_loss": 0.78701448,
      "learning_rate": 3.823418314645243e-06,
      "loss": 0.81255352,
      "num_input_tokens_seen": 28171590,
      "step": 1337,
      "time_per_iteration": 2.9089746475219727
    },
    {
      "auxiliary_loss_clip": 0.01414444,
      "auxiliary_loss_mlp": 0.0103463,
      "balance_loss_clip": 1.05757141,
      "balance_loss_mlp": 1.02576637,
      "epoch": 0.1608849876751037,
      "flos": 18366476912640.0,
      "grad_norm": 2.3325519818736833,
      "language_loss": 0.74994159,
      "learning_rate": 3.823098146568588e-06,
      "loss": 0.7744323,
      "num_input_tokens_seen": 28191295,
      "step": 1338,
      "time_per_iteration": 3.1610448360443115
    },
    {
      "auxiliary_loss_clip": 0.01283805,
      "auxiliary_loss_mlp": 0.01035517,
      "balance_loss_clip": 1.06595826,
      "balance_loss_mlp": 1.02729166,
      "epoch": 0.1610052305657428,
      "flos": 29497024880640.0,
      "grad_norm": 2.0324186041133743,
      "language_loss": 0.7135278,
      "learning_rate": 3.822777701929394e-06,
      "loss": 0.73672098,
      "num_input_tokens_seen": 28213120,
      "step": 1339,
      "time_per_iteration": 2.7361841201782227
    },
    {
      "auxiliary_loss_clip": 0.01272554,
      "auxiliary_loss_mlp": 0.01035505,
      "balance_loss_clip": 1.05939221,
      "balance_loss_mlp": 1.02621293,
      "epoch": 0.1611254734563819,
      "flos": 26797871329920.0,
      "grad_norm": 1.8103763322318076,
      "language_loss": 0.73702335,
      "learning_rate": 3.8224569807762714e-06,
      "loss": 0.76010394,
      "num_input_tokens_seen": 28232440,
      "step": 1340,
      "time_per_iteration": 3.606426477432251
    },
    {
      "auxiliary_loss_clip": 0.01418114,
      "auxiliary_loss_mlp": 0.01034961,
      "balance_loss_clip": 1.05243778,
      "balance_loss_mlp": 1.02609789,
      "epoch": 0.16124571634702098,
      "flos": 22419570741120.0,
      "grad_norm": 2.131192271002674,
      "language_loss": 0.76563418,
      "learning_rate": 3.822135983157873e-06,
      "loss": 0.79016483,
      "num_input_tokens_seen": 28251715,
      "step": 1341,
      "time_per_iteration": 2.7617552280426025
    },
    {
      "auxiliary_loss_clip": 0.01227626,
      "auxiliary_loss_mlp": 0.0257551,
      "balance_loss_clip": 1.06319153,
      "balance_loss_mlp": 1.00036538,
      "epoch": 0.16136595923766006,
      "flos": 10999116103680.0,
      "grad_norm": 4.175435257609612,
      "language_loss": 0.84447563,
      "learning_rate": 3.821814709122896e-06,
      "loss": 0.88250697,
      "num_input_tokens_seen": 28269765,
      "step": 1342,
      "time_per_iteration": 2.57891845703125
    },
    {
      "auxiliary_loss_clip": 0.01326795,
      "auxiliary_loss_mlp": 0.01031932,
      "balance_loss_clip": 1.05970883,
      "balance_loss_mlp": 1.02324772,
      "epoch": 0.16148620212829917,
      "flos": 21214983214080.0,
      "grad_norm": 2.3952863262451802,
      "language_loss": 0.84989846,
      "learning_rate": 3.821493158720076e-06,
      "loss": 0.87348574,
      "num_input_tokens_seen": 28288870,
      "step": 1343,
      "time_per_iteration": 2.720737934112549
    },
    {
      "auxiliary_loss_clip": 0.0137564,
      "auxiliary_loss_mlp": 0.01043055,
      "balance_loss_clip": 1.05417824,
      "balance_loss_mlp": 1.0335362,
      "epoch": 0.16160644501893826,
      "flos": 16758468760320.0,
      "grad_norm": 2.86833282292341,
      "language_loss": 0.73568332,
      "learning_rate": 3.821171331998191e-06,
      "loss": 0.75987029,
      "num_input_tokens_seen": 28305400,
      "step": 1344,
      "time_per_iteration": 2.6849477291107178
    },
    {
      "auxiliary_loss_clip": 0.01293706,
      "auxiliary_loss_mlp": 0.01003809,
      "balance_loss_clip": 1.03328085,
      "balance_loss_mlp": 1.00175858,
      "epoch": 0.16172668790957734,
      "flos": 64444967308800.0,
      "grad_norm": 0.7093532955435258,
      "language_loss": 0.54454035,
      "learning_rate": 3.820849229006064e-06,
      "loss": 0.56751549,
      "num_input_tokens_seen": 28373150,
      "step": 1345,
      "time_per_iteration": 5.272507190704346
    },
    {
      "auxiliary_loss_clip": 0.01231247,
      "auxiliary_loss_mlp": 0.01029043,
      "balance_loss_clip": 1.06429744,
      "balance_loss_mlp": 1.02055502,
      "epoch": 0.16184693080021645,
      "flos": 23257689759360.0,
      "grad_norm": 2.609384963228575,
      "language_loss": 0.70974267,
      "learning_rate": 3.8205268497925564e-06,
      "loss": 0.73234558,
      "num_input_tokens_seen": 28393620,
      "step": 1346,
      "time_per_iteration": 2.689685821533203
    },
    {
      "auxiliary_loss_clip": 0.01231356,
      "auxiliary_loss_mlp": 0.01032488,
      "balance_loss_clip": 1.06617773,
      "balance_loss_mlp": 1.02321935,
      "epoch": 0.16196717369085553,
      "flos": 17451113696640.0,
      "grad_norm": 2.7736455609514943,
      "language_loss": 0.78439784,
      "learning_rate": 3.8202041944065725e-06,
      "loss": 0.80703628,
      "num_input_tokens_seen": 28409440,
      "step": 1347,
      "time_per_iteration": 2.640169143676758
    },
    {
      "auxiliary_loss_clip": 0.01233787,
      "auxiliary_loss_mlp": 0.01036774,
      "balance_loss_clip": 1.06659627,
      "balance_loss_mlp": 1.02852464,
      "epoch": 0.16208741658149461,
      "flos": 23873377806720.0,
      "grad_norm": 2.1441172797281034,
      "language_loss": 0.73803592,
      "learning_rate": 3.819881262897061e-06,
      "loss": 0.76074159,
      "num_input_tokens_seen": 28427575,
      "step": 1348,
      "time_per_iteration": 3.5313470363616943
    },
    {
      "auxiliary_loss_clip": 0.01378338,
      "auxiliary_loss_mlp": 0.01035449,
      "balance_loss_clip": 1.06304884,
      "balance_loss_mlp": 1.02542365,
      "epoch": 0.1622076594721337,
      "flos": 25884806584320.0,
      "grad_norm": 2.4312463585454047,
      "language_loss": 0.73557925,
      "learning_rate": 3.819558055313008e-06,
      "loss": 0.75971711,
      "num_input_tokens_seen": 28448260,
      "step": 1349,
      "time_per_iteration": 2.7929446697235107
    },
    {
      "auxiliary_loss_clip": 0.01286584,
      "auxiliary_loss_mlp": 0.01035855,
      "balance_loss_clip": 1.0644536,
      "balance_loss_mlp": 1.02756429,
      "epoch": 0.1623279023627728,
      "flos": 21539759011200.0,
      "grad_norm": 3.7028190230734266,
      "language_loss": 0.77542746,
      "learning_rate": 3.819234571703444e-06,
      "loss": 0.79865181,
      "num_input_tokens_seen": 28467085,
      "step": 1350,
      "time_per_iteration": 2.6728122234344482
    },
    {
      "auxiliary_loss_clip": 0.01279061,
      "auxiliary_loss_mlp": 0.01037537,
      "balance_loss_clip": 1.06313562,
      "balance_loss_mlp": 1.02857804,
      "epoch": 0.1624481452534119,
      "flos": 22085421494400.0,
      "grad_norm": 4.1682241401011675,
      "language_loss": 0.85605836,
      "learning_rate": 3.8189108121174435e-06,
      "loss": 0.8792243,
      "num_input_tokens_seen": 28486850,
      "step": 1351,
      "time_per_iteration": 2.6614482402801514
    },
    {
      "auxiliary_loss_clip": 0.01371447,
      "auxiliary_loss_mlp": 0.01038982,
      "balance_loss_clip": 1.06116128,
      "balance_loss_mlp": 1.02982044,
      "epoch": 0.16256838814405097,
      "flos": 27087490690560.0,
      "grad_norm": 1.679656163009191,
      "language_loss": 0.83329082,
      "learning_rate": 3.818586776604118e-06,
      "loss": 0.85739517,
      "num_input_tokens_seen": 28507490,
      "step": 1352,
      "time_per_iteration": 2.776592254638672
    },
    {
      "auxiliary_loss_clip": 0.0126841,
      "auxiliary_loss_mlp": 0.01036316,
      "balance_loss_clip": 1.05934453,
      "balance_loss_mlp": 1.02793539,
      "epoch": 0.16268863103469008,
      "flos": 20120354196480.0,
      "grad_norm": 2.1391077356467805,
      "language_loss": 0.61657262,
      "learning_rate": 3.818262465212625e-06,
      "loss": 0.63961995,
      "num_input_tokens_seen": 28527615,
      "step": 1353,
      "time_per_iteration": 2.6782467365264893
    },
    {
      "auxiliary_loss_clip": 0.01274271,
      "auxiliary_loss_mlp": 0.01038994,
      "balance_loss_clip": 1.06207228,
      "balance_loss_mlp": 1.02980864,
      "epoch": 0.16280887392532917,
      "flos": 18332792933760.0,
      "grad_norm": 3.256805700350427,
      "language_loss": 0.77046371,
      "learning_rate": 3.817937877992161e-06,
      "loss": 0.79359627,
      "num_input_tokens_seen": 28544910,
      "step": 1354,
      "time_per_iteration": 2.6224617958068848
    },
    {
      "auxiliary_loss_clip": 0.01378882,
      "auxiliary_loss_mlp": 0.02578875,
      "balance_loss_clip": 1.05547607,
      "balance_loss_mlp": 1.00029302,
      "epoch": 0.16292911681596825,
      "flos": 11874330892800.0,
      "grad_norm": 2.4942228738212586,
      "language_loss": 0.85647738,
      "learning_rate": 3.817613014991967e-06,
      "loss": 0.89605492,
      "num_input_tokens_seen": 28561050,
      "step": 1355,
      "time_per_iteration": 2.689931869506836
    },
    {
      "auxiliary_loss_clip": 0.01370313,
      "auxiliary_loss_mlp": 0.0103693,
      "balance_loss_clip": 1.05550718,
      "balance_loss_mlp": 1.02797103,
      "epoch": 0.16304935970660733,
      "flos": 26103466627200.0,
      "grad_norm": 2.118298044736565,
      "language_loss": 0.76568907,
      "learning_rate": 3.817287876261323e-06,
      "loss": 0.78976142,
      "num_input_tokens_seen": 28581385,
      "step": 1356,
      "time_per_iteration": 2.759584903717041
    },
    {
      "auxiliary_loss_clip": 0.01324114,
      "auxiliary_loss_mlp": 0.0103106,
      "balance_loss_clip": 1.06007493,
      "balance_loss_mlp": 1.02217269,
      "epoch": 0.16316960259724644,
      "flos": 29351945848320.0,
      "grad_norm": 2.6198324112894724,
      "language_loss": 0.80074048,
      "learning_rate": 3.816962461849553e-06,
      "loss": 0.82429224,
      "num_input_tokens_seen": 28603255,
      "step": 1357,
      "time_per_iteration": 2.728184700012207
    },
    {
      "auxiliary_loss_clip": 0.01329297,
      "auxiliary_loss_mlp": 0.01040325,
      "balance_loss_clip": 1.06240022,
      "balance_loss_mlp": 1.03057361,
      "epoch": 0.16328984548788553,
      "flos": 20886759711360.0,
      "grad_norm": 4.034562719425338,
      "language_loss": 0.84148502,
      "learning_rate": 3.8166367718060235e-06,
      "loss": 0.86518121,
      "num_input_tokens_seen": 28623145,
      "step": 1358,
      "time_per_iteration": 2.701451539993286
    },
    {
      "auxiliary_loss_clip": 0.0127648,
      "auxiliary_loss_mlp": 0.0103362,
      "balance_loss_clip": 1.05931163,
      "balance_loss_mlp": 1.02497113,
      "epoch": 0.1634100883785246,
      "flos": 18041090584320.0,
      "grad_norm": 2.1171706625049334,
      "language_loss": 0.77173173,
      "learning_rate": 3.816310806180139e-06,
      "loss": 0.79483271,
      "num_input_tokens_seen": 28641555,
      "step": 1359,
      "time_per_iteration": 2.6479380130767822
    },
    {
      "auxiliary_loss_clip": 0.0132635,
      "auxiliary_loss_mlp": 0.01028696,
      "balance_loss_clip": 1.06278741,
      "balance_loss_mlp": 1.01955914,
      "epoch": 0.16353033126916372,
      "flos": 24572128055040.0,
      "grad_norm": 2.808294009587239,
      "language_loss": 0.81263828,
      "learning_rate": 3.81598456502135e-06,
      "loss": 0.83618879,
      "num_input_tokens_seen": 28661575,
      "step": 1360,
      "time_per_iteration": 2.7264108657836914
    },
    {
      "auxiliary_loss_clip": 0.01323126,
      "auxiliary_loss_mlp": 0.01035453,
      "balance_loss_clip": 1.06015241,
      "balance_loss_mlp": 1.02640462,
      "epoch": 0.1636505741598028,
      "flos": 19892895321600.0,
      "grad_norm": 4.532433074698137,
      "language_loss": 0.87126297,
      "learning_rate": 3.8156580483791455e-06,
      "loss": 0.8948487,
      "num_input_tokens_seen": 28676765,
      "step": 1361,
      "time_per_iteration": 2.6319878101348877
    },
    {
      "auxiliary_loss_clip": 0.01228946,
      "auxiliary_loss_mlp": 0.01032256,
      "balance_loss_clip": 1.06340516,
      "balance_loss_mlp": 1.0236547,
      "epoch": 0.16377081705044189,
      "flos": 28402611344640.0,
      "grad_norm": 2.2440290358212276,
      "language_loss": 0.77353114,
      "learning_rate": 3.815331256303059e-06,
      "loss": 0.79614317,
      "num_input_tokens_seen": 28696795,
      "step": 1362,
      "time_per_iteration": 2.7727062702178955
    },
    {
      "auxiliary_loss_clip": 0.01377444,
      "auxiliary_loss_mlp": 0.01037516,
      "balance_loss_clip": 1.0627085,
      "balance_loss_mlp": 1.02886152,
      "epoch": 0.163891059941081,
      "flos": 21908059113600.0,
      "grad_norm": 1.968770215264132,
      "language_loss": 0.77295232,
      "learning_rate": 3.815004188842665e-06,
      "loss": 0.79710197,
      "num_input_tokens_seen": 28714835,
      "step": 1363,
      "time_per_iteration": 2.729576826095581
    },
    {
      "auxiliary_loss_clip": 0.01267794,
      "auxiliary_loss_mlp": 0.01030819,
      "balance_loss_clip": 1.05802822,
      "balance_loss_mlp": 1.02208674,
      "epoch": 0.16401130283172008,
      "flos": 26797619934720.0,
      "grad_norm": 1.7036351604068152,
      "language_loss": 0.7988646,
      "learning_rate": 3.814676846047578e-06,
      "loss": 0.82185078,
      "num_input_tokens_seen": 28735710,
      "step": 1364,
      "time_per_iteration": 2.731799840927124
    },
    {
      "auxiliary_loss_clip": 0.01278412,
      "auxiliary_loss_mlp": 0.01036004,
      "balance_loss_clip": 1.06098545,
      "balance_loss_mlp": 1.02739084,
      "epoch": 0.16413154572235916,
      "flos": 32997417160320.0,
      "grad_norm": 2.38383171730575,
      "language_loss": 0.69991207,
      "learning_rate": 3.8143492279674565e-06,
      "loss": 0.72305626,
      "num_input_tokens_seen": 28758405,
      "step": 1365,
      "time_per_iteration": 2.773869037628174
    },
    {
      "auxiliary_loss_clip": 0.0129603,
      "auxiliary_loss_mlp": 0.01003146,
      "balance_loss_clip": 1.03746367,
      "balance_loss_mlp": 1.00120282,
      "epoch": 0.16425178861299825,
      "flos": 40113622074240.0,
      "grad_norm": 0.8691801545769928,
      "language_loss": 0.58374286,
      "learning_rate": 3.8140213346519997e-06,
      "loss": 0.60673463,
      "num_input_tokens_seen": 28809000,
      "step": 1366,
      "time_per_iteration": 3.9021623134613037
    },
    {
      "auxiliary_loss_clip": 0.01370211,
      "auxiliary_loss_mlp": 0.01038144,
      "balance_loss_clip": 1.05739474,
      "balance_loss_mlp": 1.0290246,
      "epoch": 0.16437203150363736,
      "flos": 25447486498560.0,
      "grad_norm": 2.6528773441096662,
      "language_loss": 0.77067602,
      "learning_rate": 3.813693166150948e-06,
      "loss": 0.79475957,
      "num_input_tokens_seen": 28829210,
      "step": 1367,
      "time_per_iteration": 2.755739688873291
    },
    {
      "auxiliary_loss_clip": 0.01368592,
      "auxiliary_loss_mlp": 0.010326,
      "balance_loss_clip": 1.0564177,
      "balance_loss_mlp": 1.02341509,
      "epoch": 0.16449227439427644,
      "flos": 23476888506240.0,
      "grad_norm": 2.9574240021718645,
      "language_loss": 0.85433143,
      "learning_rate": 3.813364722514086e-06,
      "loss": 0.87834334,
      "num_input_tokens_seen": 28847545,
      "step": 1368,
      "time_per_iteration": 2.726825475692749
    },
    {
      "auxiliary_loss_clip": 0.01280225,
      "auxiliary_loss_mlp": 0.01034464,
      "balance_loss_clip": 1.06129801,
      "balance_loss_mlp": 1.0251298,
      "epoch": 0.16461251728491552,
      "flos": 13545217802880.0,
      "grad_norm": 2.0077819912349884,
      "language_loss": 0.80468172,
      "learning_rate": 3.8130360037912368e-06,
      "loss": 0.82782865,
      "num_input_tokens_seen": 28863990,
      "step": 1369,
      "time_per_iteration": 2.620246171951294
    },
    {
      "auxiliary_loss_clip": 0.01276554,
      "auxiliary_loss_mlp": 0.01030796,
      "balance_loss_clip": 1.06079602,
      "balance_loss_mlp": 1.02206421,
      "epoch": 0.16473276017555463,
      "flos": 23003298662400.0,
      "grad_norm": 2.0782542221248335,
      "language_loss": 0.81526929,
      "learning_rate": 3.812707010032268e-06,
      "loss": 0.83834279,
      "num_input_tokens_seen": 28883045,
      "step": 1370,
      "time_per_iteration": 3.606328248977661
    },
    {
      "auxiliary_loss_clip": 0.01285956,
      "auxiliary_loss_mlp": 0.01033534,
      "balance_loss_clip": 1.06590033,
      "balance_loss_mlp": 1.02439666,
      "epoch": 0.16485300306619372,
      "flos": 24790680357120.0,
      "grad_norm": 1.7908994318781268,
      "language_loss": 0.79382765,
      "learning_rate": 3.8123777412870863e-06,
      "loss": 0.81702256,
      "num_input_tokens_seen": 28902545,
      "step": 1371,
      "time_per_iteration": 3.5357794761657715
    },
    {
      "auxiliary_loss_clip": 0.01331162,
      "auxiliary_loss_mlp": 0.01040135,
      "balance_loss_clip": 1.06000888,
      "balance_loss_mlp": 1.03118217,
      "epoch": 0.1649732459568328,
      "flos": 21106497162240.0,
      "grad_norm": 2.4068089358760885,
      "language_loss": 0.7863096,
      "learning_rate": 3.812048197605643e-06,
      "loss": 0.81002259,
      "num_input_tokens_seen": 28921440,
      "step": 1372,
      "time_per_iteration": 2.7050952911376953
    },
    {
      "auxiliary_loss_clip": 0.01276115,
      "auxiliary_loss_mlp": 0.01029492,
      "balance_loss_clip": 1.06050324,
      "balance_loss_mlp": 1.02070594,
      "epoch": 0.16509348884747188,
      "flos": 20266726118400.0,
      "grad_norm": 3.9038257596586323,
      "language_loss": 0.81834865,
      "learning_rate": 3.8117183790379277e-06,
      "loss": 0.84140474,
      "num_input_tokens_seen": 28939890,
      "step": 1373,
      "time_per_iteration": 2.691882371902466
    },
    {
      "auxiliary_loss_clip": 0.01230745,
      "auxiliary_loss_mlp": 0.01030397,
      "balance_loss_clip": 1.0629456,
      "balance_loss_mlp": 1.02125919,
      "epoch": 0.165213731738111,
      "flos": 11035493602560.0,
      "grad_norm": 3.9364081705900267,
      "language_loss": 0.94427872,
      "learning_rate": 3.811388285633976e-06,
      "loss": 0.9668901,
      "num_input_tokens_seen": 28955875,
      "step": 1374,
      "time_per_iteration": 3.511197090148926
    },
    {
      "auxiliary_loss_clip": 0.01433386,
      "auxiliary_loss_mlp": 0.01035523,
      "balance_loss_clip": 1.05755186,
      "balance_loss_mlp": 1.02650523,
      "epoch": 0.16533397462875007,
      "flos": 29972051268480.0,
      "grad_norm": 5.1406726980669895,
      "language_loss": 0.6191659,
      "learning_rate": 3.811057917443861e-06,
      "loss": 0.64385504,
      "num_input_tokens_seen": 28975140,
      "step": 1375,
      "time_per_iteration": 2.8306100368499756
    },
    {
      "auxiliary_loss_clip": 0.01247254,
      "auxiliary_loss_mlp": 0.01001162,
      "balance_loss_clip": 1.04051113,
      "balance_loss_mlp": 0.99921858,
      "epoch": 0.16545421751938916,
      "flos": 65556763027200.0,
      "grad_norm": 0.8668394399207702,
      "language_loss": 0.68290526,
      "learning_rate": 3.8107272745177e-06,
      "loss": 0.70538938,
      "num_input_tokens_seen": 29047470,
      "step": 1376,
      "time_per_iteration": 3.381443500518799
    },
    {
      "auxiliary_loss_clip": 0.01377092,
      "auxiliary_loss_mlp": 0.01032269,
      "balance_loss_clip": 1.05926013,
      "balance_loss_mlp": 1.02332819,
      "epoch": 0.16557446041002827,
      "flos": 22492361652480.0,
      "grad_norm": 2.109304403719126,
      "language_loss": 0.78849912,
      "learning_rate": 3.8103963569056513e-06,
      "loss": 0.81259263,
      "num_input_tokens_seen": 29066605,
      "step": 1377,
      "time_per_iteration": 2.7423925399780273
    },
    {
      "auxiliary_loss_clip": 0.01323787,
      "auxiliary_loss_mlp": 0.01032525,
      "balance_loss_clip": 1.05692124,
      "balance_loss_mlp": 1.02517319,
      "epoch": 0.16569470330066735,
      "flos": 24602723464320.0,
      "grad_norm": 8.862682285935083,
      "language_loss": 0.88408411,
      "learning_rate": 3.8100651646579146e-06,
      "loss": 0.90764719,
      "num_input_tokens_seen": 29085815,
      "step": 1378,
      "time_per_iteration": 2.6791677474975586
    },
    {
      "auxiliary_loss_clip": 0.01325321,
      "auxiliary_loss_mlp": 0.01036667,
      "balance_loss_clip": 1.05716634,
      "balance_loss_mlp": 1.02731514,
      "epoch": 0.16581494619130643,
      "flos": 15006207588480.0,
      "grad_norm": 2.0908257537622346,
      "language_loss": 0.92784464,
      "learning_rate": 3.8097336978247317e-06,
      "loss": 0.95146447,
      "num_input_tokens_seen": 29102520,
      "step": 1379,
      "time_per_iteration": 2.6184322834014893
    },
    {
      "auxiliary_loss_clip": 0.01321651,
      "auxiliary_loss_mlp": 0.01033068,
      "balance_loss_clip": 1.06055295,
      "balance_loss_mlp": 1.0235846,
      "epoch": 0.16593518908194552,
      "flos": 17420338719360.0,
      "grad_norm": 2.767275555459153,
      "language_loss": 0.88983786,
      "learning_rate": 3.8094019564563854e-06,
      "loss": 0.91338503,
      "num_input_tokens_seen": 29119450,
      "step": 1380,
      "time_per_iteration": 2.704077959060669
    },
    {
      "auxiliary_loss_clip": 0.01232559,
      "auxiliary_loss_mlp": 0.0257866,
      "balance_loss_clip": 1.06525016,
      "balance_loss_mlp": 1.00035286,
      "epoch": 0.16605543197258463,
      "flos": 20412631163520.0,
      "grad_norm": 4.393308060981328,
      "language_loss": 0.75363612,
      "learning_rate": 3.809069940603201e-06,
      "loss": 0.79174829,
      "num_input_tokens_seen": 29137405,
      "step": 1381,
      "time_per_iteration": 2.5840277671813965
    },
    {
      "auxiliary_loss_clip": 0.01318229,
      "auxiliary_loss_mlp": 0.01035074,
      "balance_loss_clip": 1.0578053,
      "balance_loss_mlp": 1.02625895,
      "epoch": 0.1661756748632237,
      "flos": 14209745368320.0,
      "grad_norm": 3.2210893601946053,
      "language_loss": 0.78440088,
      "learning_rate": 3.8087376503155452e-06,
      "loss": 0.80793393,
      "num_input_tokens_seen": 29154890,
      "step": 1382,
      "time_per_iteration": 2.6589772701263428
    },
    {
      "auxiliary_loss_clip": 0.01241837,
      "auxiliary_loss_mlp": 0.01001648,
      "balance_loss_clip": 1.04067314,
      "balance_loss_mlp": 0.99964488,
      "epoch": 0.1662959177538628,
      "flos": 66080877350400.0,
      "grad_norm": 0.9074726612821137,
      "language_loss": 0.56240004,
      "learning_rate": 3.808405085643826e-06,
      "loss": 0.58483493,
      "num_input_tokens_seen": 29219770,
      "step": 1383,
      "time_per_iteration": 3.336135149002075
    },
    {
      "auxiliary_loss_clip": 0.01230112,
      "auxiliary_loss_mlp": 0.02575365,
      "balance_loss_clip": 1.06481862,
      "balance_loss_mlp": 1.0003345,
      "epoch": 0.1664161606445019,
      "flos": 20740567357440.0,
      "grad_norm": 3.1221511844361975,
      "language_loss": 0.89087605,
      "learning_rate": 3.8080722466384925e-06,
      "loss": 0.92893088,
      "num_input_tokens_seen": 29237620,
      "step": 1384,
      "time_per_iteration": 2.752171754837036
    },
    {
      "auxiliary_loss_clip": 0.0123057,
      "auxiliary_loss_mlp": 0.01033335,
      "balance_loss_clip": 1.0617429,
      "balance_loss_mlp": 1.02391756,
      "epoch": 0.166536403535141,
      "flos": 25260930236160.0,
      "grad_norm": 4.016002099062441,
      "language_loss": 0.71142459,
      "learning_rate": 3.8077391333500376e-06,
      "loss": 0.73406363,
      "num_input_tokens_seen": 29256760,
      "step": 1385,
      "time_per_iteration": 2.7324962615966797
    },
    {
      "auxiliary_loss_clip": 0.01270157,
      "auxiliary_loss_mlp": 0.0103126,
      "balance_loss_clip": 1.06391621,
      "balance_loss_mlp": 1.02279055,
      "epoch": 0.16665664642578007,
      "flos": 25447450584960.0,
      "grad_norm": 1.8113989028613198,
      "language_loss": 0.7698428,
      "learning_rate": 3.8074057458289934e-06,
      "loss": 0.79285705,
      "num_input_tokens_seen": 29277450,
      "step": 1386,
      "time_per_iteration": 2.785930633544922
    },
    {
      "auxiliary_loss_clip": 0.01328631,
      "auxiliary_loss_mlp": 0.01033165,
      "balance_loss_clip": 1.0598228,
      "balance_loss_mlp": 1.02447474,
      "epoch": 0.16677688931641918,
      "flos": 22200767043840.0,
      "grad_norm": 2.3134030922763844,
      "language_loss": 0.82510674,
      "learning_rate": 3.807072084125934e-06,
      "loss": 0.84872472,
      "num_input_tokens_seen": 29299300,
      "step": 1387,
      "time_per_iteration": 2.757077217102051
    },
    {
      "auxiliary_loss_clip": 0.01326704,
      "auxiliary_loss_mlp": 0.01034627,
      "balance_loss_clip": 1.06400728,
      "balance_loss_mlp": 1.02567422,
      "epoch": 0.16689713220705826,
      "flos": 16945958776320.0,
      "grad_norm": 17.119619939963457,
      "language_loss": 0.80491465,
      "learning_rate": 3.806738148291477e-06,
      "loss": 0.82852793,
      "num_input_tokens_seen": 29316125,
      "step": 1388,
      "time_per_iteration": 2.6498639583587646
    },
    {
      "auxiliary_loss_clip": 0.01422768,
      "auxiliary_loss_mlp": 0.01034761,
      "balance_loss_clip": 1.05524158,
      "balance_loss_mlp": 1.02642798,
      "epoch": 0.16701737509769735,
      "flos": 36244423923840.0,
      "grad_norm": 2.037399429473232,
      "language_loss": 0.71091229,
      "learning_rate": 3.8064039383762793e-06,
      "loss": 0.73548758,
      "num_input_tokens_seen": 29338490,
      "step": 1389,
      "time_per_iteration": 3.0642402172088623
    },
    {
      "auxiliary_loss_clip": 0.01281916,
      "auxiliary_loss_mlp": 0.01034609,
      "balance_loss_clip": 1.06413352,
      "balance_loss_mlp": 1.02586555,
      "epoch": 0.16713761798833643,
      "flos": 23258659426560.0,
      "grad_norm": 3.6825954905976284,
      "language_loss": 0.76738441,
      "learning_rate": 3.8060694544310396e-06,
      "loss": 0.7905497,
      "num_input_tokens_seen": 29357000,
      "step": 1390,
      "time_per_iteration": 3.0161969661712646
    },
    {
      "auxiliary_loss_clip": 0.01230769,
      "auxiliary_loss_mlp": 0.01034817,
      "balance_loss_clip": 1.06369638,
      "balance_loss_mlp": 1.02537537,
      "epoch": 0.16725786087897554,
      "flos": 25302515207040.0,
      "grad_norm": 2.0100229154452225,
      "language_loss": 0.78872997,
      "learning_rate": 3.8057346965065006e-06,
      "loss": 0.81138581,
      "num_input_tokens_seen": 29378230,
      "step": 1391,
      "time_per_iteration": 2.7953455448150635
    },
    {
      "auxiliary_loss_clip": 0.01327006,
      "auxiliary_loss_mlp": 0.01031508,
      "balance_loss_clip": 1.06265378,
      "balance_loss_mlp": 1.02227533,
      "epoch": 0.16737810376961462,
      "flos": 31831541516160.0,
      "grad_norm": 1.785046352310323,
      "language_loss": 0.8443464,
      "learning_rate": 3.805399664653443e-06,
      "loss": 0.86793154,
      "num_input_tokens_seen": 29400370,
      "step": 1392,
      "time_per_iteration": 4.227256536483765
    },
    {
      "auxiliary_loss_clip": 0.01231523,
      "auxiliary_loss_mlp": 0.01032348,
      "balance_loss_clip": 1.06425273,
      "balance_loss_mlp": 1.02293062,
      "epoch": 0.1674983466602537,
      "flos": 27961843553280.0,
      "grad_norm": 2.6667295247208007,
      "language_loss": 0.74768591,
      "learning_rate": 3.805064358922692e-06,
      "loss": 0.77032471,
      "num_input_tokens_seen": 29418660,
      "step": 1393,
      "time_per_iteration": 2.6381728649139404
    },
    {
      "auxiliary_loss_clip": 0.01283363,
      "auxiliary_loss_mlp": 0.01027359,
      "balance_loss_clip": 1.06229734,
      "balance_loss_mlp": 1.01874006,
      "epoch": 0.16761858955089282,
      "flos": 21762656858880.0,
      "grad_norm": 1.8568149339510323,
      "language_loss": 0.80850488,
      "learning_rate": 3.8047287793651136e-06,
      "loss": 0.83161211,
      "num_input_tokens_seen": 29440105,
      "step": 1394,
      "time_per_iteration": 2.665480613708496
    },
    {
      "auxiliary_loss_clip": 0.01322928,
      "auxiliary_loss_mlp": 0.01039564,
      "balance_loss_clip": 1.06143725,
      "balance_loss_mlp": 1.03037918,
      "epoch": 0.1677388324415319,
      "flos": 23805507058560.0,
      "grad_norm": 1.9310781550355527,
      "language_loss": 0.89056718,
      "learning_rate": 3.8043929260316137e-06,
      "loss": 0.91419208,
      "num_input_tokens_seen": 29458260,
      "step": 1395,
      "time_per_iteration": 2.823216676712036
    },
    {
      "auxiliary_loss_clip": 0.01334578,
      "auxiliary_loss_mlp": 0.01039489,
      "balance_loss_clip": 1.06752706,
      "balance_loss_mlp": 1.03035176,
      "epoch": 0.16785907533217098,
      "flos": 20558859431040.0,
      "grad_norm": 4.065094253931392,
      "language_loss": 0.83532095,
      "learning_rate": 3.8040567989731417e-06,
      "loss": 0.8590616,
      "num_input_tokens_seen": 29476205,
      "step": 1396,
      "time_per_iteration": 2.7605812549591064
    },
    {
      "auxiliary_loss_clip": 0.01274933,
      "auxiliary_loss_mlp": 0.01027778,
      "balance_loss_clip": 1.06297469,
      "balance_loss_mlp": 1.01928997,
      "epoch": 0.16797931822281006,
      "flos": 15669657745920.0,
      "grad_norm": 2.796588744566628,
      "language_loss": 0.79629719,
      "learning_rate": 3.8037203982406876e-06,
      "loss": 0.81932425,
      "num_input_tokens_seen": 29494370,
      "step": 1397,
      "time_per_iteration": 4.5333638191223145
    },
    {
      "auxiliary_loss_clip": 0.01233029,
      "auxiliary_loss_mlp": 0.01033491,
      "balance_loss_clip": 1.06613564,
      "balance_loss_mlp": 1.02441859,
      "epoch": 0.16809956111344918,
      "flos": 16541101607040.0,
      "grad_norm": 2.1847789795572052,
      "language_loss": 0.73215342,
      "learning_rate": 3.8033837238852835e-06,
      "loss": 0.75481856,
      "num_input_tokens_seen": 29511070,
      "step": 1398,
      "time_per_iteration": 2.646944046020508
    },
    {
      "auxiliary_loss_clip": 0.01320538,
      "auxiliary_loss_mlp": 0.01032496,
      "balance_loss_clip": 1.05707765,
      "balance_loss_mlp": 1.02355492,
      "epoch": 0.16821980400408826,
      "flos": 23258084808960.0,
      "grad_norm": 1.7863393374140721,
      "language_loss": 0.69368023,
      "learning_rate": 3.8030467759580017e-06,
      "loss": 0.71721053,
      "num_input_tokens_seen": 29531990,
      "step": 1399,
      "time_per_iteration": 2.7319626808166504
    },
    {
      "auxiliary_loss_clip": 0.01220783,
      "auxiliary_loss_mlp": 0.01033297,
      "balance_loss_clip": 1.06212199,
      "balance_loss_mlp": 1.024261,
      "epoch": 0.16834004689472734,
      "flos": 20774754126720.0,
      "grad_norm": 2.051379853411012,
      "language_loss": 0.87261915,
      "learning_rate": 3.802709554509958e-06,
      "loss": 0.89515996,
      "num_input_tokens_seen": 29549790,
      "step": 1400,
      "time_per_iteration": 3.604795455932617
    },
    {
      "auxiliary_loss_clip": 0.01322843,
      "auxiliary_loss_mlp": 0.01031495,
      "balance_loss_clip": 1.05666697,
      "balance_loss_mlp": 1.02257788,
      "epoch": 0.16846028978536645,
      "flos": 26687302289280.0,
      "grad_norm": 2.956830380752571,
      "language_loss": 0.79395187,
      "learning_rate": 3.8023720595923083e-06,
      "loss": 0.81749523,
      "num_input_tokens_seen": 29569045,
      "step": 1401,
      "time_per_iteration": 2.703732490539551
    },
    {
      "auxiliary_loss_clip": 0.01430256,
      "auxiliary_loss_mlp": 0.01035442,
      "balance_loss_clip": 1.05605042,
      "balance_loss_mlp": 1.0265727,
      "epoch": 0.16858053267600553,
      "flos": 18843298980480.0,
      "grad_norm": 2.2678685078038807,
      "language_loss": 0.87638593,
      "learning_rate": 3.80203429125625e-06,
      "loss": 0.90104294,
      "num_input_tokens_seen": 29587220,
      "step": 1402,
      "time_per_iteration": 2.7590084075927734
    },
    {
      "auxiliary_loss_clip": 0.01412985,
      "auxiliary_loss_mlp": 0.01030071,
      "balance_loss_clip": 1.05732322,
      "balance_loss_mlp": 1.02129138,
      "epoch": 0.16870077556664462,
      "flos": 27744548227200.0,
      "grad_norm": 1.743983132025934,
      "language_loss": 0.70349479,
      "learning_rate": 3.8016962495530225e-06,
      "loss": 0.72792542,
      "num_input_tokens_seen": 29606410,
      "step": 1403,
      "time_per_iteration": 2.830263376235962
    },
    {
      "auxiliary_loss_clip": 0.01233261,
      "auxiliary_loss_mlp": 0.01033845,
      "balance_loss_clip": 1.06473422,
      "balance_loss_mlp": 1.02486277,
      "epoch": 0.1688210184572837,
      "flos": 13730768484480.0,
      "grad_norm": 2.4113715700009255,
      "language_loss": 0.77217418,
      "learning_rate": 3.8013579345339063e-06,
      "loss": 0.79484522,
      "num_input_tokens_seen": 29621275,
      "step": 1404,
      "time_per_iteration": 2.573124885559082
    },
    {
      "auxiliary_loss_clip": 0.01379021,
      "auxiliary_loss_mlp": 0.01036727,
      "balance_loss_clip": 1.05882061,
      "balance_loss_mlp": 1.0263083,
      "epoch": 0.1689412613479228,
      "flos": 26468785900800.0,
      "grad_norm": 3.6265667428202706,
      "language_loss": 0.6984179,
      "learning_rate": 3.801019346250224e-06,
      "loss": 0.72257537,
      "num_input_tokens_seen": 29641420,
      "step": 1405,
      "time_per_iteration": 2.708192825317383
    },
    {
      "auxiliary_loss_clip": 0.01281499,
      "auxiliary_loss_mlp": 0.01030866,
      "balance_loss_clip": 1.0642736,
      "balance_loss_mlp": 1.02175879,
      "epoch": 0.1690615042385619,
      "flos": 21138852337920.0,
      "grad_norm": 2.852237601067135,
      "language_loss": 0.83655792,
      "learning_rate": 3.8006804847533395e-06,
      "loss": 0.85968161,
      "num_input_tokens_seen": 29660935,
      "step": 1406,
      "time_per_iteration": 2.6203277111053467
    },
    {
      "auxiliary_loss_clip": 0.01229686,
      "auxiliary_loss_mlp": 0.01028251,
      "balance_loss_clip": 1.06347167,
      "balance_loss_mlp": 1.01940513,
      "epoch": 0.16918174712920098,
      "flos": 20849340718080.0,
      "grad_norm": 2.2757413123417116,
      "language_loss": 0.85402352,
      "learning_rate": 3.8003413500946556e-06,
      "loss": 0.87660289,
      "num_input_tokens_seen": 29681045,
      "step": 1407,
      "time_per_iteration": 2.523566484451294
    },
    {
      "auxiliary_loss_clip": 0.01332174,
      "auxiliary_loss_mlp": 0.01033824,
      "balance_loss_clip": 1.06249094,
      "balance_loss_mlp": 1.02406645,
      "epoch": 0.1693019900198401,
      "flos": 16983270028800.0,
      "grad_norm": 2.9593504804894266,
      "language_loss": 0.83205879,
      "learning_rate": 3.8000019423256216e-06,
      "loss": 0.85571885,
      "num_input_tokens_seen": 29698810,
      "step": 1408,
      "time_per_iteration": 2.5316426753997803
    },
    {
      "auxiliary_loss_clip": 0.01318934,
      "auxiliary_loss_mlp": 0.01033374,
      "balance_loss_clip": 1.05895126,
      "balance_loss_mlp": 1.02440321,
      "epoch": 0.16942223291047917,
      "flos": 26796901662720.0,
      "grad_norm": 2.3292068967330466,
      "language_loss": 0.87989616,
      "learning_rate": 3.7996622614977234e-06,
      "loss": 0.90341926,
      "num_input_tokens_seen": 29720000,
      "step": 1409,
      "time_per_iteration": 2.6160335540771484
    },
    {
      "auxiliary_loss_clip": 0.01269479,
      "auxiliary_loss_mlp": 0.01031093,
      "balance_loss_clip": 1.06537724,
      "balance_loss_mlp": 1.02196789,
      "epoch": 0.16954247580111825,
      "flos": 18583700411520.0,
      "grad_norm": 2.160624113817759,
      "language_loss": 0.79175413,
      "learning_rate": 3.799322307662492e-06,
      "loss": 0.81475985,
      "num_input_tokens_seen": 29737820,
      "step": 1410,
      "time_per_iteration": 2.6830177307128906
    },
    {
      "auxiliary_loss_clip": 0.01430809,
      "auxiliary_loss_mlp": 0.01034722,
      "balance_loss_clip": 1.05563259,
      "balance_loss_mlp": 1.02517378,
      "epoch": 0.16966271869175734,
      "flos": 13983651210240.0,
      "grad_norm": 2.3744398478380195,
      "language_loss": 0.84154308,
      "learning_rate": 3.798982080871496e-06,
      "loss": 0.86619842,
      "num_input_tokens_seen": 29752960,
      "step": 1411,
      "time_per_iteration": 2.6599204540252686
    },
    {
      "auxiliary_loss_clip": 0.01233852,
      "auxiliary_loss_mlp": 0.01033342,
      "balance_loss_clip": 1.06452203,
      "balance_loss_mlp": 1.02422285,
      "epoch": 0.16978296158239645,
      "flos": 37487328284160.0,
      "grad_norm": 2.3792963468768393,
      "language_loss": 0.68113679,
      "learning_rate": 3.798641581176349e-06,
      "loss": 0.70380872,
      "num_input_tokens_seen": 29775240,
      "step": 1412,
      "time_per_iteration": 2.765388250350952
    },
    {
      "auxiliary_loss_clip": 0.01328943,
      "auxiliary_loss_mlp": 0.01025971,
      "balance_loss_clip": 1.05843067,
      "balance_loss_mlp": 1.01674414,
      "epoch": 0.16990320447303553,
      "flos": 28328958506880.0,
      "grad_norm": 1.8363684498923152,
      "language_loss": 0.7470938,
      "learning_rate": 3.7983008086287044e-06,
      "loss": 0.770643,
      "num_input_tokens_seen": 29796560,
      "step": 1413,
      "time_per_iteration": 2.7862298488616943
    },
    {
      "auxiliary_loss_clip": 0.0132596,
      "auxiliary_loss_mlp": 0.0103402,
      "balance_loss_clip": 1.0577507,
      "balance_loss_mlp": 1.02419114,
      "epoch": 0.1700234473636746,
      "flos": 20188189031040.0,
      "grad_norm": 2.7489870971190875,
      "language_loss": 0.78935724,
      "learning_rate": 3.797959763280257e-06,
      "loss": 0.81295705,
      "num_input_tokens_seen": 29815245,
      "step": 1414,
      "time_per_iteration": 2.8263020515441895
    },
    {
      "auxiliary_loss_clip": 0.01282915,
      "auxiliary_loss_mlp": 0.01035126,
      "balance_loss_clip": 1.06363797,
      "balance_loss_mlp": 1.02631021,
      "epoch": 0.17014369025431372,
      "flos": 24858658846080.0,
      "grad_norm": 2.084845545057407,
      "language_loss": 0.79135329,
      "learning_rate": 3.797618445182743e-06,
      "loss": 0.81453365,
      "num_input_tokens_seen": 29836640,
      "step": 1415,
      "time_per_iteration": 2.7929599285125732
    },
    {
      "auxiliary_loss_clip": 0.01424049,
      "auxiliary_loss_mlp": 0.0103686,
      "balance_loss_clip": 1.05481732,
      "balance_loss_mlp": 1.02799678,
      "epoch": 0.1702639331449528,
      "flos": 16467233287680.0,
      "grad_norm": 2.1060684280672124,
      "language_loss": 0.84544581,
      "learning_rate": 3.79727685438794e-06,
      "loss": 0.87005484,
      "num_input_tokens_seen": 29850830,
      "step": 1416,
      "time_per_iteration": 2.672224521636963
    },
    {
      "auxiliary_loss_clip": 0.01189329,
      "auxiliary_loss_mlp": 0.01009172,
      "balance_loss_clip": 1.03551054,
      "balance_loss_mlp": 1.00720525,
      "epoch": 0.1703841760355919,
      "flos": 52508870979840.0,
      "grad_norm": 0.8327373179116119,
      "language_loss": 0.61593682,
      "learning_rate": 3.796934990947667e-06,
      "loss": 0.63792181,
      "num_input_tokens_seen": 29912515,
      "step": 1417,
      "time_per_iteration": 3.2633559703826904
    },
    {
      "auxiliary_loss_clip": 0.01191734,
      "auxiliary_loss_mlp": 0.01007347,
      "balance_loss_clip": 1.03567338,
      "balance_loss_mlp": 1.00537968,
      "epoch": 0.170504418926231,
      "flos": 49370637576960.0,
      "grad_norm": 0.8797444660114971,
      "language_loss": 0.6249702,
      "learning_rate": 3.7965928549137854e-06,
      "loss": 0.64696097,
      "num_input_tokens_seen": 29969330,
      "step": 1418,
      "time_per_iteration": 4.340043544769287
    },
    {
      "auxiliary_loss_clip": 0.01384555,
      "auxiliary_loss_mlp": 0.01033061,
      "balance_loss_clip": 1.05453324,
      "balance_loss_mlp": 1.02309489,
      "epoch": 0.17062466181687008,
      "flos": 25849219184640.0,
      "grad_norm": 2.0965106122078465,
      "language_loss": 0.7730605,
      "learning_rate": 3.7962504463381953e-06,
      "loss": 0.79723662,
      "num_input_tokens_seen": 29990820,
      "step": 1419,
      "time_per_iteration": 2.8291544914245605
    },
    {
      "auxiliary_loss_clip": 0.01331987,
      "auxiliary_loss_mlp": 0.02585676,
      "balance_loss_clip": 1.06393385,
      "balance_loss_mlp": 1.00024652,
      "epoch": 0.17074490470750917,
      "flos": 20960412549120.0,
      "grad_norm": 1.9377327529497037,
      "language_loss": 0.78539962,
      "learning_rate": 3.7959077652728412e-06,
      "loss": 0.82457626,
      "num_input_tokens_seen": 30009275,
      "step": 1420,
      "time_per_iteration": 2.825260877609253
    },
    {
      "auxiliary_loss_clip": 0.01328875,
      "auxiliary_loss_mlp": 0.01036362,
      "balance_loss_clip": 1.05876064,
      "balance_loss_mlp": 1.02599716,
      "epoch": 0.17086514759814825,
      "flos": 20959766104320.0,
      "grad_norm": 2.467483229887004,
      "language_loss": 0.77355587,
      "learning_rate": 3.795564811769707e-06,
      "loss": 0.79720831,
      "num_input_tokens_seen": 30027630,
      "step": 1421,
      "time_per_iteration": 2.7389471530914307
    },
    {
      "auxiliary_loss_clip": 0.01331775,
      "auxiliary_loss_mlp": 0.0103186,
      "balance_loss_clip": 1.0655421,
      "balance_loss_mlp": 1.02197194,
      "epoch": 0.17098539048878736,
      "flos": 28474073452800.0,
      "grad_norm": 2.062660702623551,
      "language_loss": 0.7816745,
      "learning_rate": 3.795221585880818e-06,
      "loss": 0.80531085,
      "num_input_tokens_seen": 30048310,
      "step": 1422,
      "time_per_iteration": 2.8747854232788086
    },
    {
      "auxiliary_loss_clip": 0.01378656,
      "auxiliary_loss_mlp": 0.01041352,
      "balance_loss_clip": 1.06398749,
      "balance_loss_mlp": 1.03200555,
      "epoch": 0.17110563337942644,
      "flos": 16290014561280.0,
      "grad_norm": 1.9944262246827513,
      "language_loss": 0.91877079,
      "learning_rate": 3.794878087658242e-06,
      "loss": 0.94297087,
      "num_input_tokens_seen": 30066080,
      "step": 1423,
      "time_per_iteration": 4.531185865402222
    },
    {
      "auxiliary_loss_clip": 0.01281077,
      "auxiliary_loss_mlp": 0.01037342,
      "balance_loss_clip": 1.06195426,
      "balance_loss_mlp": 1.02851427,
      "epoch": 0.17122587627006552,
      "flos": 29674207693440.0,
      "grad_norm": 2.0050941179441644,
      "language_loss": 0.78331375,
      "learning_rate": 3.7945343171540873e-06,
      "loss": 0.80649799,
      "num_input_tokens_seen": 30086955,
      "step": 1424,
      "time_per_iteration": 2.7561933994293213
    },
    {
      "auxiliary_loss_clip": 0.01229119,
      "auxiliary_loss_mlp": 0.01030148,
      "balance_loss_clip": 1.06204057,
      "balance_loss_mlp": 1.02064133,
      "epoch": 0.17134611916070464,
      "flos": 25338389915520.0,
      "grad_norm": 1.967433975963455,
      "language_loss": 0.7903229,
      "learning_rate": 3.7941902744205033e-06,
      "loss": 0.81291556,
      "num_input_tokens_seen": 30107990,
      "step": 1425,
      "time_per_iteration": 2.698808193206787
    },
    {
      "auxiliary_loss_clip": 0.01331752,
      "auxiliary_loss_mlp": 0.01032373,
      "balance_loss_clip": 1.05994487,
      "balance_loss_mlp": 1.02282381,
      "epoch": 0.17146636205134372,
      "flos": 13953845900160.0,
      "grad_norm": 1.946977014345416,
      "language_loss": 0.83613271,
      "learning_rate": 3.7938459595096817e-06,
      "loss": 0.85977399,
      "num_input_tokens_seen": 30126535,
      "step": 1426,
      "time_per_iteration": 3.62626576423645
    },
    {
      "auxiliary_loss_clip": 0.01287796,
      "auxiliary_loss_mlp": 0.01039952,
      "balance_loss_clip": 1.06281924,
      "balance_loss_mlp": 1.03043365,
      "epoch": 0.1715866049419828,
      "flos": 23915214172800.0,
      "grad_norm": 2.0153613405672353,
      "language_loss": 0.8643223,
      "learning_rate": 3.7935013724738545e-06,
      "loss": 0.88759983,
      "num_input_tokens_seen": 30147035,
      "step": 1427,
      "time_per_iteration": 2.7154006958007812
    },
    {
      "auxiliary_loss_clip": 0.01274629,
      "auxiliary_loss_mlp": 0.01035458,
      "balance_loss_clip": 1.0625391,
      "balance_loss_mlp": 1.02591562,
      "epoch": 0.17170684783262188,
      "flos": 22709369669760.0,
      "grad_norm": 1.9030502695278126,
      "language_loss": 0.78028899,
      "learning_rate": 3.7931565133652945e-06,
      "loss": 0.80338985,
      "num_input_tokens_seen": 30167110,
      "step": 1428,
      "time_per_iteration": 2.8286454677581787
    },
    {
      "auxiliary_loss_clip": 0.01229737,
      "auxiliary_loss_mlp": 0.01032546,
      "balance_loss_clip": 1.06264019,
      "balance_loss_mlp": 1.02311015,
      "epoch": 0.171827090723261,
      "flos": 26613290315520.0,
      "grad_norm": 2.875852798619069,
      "language_loss": 0.67820776,
      "learning_rate": 3.792811382236317e-06,
      "loss": 0.70083052,
      "num_input_tokens_seen": 30185620,
      "step": 1429,
      "time_per_iteration": 2.724975109100342
    },
    {
      "auxiliary_loss_clip": 0.0128143,
      "auxiliary_loss_mlp": 0.01035698,
      "balance_loss_clip": 1.05968606,
      "balance_loss_mlp": 1.02623856,
      "epoch": 0.17194733361390008,
      "flos": 28148507556480.0,
      "grad_norm": 3.505298686708056,
      "language_loss": 0.7834971,
      "learning_rate": 3.792465979139279e-06,
      "loss": 0.8066684,
      "num_input_tokens_seen": 30208225,
      "step": 1430,
      "time_per_iteration": 2.783107042312622
    },
    {
      "auxiliary_loss_clip": 0.012871,
      "auxiliary_loss_mlp": 0.01002842,
      "balance_loss_clip": 1.03154159,
      "balance_loss_mlp": 1.00069618,
      "epoch": 0.17206757650453916,
      "flos": 65530689753600.0,
      "grad_norm": 0.9437133131710864,
      "language_loss": 0.65640104,
      "learning_rate": 3.792120304126576e-06,
      "loss": 0.67930049,
      "num_input_tokens_seen": 30271600,
      "step": 1431,
      "time_per_iteration": 3.3497447967529297
    },
    {
      "auxiliary_loss_clip": 0.01525756,
      "auxiliary_loss_mlp": 0.01034707,
      "balance_loss_clip": 1.05229616,
      "balance_loss_mlp": 1.02548051,
      "epoch": 0.17218781939517827,
      "flos": 22273486128000.0,
      "grad_norm": 2.0439850660908325,
      "language_loss": 0.83531308,
      "learning_rate": 3.791774357250649e-06,
      "loss": 0.86091769,
      "num_input_tokens_seen": 30290430,
      "step": 1432,
      "time_per_iteration": 3.1649739742279053
    },
    {
      "auxiliary_loss_clip": 0.01327832,
      "auxiliary_loss_mlp": 0.01049302,
      "balance_loss_clip": 1.059618,
      "balance_loss_mlp": 1.03935361,
      "epoch": 0.17230806228581735,
      "flos": 14137313592960.0,
      "grad_norm": 3.6732116521001625,
      "language_loss": 0.79117751,
      "learning_rate": 3.7914281385639757e-06,
      "loss": 0.81494892,
      "num_input_tokens_seen": 30308305,
      "step": 1433,
      "time_per_iteration": 3.1704022884368896
    },
    {
      "auxiliary_loss_clip": 0.01280593,
      "auxiliary_loss_mlp": 0.01030205,
      "balance_loss_clip": 1.05975413,
      "balance_loss_mlp": 1.02079368,
      "epoch": 0.17242830517645644,
      "flos": 20704836303360.0,
      "grad_norm": 2.0697825607142826,
      "language_loss": 0.79114068,
      "learning_rate": 3.7910816481190784e-06,
      "loss": 0.81424862,
      "num_input_tokens_seen": 30328120,
      "step": 1434,
      "time_per_iteration": 2.6796376705169678
    },
    {
      "auxiliary_loss_clip": 0.01324435,
      "auxiliary_loss_mlp": 0.0103235,
      "balance_loss_clip": 1.05942321,
      "balance_loss_mlp": 1.02246141,
      "epoch": 0.17254854806709552,
      "flos": 30774582887040.0,
      "grad_norm": 2.120165652512844,
      "language_loss": 0.75059319,
      "learning_rate": 3.7907348859685193e-06,
      "loss": 0.7741611,
      "num_input_tokens_seen": 30349825,
      "step": 1435,
      "time_per_iteration": 2.8358755111694336
    },
    {
      "auxiliary_loss_clip": 0.01280503,
      "auxiliary_loss_mlp": 0.01037086,
      "balance_loss_clip": 1.06641614,
      "balance_loss_mlp": 1.02813923,
      "epoch": 0.17266879095773463,
      "flos": 26614726859520.0,
      "grad_norm": 2.5968870836870868,
      "language_loss": 0.80766618,
      "learning_rate": 3.790387852164902e-06,
      "loss": 0.83084202,
      "num_input_tokens_seen": 30370555,
      "step": 1436,
      "time_per_iteration": 2.7645745277404785
    },
    {
      "auxiliary_loss_clip": 0.01279847,
      "auxiliary_loss_mlp": 0.01042619,
      "balance_loss_clip": 1.06176698,
      "balance_loss_mlp": 1.03295755,
      "epoch": 0.1727890338483737,
      "flos": 20266295155200.0,
      "grad_norm": 1.9270882459162937,
      "language_loss": 0.76630449,
      "learning_rate": 3.7900405467608707e-06,
      "loss": 0.7895292,
      "num_input_tokens_seen": 30390100,
      "step": 1437,
      "time_per_iteration": 2.7616662979125977
    },
    {
      "auxiliary_loss_clip": 0.01416816,
      "auxiliary_loss_mlp": 0.01037918,
      "balance_loss_clip": 1.05074549,
      "balance_loss_mlp": 1.02867317,
      "epoch": 0.1729092767390128,
      "flos": 18179812909440.0,
      "grad_norm": 3.559547932625824,
      "language_loss": 0.79638648,
      "learning_rate": 3.7896929698091114e-06,
      "loss": 0.82093382,
      "num_input_tokens_seen": 30402915,
      "step": 1438,
      "time_per_iteration": 2.8744747638702393
    },
    {
      "auxiliary_loss_clip": 0.01237108,
      "auxiliary_loss_mlp": 0.01039056,
      "balance_loss_clip": 1.06854546,
      "balance_loss_mlp": 1.02987099,
      "epoch": 0.1730295196296519,
      "flos": 26759518583040.0,
      "grad_norm": 6.135528435683419,
      "language_loss": 0.67972445,
      "learning_rate": 3.7893451213623518e-06,
      "loss": 0.7024861,
      "num_input_tokens_seen": 30420145,
      "step": 1439,
      "time_per_iteration": 2.710486650466919
    },
    {
      "auxiliary_loss_clip": 0.01283155,
      "auxiliary_loss_mlp": 0.02582461,
      "balance_loss_clip": 1.06663108,
      "balance_loss_mlp": 1.00024033,
      "epoch": 0.173149762520291,
      "flos": 23842531002240.0,
      "grad_norm": 3.6494198189163,
      "language_loss": 0.82770598,
      "learning_rate": 3.7889970014733606e-06,
      "loss": 0.86636209,
      "num_input_tokens_seen": 30439250,
      "step": 1440,
      "time_per_iteration": 2.7459568977355957
    },
    {
      "auxiliary_loss_clip": 0.01416156,
      "auxiliary_loss_mlp": 0.01029651,
      "balance_loss_clip": 1.05253935,
      "balance_loss_mlp": 1.0205791,
      "epoch": 0.17327000541093007,
      "flos": 23368186972800.0,
      "grad_norm": 1.9110900615034798,
      "language_loss": 0.77854097,
      "learning_rate": 3.7886486101949463e-06,
      "loss": 0.80299902,
      "num_input_tokens_seen": 30460430,
      "step": 1441,
      "time_per_iteration": 2.8128662109375
    },
    {
      "auxiliary_loss_clip": 0.01359489,
      "auxiliary_loss_mlp": 0.01033202,
      "balance_loss_clip": 1.05326009,
      "balance_loss_mlp": 1.02501774,
      "epoch": 0.17339024830156918,
      "flos": 18221290139520.0,
      "grad_norm": 7.54654069924172,
      "language_loss": 0.88405758,
      "learning_rate": 3.7882999475799594e-06,
      "loss": 0.9079845,
      "num_input_tokens_seen": 30478465,
      "step": 1442,
      "time_per_iteration": 2.762700319290161
    },
    {
      "auxiliary_loss_clip": 0.01411436,
      "auxiliary_loss_mlp": 0.01032719,
      "balance_loss_clip": 1.05333281,
      "balance_loss_mlp": 1.02327204,
      "epoch": 0.17351049119220827,
      "flos": 23332024955520.0,
      "grad_norm": 2.0511615383156094,
      "language_loss": 0.81366193,
      "learning_rate": 3.787951013681293e-06,
      "loss": 0.83810353,
      "num_input_tokens_seen": 30496510,
      "step": 1443,
      "time_per_iteration": 2.786680221557617
    },
    {
      "auxiliary_loss_clip": 0.01279086,
      "auxiliary_loss_mlp": 0.010364,
      "balance_loss_clip": 1.05981517,
      "balance_loss_mlp": 1.02626693,
      "epoch": 0.17363073408284735,
      "flos": 23803495896960.0,
      "grad_norm": 2.8292372765244984,
      "language_loss": 0.77748531,
      "learning_rate": 3.787601808551879e-06,
      "loss": 0.80064017,
      "num_input_tokens_seen": 30516325,
      "step": 1444,
      "time_per_iteration": 4.357468605041504
    },
    {
      "auxiliary_loss_clip": 0.0131381,
      "auxiliary_loss_mlp": 0.01041944,
      "balance_loss_clip": 1.05935144,
      "balance_loss_mlp": 1.0326097,
      "epoch": 0.17375097697348643,
      "flos": 18515290959360.0,
      "grad_norm": 2.3494301329433216,
      "language_loss": 0.84023106,
      "learning_rate": 3.7872523322446926e-06,
      "loss": 0.8637886,
      "num_input_tokens_seen": 30535210,
      "step": 1445,
      "time_per_iteration": 2.7460668087005615
    },
    {
      "auxiliary_loss_clip": 0.01429757,
      "auxiliary_loss_mlp": 0.01032326,
      "balance_loss_clip": 1.05117369,
      "balance_loss_mlp": 1.02283072,
      "epoch": 0.17387121986412554,
      "flos": 38877897456000.0,
      "grad_norm": 2.158983530936075,
      "language_loss": 0.60267901,
      "learning_rate": 3.7869025848127478e-06,
      "loss": 0.62729985,
      "num_input_tokens_seen": 30559405,
      "step": 1446,
      "time_per_iteration": 2.9756698608398438
    },
    {
      "auxiliary_loss_clip": 0.01280821,
      "auxiliary_loss_mlp": 0.01034756,
      "balance_loss_clip": 1.0599947,
      "balance_loss_mlp": 1.02549386,
      "epoch": 0.17399146275476463,
      "flos": 20375714960640.0,
      "grad_norm": 2.3422628008226907,
      "language_loss": 0.80590487,
      "learning_rate": 3.786552566309102e-06,
      "loss": 0.82906067,
      "num_input_tokens_seen": 30577615,
      "step": 1447,
      "time_per_iteration": 2.6814067363739014
    },
    {
      "auxiliary_loss_clip": 0.01323468,
      "auxiliary_loss_mlp": 0.02577754,
      "balance_loss_clip": 1.06352997,
      "balance_loss_mlp": 1.00016236,
      "epoch": 0.1741117056454037,
      "flos": 19164339763200.0,
      "grad_norm": 2.4694341390728476,
      "language_loss": 0.86268437,
      "learning_rate": 3.7862022767868517e-06,
      "loss": 0.90169656,
      "num_input_tokens_seen": 30595205,
      "step": 1448,
      "time_per_iteration": 2.6798505783081055
    },
    {
      "auxiliary_loss_clip": 0.01380991,
      "auxiliary_loss_mlp": 0.01035738,
      "balance_loss_clip": 1.06357431,
      "balance_loss_mlp": 1.02664852,
      "epoch": 0.17423194853604282,
      "flos": 25374300537600.0,
      "grad_norm": 2.309479643719411,
      "language_loss": 0.84343112,
      "learning_rate": 3.7858517162991367e-06,
      "loss": 0.86759841,
      "num_input_tokens_seen": 30615280,
      "step": 1449,
      "time_per_iteration": 4.5342793464660645
    },
    {
      "auxiliary_loss_clip": 0.01381007,
      "auxiliary_loss_mlp": 0.01037722,
      "balance_loss_clip": 1.05650949,
      "balance_loss_mlp": 1.02828097,
      "epoch": 0.1743521914266819,
      "flos": 25191874339200.0,
      "grad_norm": 2.428141744567282,
      "language_loss": 0.60936451,
      "learning_rate": 3.7855008848991363e-06,
      "loss": 0.63355184,
      "num_input_tokens_seen": 30633485,
      "step": 1450,
      "time_per_iteration": 2.87052059173584
    },
    {
      "auxiliary_loss_clip": 0.01324713,
      "auxiliary_loss_mlp": 0.01032673,
      "balance_loss_clip": 1.06288934,
      "balance_loss_mlp": 1.02365494,
      "epoch": 0.17447243431732098,
      "flos": 25666577504640.0,
      "grad_norm": 1.9383111455127349,
      "language_loss": 0.77895725,
      "learning_rate": 3.7851497826400714e-06,
      "loss": 0.80253106,
      "num_input_tokens_seen": 30653625,
      "step": 1451,
      "time_per_iteration": 2.9122369289398193
    },
    {
      "auxiliary_loss_clip": 0.01230362,
      "auxiliary_loss_mlp": 0.0103688,
      "balance_loss_clip": 1.06435442,
      "balance_loss_mlp": 1.02710462,
      "epoch": 0.17459267720796007,
      "flos": 36281950657920.0,
      "grad_norm": 1.8451749998410498,
      "language_loss": 0.76041055,
      "learning_rate": 3.7847984095752034e-06,
      "loss": 0.78308296,
      "num_input_tokens_seen": 30677080,
      "step": 1452,
      "time_per_iteration": 3.7654407024383545
    },
    {
      "auxiliary_loss_clip": 0.01226718,
      "auxiliary_loss_mlp": 0.01030387,
      "balance_loss_clip": 1.06063378,
      "balance_loss_mlp": 1.02087402,
      "epoch": 0.17471292009859918,
      "flos": 20011113959040.0,
      "grad_norm": 2.2462920665764208,
      "language_loss": 0.80244434,
      "learning_rate": 3.784446765757836e-06,
      "loss": 0.82501543,
      "num_input_tokens_seen": 30695725,
      "step": 1453,
      "time_per_iteration": 2.63551926612854
    },
    {
      "auxiliary_loss_clip": 0.0136218,
      "auxiliary_loss_mlp": 0.01032053,
      "balance_loss_clip": 1.05565524,
      "balance_loss_mlp": 1.02250457,
      "epoch": 0.17483316298923826,
      "flos": 27819242559360.0,
      "grad_norm": 5.9131343569252115,
      "language_loss": 0.77814299,
      "learning_rate": 3.7840948512413133e-06,
      "loss": 0.80208534,
      "num_input_tokens_seen": 30713310,
      "step": 1454,
      "time_per_iteration": 2.7685084342956543
    },
    {
      "auxiliary_loss_clip": 0.01377878,
      "auxiliary_loss_mlp": 0.01036984,
      "balance_loss_clip": 1.06110775,
      "balance_loss_mlp": 1.02738714,
      "epoch": 0.17495340587987734,
      "flos": 44017934791680.0,
      "grad_norm": 1.8663053936351666,
      "language_loss": 0.78624678,
      "learning_rate": 3.7837426660790196e-06,
      "loss": 0.81039536,
      "num_input_tokens_seen": 30734725,
      "step": 1455,
      "time_per_iteration": 2.924888849258423
    },
    {
      "auxiliary_loss_clip": 0.01226096,
      "auxiliary_loss_mlp": 0.01035743,
      "balance_loss_clip": 1.06146288,
      "balance_loss_mlp": 1.02673697,
      "epoch": 0.17507364877051645,
      "flos": 20885825957760.0,
      "grad_norm": 2.1751820922183778,
      "language_loss": 0.82056499,
      "learning_rate": 3.783390210324382e-06,
      "loss": 0.8431834,
      "num_input_tokens_seen": 30754450,
      "step": 1456,
      "time_per_iteration": 2.6471166610717773
    },
    {
      "auxiliary_loss_clip": 0.01309027,
      "auxiliary_loss_mlp": 0.01029267,
      "balance_loss_clip": 1.05836439,
      "balance_loss_mlp": 1.02049327,
      "epoch": 0.17519389166115554,
      "flos": 24717602136960.0,
      "grad_norm": 1.9389041578147124,
      "language_loss": 0.72767115,
      "learning_rate": 3.7830374840308676e-06,
      "loss": 0.75105405,
      "num_input_tokens_seen": 30774605,
      "step": 1457,
      "time_per_iteration": 2.7452762126922607
    },
    {
      "auxiliary_loss_clip": 0.01279798,
      "auxiliary_loss_mlp": 0.01030999,
      "balance_loss_clip": 1.06423616,
      "balance_loss_mlp": 1.02156973,
      "epoch": 0.17531413455179462,
      "flos": 23798144770560.0,
      "grad_norm": 3.0841208009712586,
      "language_loss": 0.82497811,
      "learning_rate": 3.7826844872519842e-06,
      "loss": 0.84808606,
      "num_input_tokens_seen": 30792460,
      "step": 1458,
      "time_per_iteration": 2.6837852001190186
    },
    {
      "auxiliary_loss_clip": 0.01324512,
      "auxiliary_loss_mlp": 0.0103364,
      "balance_loss_clip": 1.06009066,
      "balance_loss_mlp": 1.02422881,
      "epoch": 0.1754343774424337,
      "flos": 24572379450240.0,
      "grad_norm": 5.234716046956519,
      "language_loss": 0.73220539,
      "learning_rate": 3.782331220041282e-06,
      "loss": 0.75578701,
      "num_input_tokens_seen": 30812525,
      "step": 1459,
      "time_per_iteration": 2.7159008979797363
    },
    {
      "auxiliary_loss_clip": 0.01322,
      "auxiliary_loss_mlp": 0.01032036,
      "balance_loss_clip": 1.05945849,
      "balance_loss_mlp": 1.02302384,
      "epoch": 0.17555462033307281,
      "flos": 18114599767680.0,
      "grad_norm": 2.3606295721581856,
      "language_loss": 0.82972789,
      "learning_rate": 3.7819776824523504e-06,
      "loss": 0.85326827,
      "num_input_tokens_seen": 30830390,
      "step": 1460,
      "time_per_iteration": 2.7380359172821045
    },
    {
      "auxiliary_loss_clip": 0.01336112,
      "auxiliary_loss_mlp": 0.01030694,
      "balance_loss_clip": 1.05964303,
      "balance_loss_mlp": 1.02169955,
      "epoch": 0.1756748632237119,
      "flos": 28366018364160.0,
      "grad_norm": 2.092563332977659,
      "language_loss": 0.8420521,
      "learning_rate": 3.7816238745388213e-06,
      "loss": 0.86572015,
      "num_input_tokens_seen": 30849935,
      "step": 1461,
      "time_per_iteration": 2.7182483673095703
    },
    {
      "auxiliary_loss_clip": 0.0133174,
      "auxiliary_loss_mlp": 0.01037427,
      "balance_loss_clip": 1.05693817,
      "balance_loss_mlp": 1.02833712,
      "epoch": 0.17579510611435098,
      "flos": 25732939881600.0,
      "grad_norm": 2.7356513669948233,
      "language_loss": 0.87058604,
      "learning_rate": 3.781269796354367e-06,
      "loss": 0.89427769,
      "num_input_tokens_seen": 30869555,
      "step": 1462,
      "time_per_iteration": 2.72864031791687
    },
    {
      "auxiliary_loss_clip": 0.01327602,
      "auxiliary_loss_mlp": 0.01038699,
      "balance_loss_clip": 1.05981028,
      "balance_loss_mlp": 1.02959692,
      "epoch": 0.1759153490049901,
      "flos": 18588081870720.0,
      "grad_norm": 6.5309277094432066,
      "language_loss": 0.86036015,
      "learning_rate": 3.7809154479527006e-06,
      "loss": 0.88402319,
      "num_input_tokens_seen": 30888760,
      "step": 1463,
      "time_per_iteration": 2.706270933151245
    },
    {
      "auxiliary_loss_clip": 0.01370317,
      "auxiliary_loss_mlp": 0.01030407,
      "balance_loss_clip": 1.05862761,
      "balance_loss_mlp": 1.02120447,
      "epoch": 0.17603559189562917,
      "flos": 18619323724800.0,
      "grad_norm": 2.482541566122146,
      "language_loss": 0.84143746,
      "learning_rate": 3.780560829387577e-06,
      "loss": 0.86544466,
      "num_input_tokens_seen": 30907260,
      "step": 1464,
      "time_per_iteration": 2.7423558235168457
    },
    {
      "auxiliary_loss_clip": 0.01183969,
      "auxiliary_loss_mlp": 0.01001732,
      "balance_loss_clip": 1.03580415,
      "balance_loss_mlp": 0.99988431,
      "epoch": 0.17615583478626826,
      "flos": 60530775373440.0,
      "grad_norm": 0.8550779020794265,
      "language_loss": 0.57860905,
      "learning_rate": 3.7802059407127915e-06,
      "loss": 0.60046607,
      "num_input_tokens_seen": 30965810,
      "step": 1465,
      "time_per_iteration": 3.272942066192627
    },
    {
      "auxiliary_loss_clip": 0.01325191,
      "auxiliary_loss_mlp": 0.01035508,
      "balance_loss_clip": 1.05676091,
      "balance_loss_mlp": 1.02654338,
      "epoch": 0.17627607767690734,
      "flos": 23616221362560.0,
      "grad_norm": 2.75144633195439,
      "language_loss": 0.86459816,
      "learning_rate": 3.7798507819821797e-06,
      "loss": 0.88820517,
      "num_input_tokens_seen": 30982935,
      "step": 1466,
      "time_per_iteration": 2.75236177444458
    },
    {
      "auxiliary_loss_clip": 0.01368655,
      "auxiliary_loss_mlp": 0.01034988,
      "balance_loss_clip": 1.05847907,
      "balance_loss_mlp": 1.02532029,
      "epoch": 0.17639632056754645,
      "flos": 17639070589440.0,
      "grad_norm": 2.7058859849176984,
      "language_loss": 0.79429662,
      "learning_rate": 3.7794953532496197e-06,
      "loss": 0.81833303,
      "num_input_tokens_seen": 30998840,
      "step": 1467,
      "time_per_iteration": 2.7081191539764404
    },
    {
      "auxiliary_loss_clip": 0.01328593,
      "auxiliary_loss_mlp": 0.02533877,
      "balance_loss_clip": 1.0282867,
      "balance_loss_mlp": 1.00021386,
      "epoch": 0.17651656345818553,
      "flos": 57932604910080.0,
      "grad_norm": 0.8551365450203078,
      "language_loss": 0.57899261,
      "learning_rate": 3.7791396545690295e-06,
      "loss": 0.61761737,
      "num_input_tokens_seen": 31060075,
      "step": 1468,
      "time_per_iteration": 3.3379290103912354
    },
    {
      "auxiliary_loss_clip": 0.01279255,
      "auxiliary_loss_mlp": 0.0103494,
      "balance_loss_clip": 1.06454599,
      "balance_loss_mlp": 1.0253377,
      "epoch": 0.17663680634882462,
      "flos": 22929502170240.0,
      "grad_norm": 4.09781220569232,
      "language_loss": 0.80684823,
      "learning_rate": 3.7787836859943685e-06,
      "loss": 0.82999021,
      "num_input_tokens_seen": 31078800,
      "step": 1469,
      "time_per_iteration": 2.825531244277954
    },
    {
      "auxiliary_loss_clip": 0.01279296,
      "auxiliary_loss_mlp": 0.01029743,
      "balance_loss_clip": 1.06315148,
      "balance_loss_mlp": 1.02018869,
      "epoch": 0.17675704923946373,
      "flos": 22637979388800.0,
      "grad_norm": 3.454837933064007,
      "language_loss": 0.79401231,
      "learning_rate": 3.7784274475796363e-06,
      "loss": 0.81710267,
      "num_input_tokens_seen": 31097430,
      "step": 1470,
      "time_per_iteration": 3.9424755573272705
    },
    {
      "auxiliary_loss_clip": 0.0131137,
      "auxiliary_loss_mlp": 0.01037808,
      "balance_loss_clip": 1.05769062,
      "balance_loss_mlp": 1.0275383,
      "epoch": 0.1768772921301028,
      "flos": 27126525795840.0,
      "grad_norm": 2.175526359020434,
      "language_loss": 0.76652741,
      "learning_rate": 3.7780709393788745e-06,
      "loss": 0.79001927,
      "num_input_tokens_seen": 31117905,
      "step": 1471,
      "time_per_iteration": 2.7441909313201904
    },
    {
      "auxiliary_loss_clip": 0.01231478,
      "auxiliary_loss_mlp": 0.01041386,
      "balance_loss_clip": 1.06627774,
      "balance_loss_mlp": 1.03264189,
      "epoch": 0.1769975350207419,
      "flos": 19172133014400.0,
      "grad_norm": 3.984130105632141,
      "language_loss": 0.75012577,
      "learning_rate": 3.777714161446165e-06,
      "loss": 0.77285445,
      "num_input_tokens_seen": 31137610,
      "step": 1472,
      "time_per_iteration": 2.667877197265625
    },
    {
      "auxiliary_loss_clip": 0.0128352,
      "auxiliary_loss_mlp": 0.01036374,
      "balance_loss_clip": 1.06637788,
      "balance_loss_mlp": 1.02768397,
      "epoch": 0.177117777911381,
      "flos": 36134932291200.0,
      "grad_norm": 3.441428095454451,
      "language_loss": 0.69460058,
      "learning_rate": 3.7773571138356304e-06,
      "loss": 0.71779954,
      "num_input_tokens_seen": 31157780,
      "step": 1473,
      "time_per_iteration": 2.7804720401763916
    },
    {
      "auxiliary_loss_clip": 0.01417972,
      "auxiliary_loss_mlp": 0.010243,
      "balance_loss_clip": 1.05823505,
      "balance_loss_mlp": 1.01508522,
      "epoch": 0.17723802080202009,
      "flos": 22090593052800.0,
      "grad_norm": 2.93087618770942,
      "language_loss": 0.89364392,
      "learning_rate": 3.776999796601435e-06,
      "loss": 0.91806662,
      "num_input_tokens_seen": 31176540,
      "step": 1474,
      "time_per_iteration": 2.779010772705078
    },
    {
      "auxiliary_loss_clip": 0.01282214,
      "auxiliary_loss_mlp": 0.01033518,
      "balance_loss_clip": 1.06229401,
      "balance_loss_mlp": 1.02456534,
      "epoch": 0.17735826369265917,
      "flos": 30222671437440.0,
      "grad_norm": 2.0777515325293487,
      "language_loss": 0.72702146,
      "learning_rate": 3.776642209797783e-06,
      "loss": 0.75017881,
      "num_input_tokens_seen": 31198370,
      "step": 1475,
      "time_per_iteration": 4.547362804412842
    },
    {
      "auxiliary_loss_clip": 0.0127396,
      "auxiliary_loss_mlp": 0.01030903,
      "balance_loss_clip": 1.05999541,
      "balance_loss_mlp": 1.02158737,
      "epoch": 0.17747850658329825,
      "flos": 21397588980480.0,
      "grad_norm": 2.598507690838947,
      "language_loss": 0.78010935,
      "learning_rate": 3.7762843534789205e-06,
      "loss": 0.80315799,
      "num_input_tokens_seen": 31217120,
      "step": 1476,
      "time_per_iteration": 2.6544530391693115
    },
    {
      "auxiliary_loss_clip": 0.01333503,
      "auxiliary_loss_mlp": 0.01042011,
      "balance_loss_clip": 1.05945373,
      "balance_loss_mlp": 1.03259945,
      "epoch": 0.17759874947393736,
      "flos": 16983341856000.0,
      "grad_norm": 2.9584299073374756,
      "language_loss": 0.8837437,
      "learning_rate": 3.7759262276991343e-06,
      "loss": 0.90749884,
      "num_input_tokens_seen": 31234730,
      "step": 1477,
      "time_per_iteration": 2.7369773387908936
    },
    {
      "auxiliary_loss_clip": 0.01336548,
      "auxiliary_loss_mlp": 0.01033024,
      "balance_loss_clip": 1.06436145,
      "balance_loss_mlp": 1.02320075,
      "epoch": 0.17771899236457644,
      "flos": 11546107390080.0,
      "grad_norm": 2.3484327079737475,
      "language_loss": 0.81126523,
      "learning_rate": 3.7755678325127506e-06,
      "loss": 0.83496094,
      "num_input_tokens_seen": 31252410,
      "step": 1478,
      "time_per_iteration": 3.569138765335083
    },
    {
      "auxiliary_loss_clip": 0.01424315,
      "auxiliary_loss_mlp": 0.0103567,
      "balance_loss_clip": 1.06265521,
      "balance_loss_mlp": 1.02626491,
      "epoch": 0.17783923525521553,
      "flos": 18807747494400.0,
      "grad_norm": 1.706897067019515,
      "language_loss": 0.75820994,
      "learning_rate": 3.7752091679741393e-06,
      "loss": 0.78280979,
      "num_input_tokens_seen": 31270200,
      "step": 1479,
      "time_per_iteration": 2.767169237136841
    },
    {
      "auxiliary_loss_clip": 0.01278074,
      "auxiliary_loss_mlp": 0.01032693,
      "balance_loss_clip": 1.06252933,
      "balance_loss_mlp": 1.02276897,
      "epoch": 0.17795947814585464,
      "flos": 30408365773440.0,
      "grad_norm": 3.2580270892307,
      "language_loss": 0.7714209,
      "learning_rate": 3.774850234137708e-06,
      "loss": 0.7945286,
      "num_input_tokens_seen": 31287495,
      "step": 1480,
      "time_per_iteration": 2.746973752975464
    },
    {
      "auxiliary_loss_clip": 0.01277366,
      "auxiliary_loss_mlp": 0.01033607,
      "balance_loss_clip": 1.06201053,
      "balance_loss_mlp": 1.02490473,
      "epoch": 0.17807972103649372,
      "flos": 24389055411840.0,
      "grad_norm": 2.2510960557049056,
      "language_loss": 0.82458127,
      "learning_rate": 3.7744910310579076e-06,
      "loss": 0.84769094,
      "num_input_tokens_seen": 31306420,
      "step": 1481,
      "time_per_iteration": 2.6865270137786865
    },
    {
      "auxiliary_loss_clip": 0.01228513,
      "auxiliary_loss_mlp": 0.01032905,
      "balance_loss_clip": 1.06552815,
      "balance_loss_mlp": 1.02451277,
      "epoch": 0.1781999639271328,
      "flos": 20301559332480.0,
      "grad_norm": 2.0675181027770204,
      "language_loss": 0.84833181,
      "learning_rate": 3.774131558789229e-06,
      "loss": 0.87094605,
      "num_input_tokens_seen": 31325750,
      "step": 1482,
      "time_per_iteration": 2.6404054164886475
    },
    {
      "auxiliary_loss_clip": 0.01230497,
      "auxiliary_loss_mlp": 0.0257497,
      "balance_loss_clip": 1.06610131,
      "balance_loss_mlp": 1.00002873,
      "epoch": 0.1783202068177719,
      "flos": 15924479806080.0,
      "grad_norm": 2.375245839528329,
      "language_loss": 0.69722152,
      "learning_rate": 3.773771817386203e-06,
      "loss": 0.73527616,
      "num_input_tokens_seen": 31343080,
      "step": 1483,
      "time_per_iteration": 2.6037697792053223
    },
    {
      "auxiliary_loss_clip": 0.01324356,
      "auxiliary_loss_mlp": 0.01032946,
      "balance_loss_clip": 1.05843139,
      "balance_loss_mlp": 1.02367175,
      "epoch": 0.178440449708411,
      "flos": 20631758083200.0,
      "grad_norm": 1.7506668385089612,
      "language_loss": 0.79284656,
      "learning_rate": 3.773411806903403e-06,
      "loss": 0.8164196,
      "num_input_tokens_seen": 31362160,
      "step": 1484,
      "time_per_iteration": 2.7362802028656006
    },
    {
      "auxiliary_loss_clip": 0.01415006,
      "auxiliary_loss_mlp": 0.01030874,
      "balance_loss_clip": 1.05692434,
      "balance_loss_mlp": 1.02250516,
      "epoch": 0.17856069259905008,
      "flos": 21686059105920.0,
      "grad_norm": 2.2418675071114933,
      "language_loss": 0.94888389,
      "learning_rate": 3.7730515273954415e-06,
      "loss": 0.97334266,
      "num_input_tokens_seen": 31380770,
      "step": 1485,
      "time_per_iteration": 2.820146083831787
    },
    {
      "auxiliary_loss_clip": 0.01227595,
      "auxiliary_loss_mlp": 0.01036676,
      "balance_loss_clip": 1.06530988,
      "balance_loss_mlp": 1.02785408,
      "epoch": 0.17868093548968916,
      "flos": 26572962320640.0,
      "grad_norm": 2.145603132708544,
      "language_loss": 0.84891021,
      "learning_rate": 3.772690978916973e-06,
      "loss": 0.87155294,
      "num_input_tokens_seen": 31400525,
      "step": 1486,
      "time_per_iteration": 2.7018401622772217
    },
    {
      "auxiliary_loss_clip": 0.01277051,
      "auxiliary_loss_mlp": 0.01032149,
      "balance_loss_clip": 1.06218028,
      "balance_loss_mlp": 1.02205753,
      "epoch": 0.17880117838032827,
      "flos": 18581006891520.0,
      "grad_norm": 2.146817649113855,
      "language_loss": 0.86520934,
      "learning_rate": 3.772330161522693e-06,
      "loss": 0.88830137,
      "num_input_tokens_seen": 31418435,
      "step": 1487,
      "time_per_iteration": 2.5915093421936035
    },
    {
      "auxiliary_loss_clip": 0.01325626,
      "auxiliary_loss_mlp": 0.01031545,
      "balance_loss_clip": 1.06602716,
      "balance_loss_mlp": 1.02179337,
      "epoch": 0.17892142127096736,
      "flos": 26541217676160.0,
      "grad_norm": 2.1058117640023535,
      "language_loss": 0.80232644,
      "learning_rate": 3.7719690752673365e-06,
      "loss": 0.82589811,
      "num_input_tokens_seen": 31439230,
      "step": 1488,
      "time_per_iteration": 2.6561591625213623
    },
    {
      "auxiliary_loss_clip": 0.01376163,
      "auxiliary_loss_mlp": 0.01030897,
      "balance_loss_clip": 1.06050229,
      "balance_loss_mlp": 1.02243304,
      "epoch": 0.17904166416160644,
      "flos": 23872623621120.0,
      "grad_norm": 2.182595468529718,
      "language_loss": 0.78594196,
      "learning_rate": 3.7716077202056796e-06,
      "loss": 0.81001258,
      "num_input_tokens_seen": 31457705,
      "step": 1489,
      "time_per_iteration": 2.676953077316284
    },
    {
      "auxiliary_loss_clip": 0.01319287,
      "auxiliary_loss_mlp": 0.01036791,
      "balance_loss_clip": 1.06159067,
      "balance_loss_mlp": 1.02798772,
      "epoch": 0.17916190705224552,
      "flos": 19134426712320.0,
      "grad_norm": 2.4980134345616536,
      "language_loss": 0.93483341,
      "learning_rate": 3.7712460963925404e-06,
      "loss": 0.95839423,
      "num_input_tokens_seen": 31473645,
      "step": 1490,
      "time_per_iteration": 2.73061203956604
    },
    {
      "auxiliary_loss_clip": 0.01325806,
      "auxiliary_loss_mlp": 0.01028861,
      "balance_loss_clip": 1.06025743,
      "balance_loss_mlp": 1.02021813,
      "epoch": 0.17928214994288463,
      "flos": 25152120961920.0,
      "grad_norm": 21.113277242333755,
      "language_loss": 0.75687009,
      "learning_rate": 3.7708842038827775e-06,
      "loss": 0.78041679,
      "num_input_tokens_seen": 31492605,
      "step": 1491,
      "time_per_iteration": 2.732940435409546
    },
    {
      "auxiliary_loss_clip": 0.01275935,
      "auxiliary_loss_mlp": 0.01037,
      "balance_loss_clip": 1.06141734,
      "balance_loss_mlp": 1.02812457,
      "epoch": 0.17940239283352372,
      "flos": 22384629786240.0,
      "grad_norm": 1.9439926822514368,
      "language_loss": 0.86165112,
      "learning_rate": 3.770522042731288e-06,
      "loss": 0.88478047,
      "num_input_tokens_seen": 31514500,
      "step": 1492,
      "time_per_iteration": 2.774278402328491
    },
    {
      "auxiliary_loss_clip": 0.01420072,
      "auxiliary_loss_mlp": 0.01037153,
      "balance_loss_clip": 1.0578171,
      "balance_loss_mlp": 1.02764058,
      "epoch": 0.1795226357241628,
      "flos": 23178685795200.0,
      "grad_norm": 2.4550559052659797,
      "language_loss": 0.88114321,
      "learning_rate": 3.7701596129930122e-06,
      "loss": 0.90571541,
      "num_input_tokens_seen": 31533225,
      "step": 1493,
      "time_per_iteration": 2.791640520095825
    },
    {
      "auxiliary_loss_clip": 0.01315812,
      "auxiliary_loss_mlp": 0.01036042,
      "balance_loss_clip": 1.06088328,
      "balance_loss_mlp": 1.02698779,
      "epoch": 0.1796428786148019,
      "flos": 22090413484800.0,
      "grad_norm": 2.2327119589176863,
      "language_loss": 0.73243594,
      "learning_rate": 3.7697969147229315e-06,
      "loss": 0.7559545,
      "num_input_tokens_seen": 31551385,
      "step": 1494,
      "time_per_iteration": 2.7167084217071533
    },
    {
      "auxiliary_loss_clip": 0.01279408,
      "auxiliary_loss_mlp": 0.01035046,
      "balance_loss_clip": 1.06406069,
      "balance_loss_mlp": 1.02680862,
      "epoch": 0.179763121505441,
      "flos": 21324618501120.0,
      "grad_norm": 2.603731084848324,
      "language_loss": 0.85635018,
      "learning_rate": 3.7694339479760647e-06,
      "loss": 0.87949467,
      "num_input_tokens_seen": 31570415,
      "step": 1495,
      "time_per_iteration": 2.6924164295196533
    },
    {
      "auxiliary_loss_clip": 0.01235765,
      "auxiliary_loss_mlp": 0.01000301,
      "balance_loss_clip": 1.03686571,
      "balance_loss_mlp": 0.99840564,
      "epoch": 0.17988336439608008,
      "flos": 68161864815360.0,
      "grad_norm": 0.7704122207748717,
      "language_loss": 0.57317054,
      "learning_rate": 3.769070712807476e-06,
      "loss": 0.59553123,
      "num_input_tokens_seen": 31632445,
      "step": 1496,
      "time_per_iteration": 4.483577013015747
    },
    {
      "auxiliary_loss_clip": 0.01464299,
      "auxiliary_loss_mlp": 0.01043424,
      "balance_loss_clip": 1.05487633,
      "balance_loss_mlp": 1.03425062,
      "epoch": 0.18000360728671919,
      "flos": 21945047143680.0,
      "grad_norm": 2.377599223397255,
      "language_loss": 0.79074788,
      "learning_rate": 3.768707209272266e-06,
      "loss": 0.8158251,
      "num_input_tokens_seen": 31652575,
      "step": 1497,
      "time_per_iteration": 2.7813940048217773
    },
    {
      "auxiliary_loss_clip": 0.01325336,
      "auxiliary_loss_mlp": 0.01033055,
      "balance_loss_clip": 1.05929589,
      "balance_loss_mlp": 1.02384579,
      "epoch": 0.18012385017735827,
      "flos": 18986330937600.0,
      "grad_norm": 2.8792183391443995,
      "language_loss": 0.76986337,
      "learning_rate": 3.768343437425579e-06,
      "loss": 0.79344738,
      "num_input_tokens_seen": 31671145,
      "step": 1498,
      "time_per_iteration": 2.6971933841705322
    },
    {
      "auxiliary_loss_clip": 0.0152518,
      "auxiliary_loss_mlp": 0.01041441,
      "balance_loss_clip": 1.05379033,
      "balance_loss_mlp": 1.0318445,
      "epoch": 0.18024409306799735,
      "flos": 19748103598080.0,
      "grad_norm": 3.4281237012024977,
      "language_loss": 0.86073148,
      "learning_rate": 3.7679793973225987e-06,
      "loss": 0.88639772,
      "num_input_tokens_seen": 31686955,
      "step": 1499,
      "time_per_iteration": 2.8400216102600098
    },
    {
      "auxiliary_loss_clip": 0.01330292,
      "auxiliary_loss_mlp": 0.0100339,
      "balance_loss_clip": 1.02846074,
      "balance_loss_mlp": 1.00150609,
      "epoch": 0.18036433595863643,
      "flos": 67227183060480.0,
      "grad_norm": 0.8535107532193602,
      "language_loss": 0.6158998,
      "learning_rate": 3.767615089018549e-06,
      "loss": 0.63923669,
      "num_input_tokens_seen": 31749300,
      "step": 1500,
      "time_per_iteration": 3.4407029151916504
    },
    {
      "auxiliary_loss_clip": 0.01321488,
      "auxiliary_loss_mlp": 0.01032341,
      "balance_loss_clip": 1.06123066,
      "balance_loss_mlp": 1.02304316,
      "epoch": 0.18048457884927555,
      "flos": 18181464935040.0,
      "grad_norm": 2.1523205918471113,
      "language_loss": 0.86611426,
      "learning_rate": 3.7672505125686966e-06,
      "loss": 0.88965255,
      "num_input_tokens_seen": 31765665,
      "step": 1501,
      "time_per_iteration": 4.705912828445435
    },
    {
      "auxiliary_loss_clip": 0.01422103,
      "auxiliary_loss_mlp": 0.01034116,
      "balance_loss_clip": 1.05455267,
      "balance_loss_mlp": 1.02449608,
      "epoch": 0.18060482173991463,
      "flos": 15813767111040.0,
      "grad_norm": 3.3889831094584224,
      "language_loss": 0.84828508,
      "learning_rate": 3.7668856680283455e-06,
      "loss": 0.87284732,
      "num_input_tokens_seen": 31782690,
      "step": 1502,
      "time_per_iteration": 2.8574633598327637
    },
    {
      "auxiliary_loss_clip": 0.01334648,
      "auxiliary_loss_mlp": 0.01031809,
      "balance_loss_clip": 1.06349707,
      "balance_loss_mlp": 1.02294552,
      "epoch": 0.1807250646305537,
      "flos": 18587399512320.0,
      "grad_norm": 1.894444001222212,
      "language_loss": 0.82782638,
      "learning_rate": 3.7665205554528437e-06,
      "loss": 0.85149097,
      "num_input_tokens_seen": 31802045,
      "step": 1503,
      "time_per_iteration": 3.7274699211120605
    },
    {
      "auxiliary_loss_clip": 0.01323054,
      "auxiliary_loss_mlp": 0.01035062,
      "balance_loss_clip": 1.06204021,
      "balance_loss_mlp": 1.02591252,
      "epoch": 0.18084530752119282,
      "flos": 23149131880320.0,
      "grad_norm": 1.8444919059563152,
      "language_loss": 0.74240965,
      "learning_rate": 3.7661551748975782e-06,
      "loss": 0.76599079,
      "num_input_tokens_seen": 31820220,
      "step": 1504,
      "time_per_iteration": 2.7413785457611084
    },
    {
      "auxiliary_loss_clip": 0.01230291,
      "auxiliary_loss_mlp": 0.01006947,
      "balance_loss_clip": 1.03304541,
      "balance_loss_mlp": 1.00501585,
      "epoch": 0.1809655504118319,
      "flos": 59803153568640.0,
      "grad_norm": 0.807651729145999,
      "language_loss": 0.60462391,
      "learning_rate": 3.7657895264179772e-06,
      "loss": 0.62699634,
      "num_input_tokens_seen": 31876195,
      "step": 1505,
      "time_per_iteration": 3.2627556324005127
    },
    {
      "auxiliary_loss_clip": 0.01252354,
      "auxiliary_loss_mlp": 0.01029796,
      "balance_loss_clip": 1.05801511,
      "balance_loss_mlp": 1.02038455,
      "epoch": 0.181085793302471,
      "flos": 44201941188480.0,
      "grad_norm": 1.9605460161251065,
      "language_loss": 0.74890459,
      "learning_rate": 3.765423610069509e-06,
      "loss": 0.77172601,
      "num_input_tokens_seen": 31901585,
      "step": 1506,
      "time_per_iteration": 2.9200267791748047
    },
    {
      "auxiliary_loss_clip": 0.0132971,
      "auxiliary_loss_mlp": 0.01037796,
      "balance_loss_clip": 1.06527495,
      "balance_loss_mlp": 1.02856946,
      "epoch": 0.18120603619311007,
      "flos": 34898384638080.0,
      "grad_norm": 1.8586833885302507,
      "language_loss": 0.72465575,
      "learning_rate": 3.765057425907683e-06,
      "loss": 0.74833083,
      "num_input_tokens_seen": 31923045,
      "step": 1507,
      "time_per_iteration": 2.823455572128296
    },
    {
      "auxiliary_loss_clip": 0.01281601,
      "auxiliary_loss_mlp": 0.01030099,
      "balance_loss_clip": 1.0624876,
      "balance_loss_mlp": 1.02076519,
      "epoch": 0.18132627908374918,
      "flos": 21506757390720.0,
      "grad_norm": 1.8608123048885692,
      "language_loss": 0.78719658,
      "learning_rate": 3.764690973988048e-06,
      "loss": 0.81031358,
      "num_input_tokens_seen": 31943385,
      "step": 1508,
      "time_per_iteration": 2.6938164234161377
    },
    {
      "auxiliary_loss_clip": 0.01312622,
      "auxiliary_loss_mlp": 0.01035017,
      "balance_loss_clip": 1.06147575,
      "balance_loss_mlp": 1.02615356,
      "epoch": 0.18144652197438826,
      "flos": 29057693633280.0,
      "grad_norm": 1.8445285071634894,
      "language_loss": 0.73922729,
      "learning_rate": 3.7643242543661967e-06,
      "loss": 0.76270372,
      "num_input_tokens_seen": 31966045,
      "step": 1509,
      "time_per_iteration": 2.8014402389526367
    },
    {
      "auxiliary_loss_clip": 0.01219825,
      "auxiliary_loss_mlp": 0.01009274,
      "balance_loss_clip": 1.02712035,
      "balance_loss_mlp": 1.00750983,
      "epoch": 0.18156676486502735,
      "flos": 68675064382080.0,
      "grad_norm": 0.8902578050279709,
      "language_loss": 0.60464346,
      "learning_rate": 3.7639572670977573e-06,
      "loss": 0.62693447,
      "num_input_tokens_seen": 32021540,
      "step": 1510,
      "time_per_iteration": 3.122131109237671
    },
    {
      "auxiliary_loss_clip": 0.01380819,
      "auxiliary_loss_mlp": 0.01030365,
      "balance_loss_clip": 1.0601542,
      "balance_loss_mlp": 1.02110267,
      "epoch": 0.18168700775566646,
      "flos": 26471515334400.0,
      "grad_norm": 1.7432783812717951,
      "language_loss": 0.76802933,
      "learning_rate": 3.7635900122384042e-06,
      "loss": 0.79214108,
      "num_input_tokens_seen": 32044535,
      "step": 1511,
      "time_per_iteration": 2.9694597721099854
    },
    {
      "auxiliary_loss_clip": 0.01259324,
      "auxiliary_loss_mlp": 0.01032201,
      "balance_loss_clip": 1.05944681,
      "balance_loss_mlp": 1.02214003,
      "epoch": 0.18180725064630554,
      "flos": 15005668884480.0,
      "grad_norm": 2.1993748090054592,
      "language_loss": 0.87071532,
      "learning_rate": 3.7632224898438477e-06,
      "loss": 0.89363062,
      "num_input_tokens_seen": 32061010,
      "step": 1512,
      "time_per_iteration": 2.6210787296295166
    },
    {
      "auxiliary_loss_clip": 0.01374456,
      "auxiliary_loss_mlp": 0.01033062,
      "balance_loss_clip": 1.05495,
      "balance_loss_mlp": 1.02475929,
      "epoch": 0.18192749353694462,
      "flos": 19682387665920.0,
      "grad_norm": 1.6299524774624237,
      "language_loss": 0.78973544,
      "learning_rate": 3.762854699969842e-06,
      "loss": 0.81381065,
      "num_input_tokens_seen": 32081520,
      "step": 1513,
      "time_per_iteration": 2.8520002365112305
    },
    {
      "auxiliary_loss_clip": 0.01274512,
      "auxiliary_loss_mlp": 0.01025952,
      "balance_loss_clip": 1.06360447,
      "balance_loss_mlp": 1.01757753,
      "epoch": 0.1820477364275837,
      "flos": 20702717400960.0,
      "grad_norm": 1.8337879940906343,
      "language_loss": 0.73211563,
      "learning_rate": 3.762486642672179e-06,
      "loss": 0.75512028,
      "num_input_tokens_seen": 32098460,
      "step": 1514,
      "time_per_iteration": 2.666271448135376
    },
    {
      "auxiliary_loss_clip": 0.01329466,
      "auxiliary_loss_mlp": 0.01035185,
      "balance_loss_clip": 1.06320477,
      "balance_loss_mlp": 1.02629209,
      "epoch": 0.18216797931822282,
      "flos": 17128708197120.0,
      "grad_norm": 6.837352379259871,
      "language_loss": 0.86674875,
      "learning_rate": 3.7621183180066946e-06,
      "loss": 0.89039528,
      "num_input_tokens_seen": 32116420,
      "step": 1515,
      "time_per_iteration": 2.679243803024292
    },
    {
      "auxiliary_loss_clip": 0.01327198,
      "auxiliary_loss_mlp": 0.01032149,
      "balance_loss_clip": 1.0589962,
      "balance_loss_mlp": 1.02317834,
      "epoch": 0.1822882222088619,
      "flos": 29242561956480.0,
      "grad_norm": 1.555622723963142,
      "language_loss": 0.74022561,
      "learning_rate": 3.7617497260292625e-06,
      "loss": 0.7638191,
      "num_input_tokens_seen": 32138475,
      "step": 1516,
      "time_per_iteration": 2.794055223464966
    },
    {
      "auxiliary_loss_clip": 0.01326379,
      "auxiliary_loss_mlp": 0.01039129,
      "balance_loss_clip": 1.06300831,
      "balance_loss_mlp": 1.02898479,
      "epoch": 0.18240846509950098,
      "flos": 17702739446400.0,
      "grad_norm": 3.440362490795796,
      "language_loss": 0.78786051,
      "learning_rate": 3.7613808667957967e-06,
      "loss": 0.81151557,
      "num_input_tokens_seen": 32151165,
      "step": 1517,
      "time_per_iteration": 2.639190673828125
    },
    {
      "auxiliary_loss_clip": 0.01325516,
      "auxiliary_loss_mlp": 0.01033521,
      "balance_loss_clip": 1.05935669,
      "balance_loss_mlp": 1.0238235,
      "epoch": 0.1825287079901401,
      "flos": 14790025584000.0,
      "grad_norm": 2.4320221355112346,
      "language_loss": 0.91265273,
      "learning_rate": 3.7610117403622547e-06,
      "loss": 0.93624306,
      "num_input_tokens_seen": 32167725,
      "step": 1518,
      "time_per_iteration": 2.6486666202545166
    },
    {
      "auxiliary_loss_clip": 0.01374512,
      "auxiliary_loss_mlp": 0.01034596,
      "balance_loss_clip": 1.057302,
      "balance_loss_mlp": 1.02586424,
      "epoch": 0.18264895088077918,
      "flos": 21946232292480.0,
      "grad_norm": 1.8486312174951298,
      "language_loss": 0.90422314,
      "learning_rate": 3.7606423467846313e-06,
      "loss": 0.92831421,
      "num_input_tokens_seen": 32187330,
      "step": 1519,
      "time_per_iteration": 2.764415740966797
    },
    {
      "auxiliary_loss_clip": 0.01384077,
      "auxiliary_loss_mlp": 0.01040867,
      "balance_loss_clip": 1.06224668,
      "balance_loss_mlp": 1.03089476,
      "epoch": 0.18276919377141826,
      "flos": 20886759711360.0,
      "grad_norm": 1.6012033846428653,
      "language_loss": 0.79535741,
      "learning_rate": 3.760272686118964e-06,
      "loss": 0.81960684,
      "num_input_tokens_seen": 32205550,
      "step": 1520,
      "time_per_iteration": 2.7111802101135254
    },
    {
      "auxiliary_loss_clip": 0.01325306,
      "auxiliary_loss_mlp": 0.01032745,
      "balance_loss_clip": 1.05963123,
      "balance_loss_mlp": 1.02451921,
      "epoch": 0.18288943666205737,
      "flos": 21469877101440.0,
      "grad_norm": 2.2251339963712895,
      "language_loss": 0.92423701,
      "learning_rate": 3.7599027584213297e-06,
      "loss": 0.94781762,
      "num_input_tokens_seen": 32224430,
      "step": 1521,
      "time_per_iteration": 2.6793038845062256
    },
    {
      "auxiliary_loss_clip": 0.01281577,
      "auxiliary_loss_mlp": 0.01036016,
      "balance_loss_clip": 1.06198847,
      "balance_loss_mlp": 1.02630079,
      "epoch": 0.18300967955269645,
      "flos": 21539363961600.0,
      "grad_norm": 1.9237510364129342,
      "language_loss": 0.78241706,
      "learning_rate": 3.7595325637478465e-06,
      "loss": 0.80559301,
      "num_input_tokens_seen": 32242455,
      "step": 1522,
      "time_per_iteration": 3.5520341396331787
    },
    {
      "auxiliary_loss_clip": 0.01321207,
      "auxiliary_loss_mlp": 0.01037334,
      "balance_loss_clip": 1.05958033,
      "balance_loss_mlp": 1.02759516,
      "epoch": 0.18312992244333554,
      "flos": 28876237102080.0,
      "grad_norm": 2.1851659901385907,
      "language_loss": 0.81681836,
      "learning_rate": 3.7591621021546723e-06,
      "loss": 0.84040374,
      "num_input_tokens_seen": 32264450,
      "step": 1523,
      "time_per_iteration": 2.77325701713562
    },
    {
      "auxiliary_loss_clip": 0.01273079,
      "auxiliary_loss_mlp": 0.01035937,
      "balance_loss_clip": 1.05924845,
      "balance_loss_mlp": 1.02667463,
      "epoch": 0.18325016533397462,
      "flos": 20120102801280.0,
      "grad_norm": 1.7928902046503656,
      "language_loss": 0.81236994,
      "learning_rate": 3.7587913736980062e-06,
      "loss": 0.83546007,
      "num_input_tokens_seen": 32284090,
      "step": 1524,
      "time_per_iteration": 2.6930010318756104
    },
    {
      "auxiliary_loss_clip": 0.01462155,
      "auxiliary_loss_mlp": 0.01035101,
      "balance_loss_clip": 1.05108762,
      "balance_loss_mlp": 1.02611256,
      "epoch": 0.18337040822461373,
      "flos": 23329187781120.0,
      "grad_norm": 1.8905173833146378,
      "language_loss": 0.84347904,
      "learning_rate": 3.7584203784340865e-06,
      "loss": 0.8684516,
      "num_input_tokens_seen": 32303260,
      "step": 1525,
      "time_per_iteration": 2.8417577743530273
    },
    {
      "auxiliary_loss_clip": 0.013225,
      "auxiliary_loss_mlp": 0.01034627,
      "balance_loss_clip": 1.05911112,
      "balance_loss_mlp": 1.02526927,
      "epoch": 0.1834906511152528,
      "flos": 25009555881600.0,
      "grad_norm": 1.9928666285120804,
      "language_loss": 0.85959959,
      "learning_rate": 3.7580491164191938e-06,
      "loss": 0.88317084,
      "num_input_tokens_seen": 32321570,
      "step": 1526,
      "time_per_iteration": 3.542262315750122
    },
    {
      "auxiliary_loss_clip": 0.01178038,
      "auxiliary_loss_mlp": 0.01001822,
      "balance_loss_clip": 1.0310992,
      "balance_loss_mlp": 0.99997389,
      "epoch": 0.1836108940058919,
      "flos": 67251493589760.0,
      "grad_norm": 0.7406119319739408,
      "language_loss": 0.61165136,
      "learning_rate": 3.757677587709648e-06,
      "loss": 0.63344997,
      "num_input_tokens_seen": 32384835,
      "step": 1527,
      "time_per_iteration": 4.1913769245147705
    },
    {
      "auxiliary_loss_clip": 0.01368846,
      "auxiliary_loss_mlp": 0.01036281,
      "balance_loss_clip": 1.06002259,
      "balance_loss_mlp": 1.02695334,
      "epoch": 0.183731136896531,
      "flos": 25738721971200.0,
      "grad_norm": 3.821276520090281,
      "language_loss": 0.7559768,
      "learning_rate": 3.7573057923618095e-06,
      "loss": 0.7800281,
      "num_input_tokens_seen": 32404930,
      "step": 1528,
      "time_per_iteration": 2.8863253593444824
    },
    {
      "auxiliary_loss_clip": 0.01421987,
      "auxiliary_loss_mlp": 0.01027926,
      "balance_loss_clip": 1.05286813,
      "balance_loss_mlp": 1.01890206,
      "epoch": 0.1838513797871701,
      "flos": 20449403712000.0,
      "grad_norm": 2.1144114731080488,
      "language_loss": 0.74298739,
      "learning_rate": 3.7569337304320793e-06,
      "loss": 0.76748651,
      "num_input_tokens_seen": 32424515,
      "step": 1529,
      "time_per_iteration": 2.712099075317383
    },
    {
      "auxiliary_loss_clip": 0.01219284,
      "auxiliary_loss_mlp": 0.01000522,
      "balance_loss_clip": 1.02473307,
      "balance_loss_mlp": 0.99880499,
      "epoch": 0.18397162267780917,
      "flos": 68565141786240.0,
      "grad_norm": 0.8485929580686579,
      "language_loss": 0.64457363,
      "learning_rate": 3.756561401976899e-06,
      "loss": 0.66677171,
      "num_input_tokens_seen": 32484220,
      "step": 1530,
      "time_per_iteration": 4.008121728897095
    },
    {
      "auxiliary_loss_clip": 0.01229928,
      "auxiliary_loss_mlp": 0.01029648,
      "balance_loss_clip": 1.0649941,
      "balance_loss_mlp": 1.02086854,
      "epoch": 0.18409186556844825,
      "flos": 31941104976000.0,
      "grad_norm": 2.1764686778484577,
      "language_loss": 0.82584298,
      "learning_rate": 3.7561888070527514e-06,
      "loss": 0.8484388,
      "num_input_tokens_seen": 32506260,
      "step": 1531,
      "time_per_iteration": 2.7367424964904785
    },
    {
      "auxiliary_loss_clip": 0.0136011,
      "auxiliary_loss_mlp": 0.02578488,
      "balance_loss_clip": 1.05562639,
      "balance_loss_mlp": 0.99994671,
      "epoch": 0.18421210845908736,
      "flos": 20120533764480.0,
      "grad_norm": 2.6240605053419586,
      "language_loss": 0.80218089,
      "learning_rate": 3.7558159457161577e-06,
      "loss": 0.84156692,
      "num_input_tokens_seen": 32524225,
      "step": 1532,
      "time_per_iteration": 2.6696903705596924
    },
    {
      "auxiliary_loss_clip": 0.0132728,
      "auxiliary_loss_mlp": 0.02577833,
      "balance_loss_clip": 1.06068873,
      "balance_loss_mlp": 0.99995685,
      "epoch": 0.18433235134972645,
      "flos": 23110491824640.0,
      "grad_norm": 2.1945317712228833,
      "language_loss": 0.7777375,
      "learning_rate": 3.755442818023681e-06,
      "loss": 0.81678867,
      "num_input_tokens_seen": 32543850,
      "step": 1533,
      "time_per_iteration": 2.7071237564086914
    },
    {
      "auxiliary_loss_clip": 0.01373499,
      "auxiliary_loss_mlp": 0.01028619,
      "balance_loss_clip": 1.05788624,
      "balance_loss_mlp": 1.01941013,
      "epoch": 0.18445259424036553,
      "flos": 18291351617280.0,
      "grad_norm": 2.1350919472001957,
      "language_loss": 0.76207221,
      "learning_rate": 3.7550694240319246e-06,
      "loss": 0.78609335,
      "num_input_tokens_seen": 32561725,
      "step": 1534,
      "time_per_iteration": 2.7147843837738037
    },
    {
      "auxiliary_loss_clip": 0.01279674,
      "auxiliary_loss_mlp": 0.0102897,
      "balance_loss_clip": 1.06064558,
      "balance_loss_mlp": 1.01908135,
      "epoch": 0.18457283713100464,
      "flos": 21324079797120.0,
      "grad_norm": 11.83530599556708,
      "language_loss": 0.76325041,
      "learning_rate": 3.7546957637975326e-06,
      "loss": 0.78633684,
      "num_input_tokens_seen": 32579135,
      "step": 1535,
      "time_per_iteration": 2.6583712100982666
    },
    {
      "auxiliary_loss_clip": 0.01469449,
      "auxiliary_loss_mlp": 0.01033146,
      "balance_loss_clip": 1.0501281,
      "balance_loss_mlp": 1.02443135,
      "epoch": 0.18469308002164372,
      "flos": 20375679047040.0,
      "grad_norm": 2.1637520201808003,
      "language_loss": 0.74233985,
      "learning_rate": 3.7543218373771873e-06,
      "loss": 0.76736581,
      "num_input_tokens_seen": 32598460,
      "step": 1536,
      "time_per_iteration": 2.7301926612854004
    },
    {
      "auxiliary_loss_clip": 0.01472319,
      "auxiliary_loss_mlp": 0.02578391,
      "balance_loss_clip": 1.05297709,
      "balance_loss_mlp": 0.99994677,
      "epoch": 0.1848133229122828,
      "flos": 26435892021120.0,
      "grad_norm": 1.4098276241501269,
      "language_loss": 0.78143287,
      "learning_rate": 3.753947644827615e-06,
      "loss": 0.82194,
      "num_input_tokens_seen": 32621920,
      "step": 1537,
      "time_per_iteration": 2.875382423400879
    },
    {
      "auxiliary_loss_clip": 0.0122207,
      "auxiliary_loss_mlp": 0.01005798,
      "balance_loss_clip": 1.02246785,
      "balance_loss_mlp": 1.00403333,
      "epoch": 0.1849335658029219,
      "flos": 70547447612160.0,
      "grad_norm": 0.9067280771338199,
      "language_loss": 0.57149976,
      "learning_rate": 3.753573186205579e-06,
      "loss": 0.59377843,
      "num_input_tokens_seen": 32690040,
      "step": 1538,
      "time_per_iteration": 3.3460488319396973
    },
    {
      "auxiliary_loss_clip": 0.01318767,
      "auxiliary_loss_mlp": 0.02576949,
      "balance_loss_clip": 1.05703688,
      "balance_loss_mlp": 0.99989891,
      "epoch": 0.185053808693561,
      "flos": 17384140788480.0,
      "grad_norm": 2.8222685286886913,
      "language_loss": 0.78159559,
      "learning_rate": 3.753198461567885e-06,
      "loss": 0.82055271,
      "num_input_tokens_seen": 32707285,
      "step": 1539,
      "time_per_iteration": 2.641657829284668
    },
    {
      "auxiliary_loss_clip": 0.01366638,
      "auxiliary_loss_mlp": 0.01034207,
      "balance_loss_clip": 1.05907369,
      "balance_loss_mlp": 1.02586246,
      "epoch": 0.18517405158420008,
      "flos": 28986159697920.0,
      "grad_norm": 2.042039610549057,
      "language_loss": 0.92178935,
      "learning_rate": 3.7528234709713783e-06,
      "loss": 0.9457978,
      "num_input_tokens_seen": 32730030,
      "step": 1540,
      "time_per_iteration": 2.8332266807556152
    },
    {
      "auxiliary_loss_clip": 0.01281269,
      "auxiliary_loss_mlp": 0.01039842,
      "balance_loss_clip": 1.06507349,
      "balance_loss_mlp": 1.03153944,
      "epoch": 0.18529429447483917,
      "flos": 26794962328320.0,
      "grad_norm": 2.6237903879913365,
      "language_loss": 0.8452059,
      "learning_rate": 3.7524482144729447e-06,
      "loss": 0.86841702,
      "num_input_tokens_seen": 32749485,
      "step": 1541,
      "time_per_iteration": 2.673372507095337
    },
    {
      "auxiliary_loss_clip": 0.0137359,
      "auxiliary_loss_mlp": 0.01033125,
      "balance_loss_clip": 1.05581796,
      "balance_loss_mlp": 1.02433372,
      "epoch": 0.18541453736547828,
      "flos": 13581595301760.0,
      "grad_norm": 2.851785837179984,
      "language_loss": 0.84144133,
      "learning_rate": 3.7520726921295106e-06,
      "loss": 0.8655085,
      "num_input_tokens_seen": 32766205,
      "step": 1542,
      "time_per_iteration": 2.7095768451690674
    },
    {
      "auxiliary_loss_clip": 0.01273229,
      "auxiliary_loss_mlp": 0.01032754,
      "balance_loss_clip": 1.05966067,
      "balance_loss_mlp": 1.02327073,
      "epoch": 0.18553478025611736,
      "flos": 24025424077440.0,
      "grad_norm": 2.6985039941937567,
      "language_loss": 0.72387791,
      "learning_rate": 3.751696903998042e-06,
      "loss": 0.74693769,
      "num_input_tokens_seen": 32784840,
      "step": 1543,
      "time_per_iteration": 2.661353826522827
    },
    {
      "auxiliary_loss_clip": 0.01273794,
      "auxiliary_loss_mlp": 0.0103548,
      "balance_loss_clip": 1.06329775,
      "balance_loss_mlp": 1.02657521,
      "epoch": 0.18565502314675644,
      "flos": 25885165720320.0,
      "grad_norm": 1.8304767926306873,
      "language_loss": 0.69988662,
      "learning_rate": 3.7513208501355456e-06,
      "loss": 0.72297937,
      "num_input_tokens_seen": 32805945,
      "step": 1544,
      "time_per_iteration": 2.7166810035705566
    },
    {
      "auxiliary_loss_clip": 0.01320864,
      "auxiliary_loss_mlp": 0.01029782,
      "balance_loss_clip": 1.06008613,
      "balance_loss_mlp": 1.02145529,
      "epoch": 0.18577526603739553,
      "flos": 19610063631360.0,
      "grad_norm": 1.9804650390671747,
      "language_loss": 0.83614463,
      "learning_rate": 3.750944530599069e-06,
      "loss": 0.85965109,
      "num_input_tokens_seen": 32825515,
      "step": 1545,
      "time_per_iteration": 2.6373379230499268
    },
    {
      "auxiliary_loss_clip": 0.01285243,
      "auxiliary_loss_mlp": 0.0103491,
      "balance_loss_clip": 1.06614494,
      "balance_loss_mlp": 1.02515316,
      "epoch": 0.18589550892803464,
      "flos": 18474891137280.0,
      "grad_norm": 2.793967263722739,
      "language_loss": 0.80846059,
      "learning_rate": 3.7505679454456992e-06,
      "loss": 0.83166206,
      "num_input_tokens_seen": 32842125,
      "step": 1546,
      "time_per_iteration": 2.6117494106292725
    },
    {
      "auxiliary_loss_clip": 0.01443704,
      "auxiliary_loss_mlp": 0.0103668,
      "balance_loss_clip": 1.05179381,
      "balance_loss_mlp": 1.02766824,
      "epoch": 0.18601575181867372,
      "flos": 23549966726400.0,
      "grad_norm": 1.8699654799144776,
      "language_loss": 0.70292699,
      "learning_rate": 3.750191094732564e-06,
      "loss": 0.72773081,
      "num_input_tokens_seen": 32862990,
      "step": 1547,
      "time_per_iteration": 2.9491324424743652
    },
    {
      "auxiliary_loss_clip": 0.01447481,
      "auxiliary_loss_mlp": 0.02578792,
      "balance_loss_clip": 1.05027676,
      "balance_loss_mlp": 0.9999249,
      "epoch": 0.1861359947093128,
      "flos": 26360192108160.0,
      "grad_norm": 1.7879162356467515,
      "language_loss": 0.75239527,
      "learning_rate": 3.7498139785168313e-06,
      "loss": 0.79265803,
      "num_input_tokens_seen": 32883595,
      "step": 1548,
      "time_per_iteration": 4.036811828613281
    },
    {
      "auxiliary_loss_clip": 0.01274634,
      "auxiliary_loss_mlp": 0.01034899,
      "balance_loss_clip": 1.06395555,
      "balance_loss_mlp": 1.02578533,
      "epoch": 0.1862562375999519,
      "flos": 23331198942720.0,
      "grad_norm": 2.008091668910084,
      "language_loss": 0.76872468,
      "learning_rate": 3.749436596855709e-06,
      "loss": 0.79182005,
      "num_input_tokens_seen": 32902895,
      "step": 1549,
      "time_per_iteration": 3.0681862831115723
    },
    {
      "auxiliary_loss_clip": 0.01276631,
      "auxiliary_loss_mlp": 0.01034499,
      "balance_loss_clip": 1.0627265,
      "balance_loss_mlp": 1.0253737,
      "epoch": 0.186376480490591,
      "flos": 16648222942080.0,
      "grad_norm": 2.264092314501058,
      "language_loss": 0.9042663,
      "learning_rate": 3.749058949806446e-06,
      "loss": 0.92737758,
      "num_input_tokens_seen": 32919620,
      "step": 1550,
      "time_per_iteration": 2.6655731201171875
    },
    {
      "auxiliary_loss_clip": 0.01277722,
      "auxiliary_loss_mlp": 0.01030356,
      "balance_loss_clip": 1.06226456,
      "balance_loss_mlp": 1.02103996,
      "epoch": 0.18649672338123008,
      "flos": 21468656039040.0,
      "grad_norm": 1.8719135861937704,
      "language_loss": 0.84353256,
      "learning_rate": 3.748681037426331e-06,
      "loss": 0.86661333,
      "num_input_tokens_seen": 32938830,
      "step": 1551,
      "time_per_iteration": 2.6278204917907715
    },
    {
      "auxiliary_loss_clip": 0.01228268,
      "auxiliary_loss_mlp": 0.01031413,
      "balance_loss_clip": 1.06573129,
      "balance_loss_mlp": 1.02174485,
      "epoch": 0.1866169662718692,
      "flos": 12312728386560.0,
      "grad_norm": 2.1626783701733974,
      "language_loss": 0.91755849,
      "learning_rate": 3.7483028597726936e-06,
      "loss": 0.94015533,
      "num_input_tokens_seen": 32955600,
      "step": 1552,
      "time_per_iteration": 3.497960329055786
    },
    {
      "auxiliary_loss_clip": 0.01380861,
      "auxiliary_loss_mlp": 0.01040657,
      "balance_loss_clip": 1.06593573,
      "balance_loss_mlp": 1.03062522,
      "epoch": 0.18673720916250827,
      "flos": 23581280407680.0,
      "grad_norm": 2.4835270216512053,
      "language_loss": 0.62320334,
      "learning_rate": 3.7479244169029017e-06,
      "loss": 0.6474185,
      "num_input_tokens_seen": 32975390,
      "step": 1553,
      "time_per_iteration": 3.617387533187866
    },
    {
      "auxiliary_loss_clip": 0.01277126,
      "auxiliary_loss_mlp": 0.01030487,
      "balance_loss_clip": 1.05933881,
      "balance_loss_mlp": 1.02196395,
      "epoch": 0.18685745205314735,
      "flos": 19718370115200.0,
      "grad_norm": 2.5170237714154116,
      "language_loss": 0.73421705,
      "learning_rate": 3.7475457088743658e-06,
      "loss": 0.75729316,
      "num_input_tokens_seen": 32992640,
      "step": 1554,
      "time_per_iteration": 2.6822307109832764
    },
    {
      "auxiliary_loss_clip": 0.01325739,
      "auxiliary_loss_mlp": 0.01034925,
      "balance_loss_clip": 1.06418347,
      "balance_loss_mlp": 1.02548397,
      "epoch": 0.18697769494378644,
      "flos": 34204123589760.0,
      "grad_norm": 1.8731931201011762,
      "language_loss": 0.74698305,
      "learning_rate": 3.7471667357445348e-06,
      "loss": 0.77058971,
      "num_input_tokens_seen": 33012470,
      "step": 1555,
      "time_per_iteration": 2.8088340759277344
    },
    {
      "auxiliary_loss_clip": 0.01483212,
      "auxiliary_loss_mlp": 0.01036432,
      "balance_loss_clip": 1.06145239,
      "balance_loss_mlp": 1.02636504,
      "epoch": 0.18709793783442555,
      "flos": 34241327101440.0,
      "grad_norm": 1.7741834837185226,
      "language_loss": 0.72532117,
      "learning_rate": 3.7467874975709e-06,
      "loss": 0.75051767,
      "num_input_tokens_seen": 33033275,
      "step": 1556,
      "time_per_iteration": 4.111306428909302
    },
    {
      "auxiliary_loss_clip": 0.01289295,
      "auxiliary_loss_mlp": 0.01034544,
      "balance_loss_clip": 1.07083786,
      "balance_loss_mlp": 1.02473295,
      "epoch": 0.18721818072506463,
      "flos": 40734550529280.0,
      "grad_norm": 4.185189063590591,
      "language_loss": 0.78387523,
      "learning_rate": 3.7464079944109904e-06,
      "loss": 0.80711365,
      "num_input_tokens_seen": 33055135,
      "step": 1557,
      "time_per_iteration": 2.927212953567505
    },
    {
      "auxiliary_loss_clip": 0.01311181,
      "auxiliary_loss_mlp": 0.01039043,
      "balance_loss_clip": 1.06198168,
      "balance_loss_mlp": 1.02992952,
      "epoch": 0.18733842361570371,
      "flos": 22157386392960.0,
      "grad_norm": 1.929568055163094,
      "language_loss": 0.77803081,
      "learning_rate": 3.746028226322376e-06,
      "loss": 0.80153304,
      "num_input_tokens_seen": 33071015,
      "step": 1558,
      "time_per_iteration": 2.7257416248321533
    },
    {
      "auxiliary_loss_clip": 0.01326512,
      "auxiliary_loss_mlp": 0.01038565,
      "balance_loss_clip": 1.06436563,
      "balance_loss_mlp": 1.0296067,
      "epoch": 0.18745866650634282,
      "flos": 18914940656640.0,
      "grad_norm": 1.9280740904146816,
      "language_loss": 0.75701046,
      "learning_rate": 3.745648193362669e-06,
      "loss": 0.78066123,
      "num_input_tokens_seen": 33090370,
      "step": 1559,
      "time_per_iteration": 2.6474809646606445
    },
    {
      "auxiliary_loss_clip": 0.01323091,
      "auxiliary_loss_mlp": 0.01036826,
      "balance_loss_clip": 1.06116152,
      "balance_loss_mlp": 1.02811801,
      "epoch": 0.1875789093969819,
      "flos": 19314626267520.0,
      "grad_norm": 2.6860096544240983,
      "language_loss": 0.72458071,
      "learning_rate": 3.745267895589518e-06,
      "loss": 0.74817991,
      "num_input_tokens_seen": 33108910,
      "step": 1560,
      "time_per_iteration": 2.7224202156066895
    },
    {
      "auxiliary_loss_clip": 0.01331074,
      "auxiliary_loss_mlp": 0.01034607,
      "balance_loss_clip": 1.06516814,
      "balance_loss_mlp": 1.02506447,
      "epoch": 0.187699152287621,
      "flos": 17018965169280.0,
      "grad_norm": 2.6479964561143676,
      "language_loss": 0.81782252,
      "learning_rate": 3.7448873330606154e-06,
      "loss": 0.8414793,
      "num_input_tokens_seen": 33126680,
      "step": 1561,
      "time_per_iteration": 2.6960654258728027
    },
    {
      "auxiliary_loss_clip": 0.01372802,
      "auxiliary_loss_mlp": 0.01033651,
      "balance_loss_clip": 1.05973029,
      "balance_loss_mlp": 1.02432847,
      "epoch": 0.18781939517826007,
      "flos": 22346384780160.0,
      "grad_norm": 2.028076361895609,
      "language_loss": 0.87238455,
      "learning_rate": 3.7445065058336914e-06,
      "loss": 0.89644909,
      "num_input_tokens_seen": 33145550,
      "step": 1562,
      "time_per_iteration": 2.7259116172790527
    },
    {
      "auxiliary_loss_clip": 0.01422023,
      "auxiliary_loss_mlp": 0.01033386,
      "balance_loss_clip": 1.0561223,
      "balance_loss_mlp": 1.02426612,
      "epoch": 0.18793963806889918,
      "flos": 14611478054400.0,
      "grad_norm": 1.8881918190024216,
      "language_loss": 0.86516607,
      "learning_rate": 3.7441254139665176e-06,
      "loss": 0.8897202,
      "num_input_tokens_seen": 33161735,
      "step": 1563,
      "time_per_iteration": 2.730700731277466
    },
    {
      "auxiliary_loss_clip": 0.01232062,
      "auxiliary_loss_mlp": 0.01033435,
      "balance_loss_clip": 1.06983304,
      "balance_loss_mlp": 1.02498877,
      "epoch": 0.18805988095953827,
      "flos": 17457075354240.0,
      "grad_norm": 2.7528215602557604,
      "language_loss": 0.82633573,
      "learning_rate": 3.743744057516905e-06,
      "loss": 0.84899068,
      "num_input_tokens_seen": 33179795,
      "step": 1564,
      "time_per_iteration": 2.6147055625915527
    },
    {
      "auxiliary_loss_clip": 0.01427303,
      "auxiliary_loss_mlp": 0.01037238,
      "balance_loss_clip": 1.05951345,
      "balance_loss_mlp": 1.02748036,
      "epoch": 0.18818012385017735,
      "flos": 15043877976960.0,
      "grad_norm": 4.024404710843535,
      "language_loss": 0.87740254,
      "learning_rate": 3.743362436542706e-06,
      "loss": 0.90204787,
      "num_input_tokens_seen": 33194485,
      "step": 1565,
      "time_per_iteration": 2.6966302394866943
    },
    {
      "auxiliary_loss_clip": 0.01229658,
      "auxiliary_loss_mlp": 0.01026655,
      "balance_loss_clip": 1.06744361,
      "balance_loss_mlp": 1.01736259,
      "epoch": 0.18830036674081646,
      "flos": 47551975136640.0,
      "grad_norm": 2.296866298136529,
      "language_loss": 0.76854807,
      "learning_rate": 3.7429805511018115e-06,
      "loss": 0.79111123,
      "num_input_tokens_seen": 33216145,
      "step": 1566,
      "time_per_iteration": 2.8523616790771484
    },
    {
      "auxiliary_loss_clip": 0.01373912,
      "auxiliary_loss_mlp": 0.02580092,
      "balance_loss_clip": 1.0603056,
      "balance_loss_mlp": 0.99986959,
      "epoch": 0.18842060963145554,
      "flos": 30044626698240.0,
      "grad_norm": 2.2875972480383595,
      "language_loss": 0.7796433,
      "learning_rate": 3.7425984012521524e-06,
      "loss": 0.81918329,
      "num_input_tokens_seen": 33236345,
      "step": 1567,
      "time_per_iteration": 2.77474308013916
    },
    {
      "auxiliary_loss_clip": 0.01276623,
      "auxiliary_loss_mlp": 0.02529073,
      "balance_loss_clip": 1.03056574,
      "balance_loss_mlp": 0.99972039,
      "epoch": 0.18854085252209463,
      "flos": 70318372625280.0,
      "grad_norm": 0.7634359496930964,
      "language_loss": 0.60425007,
      "learning_rate": 3.7422159870517025e-06,
      "loss": 0.64230704,
      "num_input_tokens_seen": 33301600,
      "step": 1568,
      "time_per_iteration": 3.4091663360595703
    },
    {
      "auxiliary_loss_clip": 0.0132778,
      "auxiliary_loss_mlp": 0.01030303,
      "balance_loss_clip": 1.06297874,
      "balance_loss_mlp": 1.02161908,
      "epoch": 0.1886610954127337,
      "flos": 21289318410240.0,
      "grad_norm": 1.5446992642519313,
      "language_loss": 0.78944993,
      "learning_rate": 3.7418333085584717e-06,
      "loss": 0.81303084,
      "num_input_tokens_seen": 33322785,
      "step": 1569,
      "time_per_iteration": 2.845991611480713
    },
    {
      "auxiliary_loss_clip": 0.01374423,
      "auxiliary_loss_mlp": 0.0103068,
      "balance_loss_clip": 1.06226826,
      "balance_loss_mlp": 1.02194154,
      "epoch": 0.18878133830337282,
      "flos": 17266819991040.0,
      "grad_norm": 2.7182896402171663,
      "language_loss": 0.90426016,
      "learning_rate": 3.7414503658305128e-06,
      "loss": 0.92831123,
      "num_input_tokens_seen": 33340020,
      "step": 1570,
      "time_per_iteration": 2.695324182510376
    },
    {
      "auxiliary_loss_clip": 0.01429116,
      "auxiliary_loss_mlp": 0.01029436,
      "balance_loss_clip": 1.05713606,
      "balance_loss_mlp": 1.02033472,
      "epoch": 0.1889015811940119,
      "flos": 25775207210880.0,
      "grad_norm": 2.4301028557683844,
      "language_loss": 0.77743316,
      "learning_rate": 3.7410671589259185e-06,
      "loss": 0.80201864,
      "num_input_tokens_seen": 33358620,
      "step": 1571,
      "time_per_iteration": 2.7833962440490723
    },
    {
      "auxiliary_loss_clip": 0.01231503,
      "auxiliary_loss_mlp": 0.01033017,
      "balance_loss_clip": 1.06830454,
      "balance_loss_mlp": 1.02376604,
      "epoch": 0.18902182408465099,
      "flos": 21032197879680.0,
      "grad_norm": 2.3977315170668527,
      "language_loss": 0.79730392,
      "learning_rate": 3.7406836879028205e-06,
      "loss": 0.81994915,
      "num_input_tokens_seen": 33378845,
      "step": 1572,
      "time_per_iteration": 2.5999042987823486
    },
    {
      "auxiliary_loss_clip": 0.01272948,
      "auxiliary_loss_mlp": 0.01033986,
      "balance_loss_clip": 1.0637815,
      "balance_loss_mlp": 1.02555788,
      "epoch": 0.1891420669752901,
      "flos": 22272121411200.0,
      "grad_norm": 6.638150757592697,
      "language_loss": 0.7694236,
      "learning_rate": 3.7402999528193907e-06,
      "loss": 0.79249287,
      "num_input_tokens_seen": 33398345,
      "step": 1573,
      "time_per_iteration": 2.694312334060669
    },
    {
      "auxiliary_loss_clip": 0.01365883,
      "auxiliary_loss_mlp": 0.02577262,
      "balance_loss_clip": 1.06030464,
      "balance_loss_mlp": 0.99991941,
      "epoch": 0.18926230986592918,
      "flos": 22017802141440.0,
      "grad_norm": 2.4635209118923482,
      "language_loss": 0.85688537,
      "learning_rate": 3.739915953733842e-06,
      "loss": 0.89631683,
      "num_input_tokens_seen": 33416390,
      "step": 1574,
      "time_per_iteration": 3.7370975017547607
    },
    {
      "auxiliary_loss_clip": 0.01225516,
      "auxiliary_loss_mlp": 0.01035934,
      "balance_loss_clip": 1.06477952,
      "balance_loss_mlp": 1.02666593,
      "epoch": 0.18938255275656826,
      "flos": 24462672336000.0,
      "grad_norm": 1.7157930581080236,
      "language_loss": 0.82129699,
      "learning_rate": 3.7395316907044264e-06,
      "loss": 0.84391147,
      "num_input_tokens_seen": 33437175,
      "step": 1575,
      "time_per_iteration": 2.68109393119812
    },
    {
      "auxiliary_loss_clip": 0.01275929,
      "auxiliary_loss_mlp": 0.01034341,
      "balance_loss_clip": 1.06317997,
      "balance_loss_mlp": 1.02525711,
      "epoch": 0.18950279564720737,
      "flos": 24427049022720.0,
      "grad_norm": 1.8989705432895705,
      "language_loss": 0.80085397,
      "learning_rate": 3.7391471637894364e-06,
      "loss": 0.82395667,
      "num_input_tokens_seen": 33459440,
      "step": 1576,
      "time_per_iteration": 2.725973129272461
    },
    {
      "auxiliary_loss_clip": 0.01375919,
      "auxiliary_loss_mlp": 0.01031865,
      "balance_loss_clip": 1.05624294,
      "balance_loss_mlp": 1.02338934,
      "epoch": 0.18962303853784646,
      "flos": 19756291898880.0,
      "grad_norm": 4.6811578748429765,
      "language_loss": 0.85144961,
      "learning_rate": 3.738762373047205e-06,
      "loss": 0.87552738,
      "num_input_tokens_seen": 33479360,
      "step": 1577,
      "time_per_iteration": 2.6754956245422363
    },
    {
      "auxiliary_loss_clip": 0.01374837,
      "auxiliary_loss_mlp": 0.01030239,
      "balance_loss_clip": 1.05881,
      "balance_loss_mlp": 1.02120245,
      "epoch": 0.18974328142848554,
      "flos": 21032054225280.0,
      "grad_norm": 2.0678455594463143,
      "language_loss": 0.83273494,
      "learning_rate": 3.738377318536103e-06,
      "loss": 0.85678566,
      "num_input_tokens_seen": 33499245,
      "step": 1578,
      "time_per_iteration": 3.641477108001709
    },
    {
      "auxiliary_loss_clip": 0.01220775,
      "auxiliary_loss_mlp": 0.01028772,
      "balance_loss_clip": 1.0635829,
      "balance_loss_mlp": 1.02039766,
      "epoch": 0.18986352431912462,
      "flos": 12966122736000.0,
      "grad_norm": 2.17665743176533,
      "language_loss": 0.70919532,
      "learning_rate": 3.7379920003145447e-06,
      "loss": 0.73169082,
      "num_input_tokens_seen": 33513520,
      "step": 1579,
      "time_per_iteration": 3.533215045928955
    },
    {
      "auxiliary_loss_clip": 0.01322061,
      "auxiliary_loss_mlp": 0.01040614,
      "balance_loss_clip": 1.06325781,
      "balance_loss_mlp": 1.03133333,
      "epoch": 0.18998376720976373,
      "flos": 23767908497280.0,
      "grad_norm": 1.7974393843664607,
      "language_loss": 0.83723629,
      "learning_rate": 3.7376064184409817e-06,
      "loss": 0.86086309,
      "num_input_tokens_seen": 33533100,
      "step": 1580,
      "time_per_iteration": 2.6971888542175293
    },
    {
      "auxiliary_loss_clip": 0.01322602,
      "auxiliary_loss_mlp": 0.01028458,
      "balance_loss_clip": 1.06209135,
      "balance_loss_mlp": 1.01984477,
      "epoch": 0.19010401010040281,
      "flos": 22966023323520.0,
      "grad_norm": 1.4550947657259516,
      "language_loss": 0.86998487,
      "learning_rate": 3.7372205729739063e-06,
      "loss": 0.89349544,
      "num_input_tokens_seen": 33554915,
      "step": 1581,
      "time_per_iteration": 2.718827724456787
    },
    {
      "auxiliary_loss_clip": 0.01277619,
      "auxiliary_loss_mlp": 0.01038229,
      "balance_loss_clip": 1.06349051,
      "balance_loss_mlp": 1.02885306,
      "epoch": 0.1902242529910419,
      "flos": 19135647774720.0,
      "grad_norm": 1.916926591146347,
      "language_loss": 0.71600741,
      "learning_rate": 3.7368344639718514e-06,
      "loss": 0.7391659,
      "num_input_tokens_seen": 33572850,
      "step": 1582,
      "time_per_iteration": 3.58052396774292
    },
    {
      "auxiliary_loss_clip": 0.01275669,
      "auxiliary_loss_mlp": 0.01031884,
      "balance_loss_clip": 1.06131983,
      "balance_loss_mlp": 1.02306294,
      "epoch": 0.190344495881681,
      "flos": 25483935824640.0,
      "grad_norm": 1.7540758258451201,
      "language_loss": 0.80553615,
      "learning_rate": 3.7364480914933895e-06,
      "loss": 0.82861173,
      "num_input_tokens_seen": 33593090,
      "step": 1583,
      "time_per_iteration": 2.7210943698883057
    },
    {
      "auxiliary_loss_clip": 0.01417488,
      "auxiliary_loss_mlp": 0.02576758,
      "balance_loss_clip": 1.05579543,
      "balance_loss_mlp": 0.99984217,
      "epoch": 0.1904647387723201,
      "flos": 26792843425920.0,
      "grad_norm": 2.0976335900764274,
      "language_loss": 0.80683792,
      "learning_rate": 3.7360614555971325e-06,
      "loss": 0.84678042,
      "num_input_tokens_seen": 33612745,
      "step": 1584,
      "time_per_iteration": 2.79917049407959
    },
    {
      "auxiliary_loss_clip": 0.01274418,
      "auxiliary_loss_mlp": 0.02574393,
      "balance_loss_clip": 1.06484199,
      "balance_loss_mlp": 0.99984276,
      "epoch": 0.19058498166295917,
      "flos": 23987753688960.0,
      "grad_norm": 2.069059938470754,
      "language_loss": 0.85134065,
      "learning_rate": 3.735674556341733e-06,
      "loss": 0.88982874,
      "num_input_tokens_seen": 33632360,
      "step": 1585,
      "time_per_iteration": 2.798358917236328
    },
    {
      "auxiliary_loss_clip": 0.01320847,
      "auxiliary_loss_mlp": 0.01037913,
      "balance_loss_clip": 1.06157863,
      "balance_loss_mlp": 1.02907956,
      "epoch": 0.19070522455359826,
      "flos": 28293299280000.0,
      "grad_norm": 2.0077150732941993,
      "language_loss": 0.82935834,
      "learning_rate": 3.7352873937858835e-06,
      "loss": 0.85294592,
      "num_input_tokens_seen": 33653895,
      "step": 1586,
      "time_per_iteration": 2.7389304637908936
    },
    {
      "auxiliary_loss_clip": 0.01370965,
      "auxiliary_loss_mlp": 0.02577213,
      "balance_loss_clip": 1.06114793,
      "balance_loss_mlp": 0.9998225,
      "epoch": 0.19082546744423737,
      "flos": 25660220797440.0,
      "grad_norm": 2.0879217892397812,
      "language_loss": 0.71900427,
      "learning_rate": 3.734899967988316e-06,
      "loss": 0.75848603,
      "num_input_tokens_seen": 33672075,
      "step": 1587,
      "time_per_iteration": 2.7403852939605713
    },
    {
      "auxiliary_loss_clip": 0.01366832,
      "auxiliary_loss_mlp": 0.01035594,
      "balance_loss_clip": 1.0576272,
      "balance_loss_mlp": 1.02707028,
      "epoch": 0.19094571033487645,
      "flos": 19719483436800.0,
      "grad_norm": 2.0060733557821995,
      "language_loss": 0.84025788,
      "learning_rate": 3.7345122790078026e-06,
      "loss": 0.86428213,
      "num_input_tokens_seen": 33689640,
      "step": 1588,
      "time_per_iteration": 2.734964609146118
    },
    {
      "auxiliary_loss_clip": 0.01273735,
      "auxiliary_loss_mlp": 0.01040315,
      "balance_loss_clip": 1.06355274,
      "balance_loss_mlp": 1.03087986,
      "epoch": 0.19106595322551553,
      "flos": 21616320850560.0,
      "grad_norm": 4.807057013860972,
      "language_loss": 0.92631257,
      "learning_rate": 3.7341243269031556e-06,
      "loss": 0.94945312,
      "num_input_tokens_seen": 33708630,
      "step": 1589,
      "time_per_iteration": 2.6260786056518555
    },
    {
      "auxiliary_loss_clip": 0.0131939,
      "auxiliary_loss_mlp": 0.01033097,
      "balance_loss_clip": 1.06278276,
      "balance_loss_mlp": 1.02510405,
      "epoch": 0.19118619611615464,
      "flos": 29896890059520.0,
      "grad_norm": 1.5946069264220581,
      "language_loss": 0.77520257,
      "learning_rate": 3.7337361117332275e-06,
      "loss": 0.79872739,
      "num_input_tokens_seen": 33730370,
      "step": 1590,
      "time_per_iteration": 2.7638638019561768
    },
    {
      "auxiliary_loss_clip": 0.01379286,
      "auxiliary_loss_mlp": 0.01035832,
      "balance_loss_clip": 1.06074703,
      "balance_loss_mlp": 1.0273149,
      "epoch": 0.19130643900679373,
      "flos": 17273428093440.0,
      "grad_norm": 5.559809598946598,
      "language_loss": 0.77221191,
      "learning_rate": 3.7333476335569087e-06,
      "loss": 0.79636306,
      "num_input_tokens_seen": 33748370,
      "step": 1591,
      "time_per_iteration": 2.686599016189575
    },
    {
      "auxiliary_loss_clip": 0.01323911,
      "auxiliary_loss_mlp": 0.01028741,
      "balance_loss_clip": 1.06160259,
      "balance_loss_mlp": 1.01994348,
      "epoch": 0.1914266818974328,
      "flos": 24826339584000.0,
      "grad_norm": 6.269812036376255,
      "language_loss": 0.67183095,
      "learning_rate": 3.7329588924331325e-06,
      "loss": 0.69535744,
      "num_input_tokens_seen": 33769575,
      "step": 1592,
      "time_per_iteration": 2.6736905574798584
    },
    {
      "auxiliary_loss_clip": 0.01368752,
      "auxiliary_loss_mlp": 0.010328,
      "balance_loss_clip": 1.0564301,
      "balance_loss_mlp": 1.02357888,
      "epoch": 0.1915469247880719,
      "flos": 18952467390720.0,
      "grad_norm": 2.704053204865874,
      "language_loss": 0.82586849,
      "learning_rate": 3.732569888420871e-06,
      "loss": 0.84988403,
      "num_input_tokens_seen": 33789110,
      "step": 1593,
      "time_per_iteration": 2.698427200317383
    },
    {
      "auxiliary_loss_clip": 0.01228268,
      "auxiliary_loss_mlp": 0.01033261,
      "balance_loss_clip": 1.0644908,
      "balance_loss_mlp": 1.02399242,
      "epoch": 0.191667167678711,
      "flos": 21032952065280.0,
      "grad_norm": 4.257591176240382,
      "language_loss": 0.82573867,
      "learning_rate": 3.732180621579134e-06,
      "loss": 0.84835398,
      "num_input_tokens_seen": 33808325,
      "step": 1594,
      "time_per_iteration": 2.624579668045044
    },
    {
      "auxiliary_loss_clip": 0.01390604,
      "auxiliary_loss_mlp": 0.01042865,
      "balance_loss_clip": 1.06684625,
      "balance_loss_mlp": 1.03349495,
      "epoch": 0.1917874105693501,
      "flos": 34237663914240.0,
      "grad_norm": 2.2733804876581423,
      "language_loss": 0.81549966,
      "learning_rate": 3.7317910919669745e-06,
      "loss": 0.83983439,
      "num_input_tokens_seen": 33829520,
      "step": 1595,
      "time_per_iteration": 2.830038547515869
    },
    {
      "auxiliary_loss_clip": 0.0127079,
      "auxiliary_loss_mlp": 0.01035446,
      "balance_loss_clip": 1.06108594,
      "balance_loss_mlp": 1.02562916,
      "epoch": 0.19190765345998917,
      "flos": 23550613171200.0,
      "grad_norm": 2.3499102044321534,
      "language_loss": 0.76747012,
      "learning_rate": 3.7314012996434826e-06,
      "loss": 0.79053247,
      "num_input_tokens_seen": 33848250,
      "step": 1596,
      "time_per_iteration": 2.6713778972625732
    },
    {
      "auxiliary_loss_clip": 0.01245501,
      "auxiliary_loss_mlp": 0.01035543,
      "balance_loss_clip": 1.05698276,
      "balance_loss_mlp": 1.02710867,
      "epoch": 0.19202789635062828,
      "flos": 19861330245120.0,
      "grad_norm": 2.0370847736743967,
      "language_loss": 0.81233031,
      "learning_rate": 3.7310112446677907e-06,
      "loss": 0.83514071,
      "num_input_tokens_seen": 33866160,
      "step": 1597,
      "time_per_iteration": 2.6779978275299072
    },
    {
      "auxiliary_loss_clip": 0.01230201,
      "auxiliary_loss_mlp": 0.01034018,
      "balance_loss_clip": 1.06773126,
      "balance_loss_mlp": 1.02460647,
      "epoch": 0.19214813924126736,
      "flos": 20922957642240.0,
      "grad_norm": 2.199490494354691,
      "language_loss": 0.69151396,
      "learning_rate": 3.7306209270990695e-06,
      "loss": 0.71415615,
      "num_input_tokens_seen": 33884165,
      "step": 1598,
      "time_per_iteration": 2.574169874191284
    },
    {
      "auxiliary_loss_clip": 0.0132718,
      "auxiliary_loss_mlp": 0.01038498,
      "balance_loss_clip": 1.06380296,
      "balance_loss_mlp": 1.02983153,
      "epoch": 0.19226838213190645,
      "flos": 26359725231360.0,
      "grad_norm": 1.794130281661613,
      "language_loss": 0.8664614,
      "learning_rate": 3.7302303469965292e-06,
      "loss": 0.89011812,
      "num_input_tokens_seen": 33903705,
      "step": 1599,
      "time_per_iteration": 2.7408499717712402
    },
    {
      "auxiliary_loss_clip": 0.01278515,
      "auxiliary_loss_mlp": 0.01037513,
      "balance_loss_clip": 1.06643903,
      "balance_loss_mlp": 1.02807152,
      "epoch": 0.19238862502254553,
      "flos": 20850525866880.0,
      "grad_norm": 1.9654174354860336,
      "language_loss": 0.7055974,
      "learning_rate": 3.7298395044194206e-06,
      "loss": 0.72875774,
      "num_input_tokens_seen": 33922515,
      "step": 1600,
      "time_per_iteration": 3.581028461456299
    },
    {
      "auxiliary_loss_clip": 0.01229848,
      "auxiliary_loss_mlp": 0.01035222,
      "balance_loss_clip": 1.06911755,
      "balance_loss_mlp": 1.0268594,
      "epoch": 0.19250886791318464,
      "flos": 21726063878400.0,
      "grad_norm": 1.9923742829773876,
      "language_loss": 0.94494545,
      "learning_rate": 3.7294483994270356e-06,
      "loss": 0.96759617,
      "num_input_tokens_seen": 33940840,
      "step": 1601,
      "time_per_iteration": 2.599031448364258
    },
    {
      "auxiliary_loss_clip": 0.01416569,
      "auxiliary_loss_mlp": 0.01036626,
      "balance_loss_clip": 1.05908775,
      "balance_loss_mlp": 1.02723217,
      "epoch": 0.19262911080382372,
      "flos": 23367827836800.0,
      "grad_norm": 2.0765211928812137,
      "language_loss": 0.78458393,
      "learning_rate": 3.7290570320787033e-06,
      "loss": 0.80911589,
      "num_input_tokens_seen": 33960420,
      "step": 1602,
      "time_per_iteration": 2.7741239070892334
    },
    {
      "auxiliary_loss_clip": 0.01282915,
      "auxiliary_loss_mlp": 0.01037471,
      "balance_loss_clip": 1.06945622,
      "balance_loss_mlp": 1.02817893,
      "epoch": 0.1927493536944628,
      "flos": 21943502858880.0,
      "grad_norm": 2.249383112238618,
      "language_loss": 0.71324074,
      "learning_rate": 3.728665402433793e-06,
      "loss": 0.73644465,
      "num_input_tokens_seen": 33978990,
      "step": 1603,
      "time_per_iteration": 2.6608128547668457
    },
    {
      "auxiliary_loss_clip": 0.01330562,
      "auxiliary_loss_mlp": 0.01033054,
      "balance_loss_clip": 1.06753898,
      "balance_loss_mlp": 1.02451897,
      "epoch": 0.19286959658510192,
      "flos": 16545590807040.0,
      "grad_norm": 2.740697480506498,
      "language_loss": 0.86061269,
      "learning_rate": 3.7282735105517164e-06,
      "loss": 0.88424885,
      "num_input_tokens_seen": 33997115,
      "step": 1604,
      "time_per_iteration": 3.606609344482422
    },
    {
      "auxiliary_loss_clip": 0.01430999,
      "auxiliary_loss_mlp": 0.01035888,
      "balance_loss_clip": 1.05964351,
      "balance_loss_mlp": 1.02656007,
      "epoch": 0.192989839475741,
      "flos": 21616967295360.0,
      "grad_norm": 2.136332928491254,
      "language_loss": 0.67508149,
      "learning_rate": 3.727881356491922e-06,
      "loss": 0.6997503,
      "num_input_tokens_seen": 34015525,
      "step": 1605,
      "time_per_iteration": 3.616151809692383
    },
    {
      "auxiliary_loss_clip": 0.01230506,
      "auxiliary_loss_mlp": 0.01035701,
      "balance_loss_clip": 1.07175064,
      "balance_loss_mlp": 1.02735639,
      "epoch": 0.19311008236638008,
      "flos": 19281516906240.0,
      "grad_norm": 1.984762290856617,
      "language_loss": 0.75520182,
      "learning_rate": 3.7274889403139002e-06,
      "loss": 0.77786386,
      "num_input_tokens_seen": 34033150,
      "step": 1606,
      "time_per_iteration": 2.615187883377075
    },
    {
      "auxiliary_loss_clip": 0.01414748,
      "auxiliary_loss_mlp": 0.01031529,
      "balance_loss_clip": 1.06115532,
      "balance_loss_mlp": 1.02323806,
      "epoch": 0.1932303252570192,
      "flos": 28652369587200.0,
      "grad_norm": 2.8677020948390517,
      "language_loss": 0.78804463,
      "learning_rate": 3.727096262077179e-06,
      "loss": 0.81250739,
      "num_input_tokens_seen": 34052145,
      "step": 1607,
      "time_per_iteration": 3.7641241550445557
    },
    {
      "auxiliary_loss_clip": 0.01276723,
      "auxiliary_loss_mlp": 0.01030042,
      "balance_loss_clip": 1.06651807,
      "balance_loss_mlp": 1.0208981,
      "epoch": 0.19335056814765827,
      "flos": 18368990864640.0,
      "grad_norm": 2.010241591249997,
      "language_loss": 0.85330868,
      "learning_rate": 3.7267033218413285e-06,
      "loss": 0.87637627,
      "num_input_tokens_seen": 34069940,
      "step": 1608,
      "time_per_iteration": 2.5748157501220703
    },
    {
      "auxiliary_loss_clip": 0.01475616,
      "auxiliary_loss_mlp": 0.01034079,
      "balance_loss_clip": 1.0514549,
      "balance_loss_mlp": 1.02539456,
      "epoch": 0.19347081103829736,
      "flos": 13260877741440.0,
      "grad_norm": 2.4044617244439808,
      "language_loss": 0.80971146,
      "learning_rate": 3.726310119665957e-06,
      "loss": 0.83480841,
      "num_input_tokens_seen": 34086275,
      "step": 1609,
      "time_per_iteration": 2.7809226512908936
    },
    {
      "auxiliary_loss_clip": 0.01274572,
      "auxiliary_loss_mlp": 0.01031666,
      "balance_loss_clip": 1.06615829,
      "balance_loss_mlp": 1.02294564,
      "epoch": 0.19359105392893644,
      "flos": 20300122788480.0,
      "grad_norm": 1.766424966613472,
      "language_loss": 0.85799384,
      "learning_rate": 3.725916655610713e-06,
      "loss": 0.88105619,
      "num_input_tokens_seen": 34105605,
      "step": 1610,
      "time_per_iteration": 2.670405387878418
    },
    {
      "auxiliary_loss_clip": 0.01320499,
      "auxiliary_loss_mlp": 0.01035335,
      "balance_loss_clip": 1.06014431,
      "balance_loss_mlp": 1.02659059,
      "epoch": 0.19371129681957555,
      "flos": 20484596062080.0,
      "grad_norm": 2.890815172117339,
      "language_loss": 0.75856501,
      "learning_rate": 3.725522929735284e-06,
      "loss": 0.78212333,
      "num_input_tokens_seen": 34122540,
      "step": 1611,
      "time_per_iteration": 2.6856460571289062
    },
    {
      "auxiliary_loss_clip": 0.01331041,
      "auxiliary_loss_mlp": 0.01030726,
      "balance_loss_clip": 1.06009614,
      "balance_loss_mlp": 1.02177978,
      "epoch": 0.19383153971021463,
      "flos": 30445497457920.0,
      "grad_norm": 1.9622279938092642,
      "language_loss": 0.74600101,
      "learning_rate": 3.725128942099399e-06,
      "loss": 0.76961869,
      "num_input_tokens_seen": 34142940,
      "step": 1612,
      "time_per_iteration": 2.7510533332824707
    },
    {
      "auxiliary_loss_clip": 0.01321311,
      "auxiliary_loss_mlp": 0.01035563,
      "balance_loss_clip": 1.06112444,
      "balance_loss_mlp": 1.02689695,
      "epoch": 0.19395178260085372,
      "flos": 24569937325440.0,
      "grad_norm": 1.6575902897539643,
      "language_loss": 0.79901987,
      "learning_rate": 3.7247346927628245e-06,
      "loss": 0.82258856,
      "num_input_tokens_seen": 34162875,
      "step": 1613,
      "time_per_iteration": 2.7569332122802734
    },
    {
      "auxiliary_loss_clip": 0.01323743,
      "auxiliary_loss_mlp": 0.02578953,
      "balance_loss_clip": 1.06364202,
      "balance_loss_mlp": 0.99982178,
      "epoch": 0.19407202549149283,
      "flos": 28950608211840.0,
      "grad_norm": 7.808048207736728,
      "language_loss": 0.78962046,
      "learning_rate": 3.7243401817853694e-06,
      "loss": 0.82864743,
      "num_input_tokens_seen": 34183565,
      "step": 1614,
      "time_per_iteration": 2.7305638790130615
    },
    {
      "auxiliary_loss_clip": 0.01272251,
      "auxiliary_loss_mlp": 0.01035937,
      "balance_loss_clip": 1.06453562,
      "balance_loss_mlp": 1.02728796,
      "epoch": 0.1941922683821319,
      "flos": 18004497603840.0,
      "grad_norm": 2.708856297993703,
      "language_loss": 0.72124892,
      "learning_rate": 3.723945409226879e-06,
      "loss": 0.74433082,
      "num_input_tokens_seen": 34202055,
      "step": 1615,
      "time_per_iteration": 2.7009339332580566
    },
    {
      "auxiliary_loss_clip": 0.01278101,
      "auxiliary_loss_mlp": 0.01028321,
      "balance_loss_clip": 1.06508887,
      "balance_loss_mlp": 1.0190047,
      "epoch": 0.194312511272771,
      "flos": 9720337034880.0,
      "grad_norm": 3.2855454468495067,
      "language_loss": 0.79849112,
      "learning_rate": 3.723550375147241e-06,
      "loss": 0.82155532,
      "num_input_tokens_seen": 34216830,
      "step": 1616,
      "time_per_iteration": 2.688101053237915
    },
    {
      "auxiliary_loss_clip": 0.01365717,
      "auxiliary_loss_mlp": 0.01034614,
      "balance_loss_clip": 1.05835581,
      "balance_loss_mlp": 1.0247252,
      "epoch": 0.19443275416341008,
      "flos": 27016208150400.0,
      "grad_norm": 1.815903514794724,
      "language_loss": 0.80272162,
      "learning_rate": 3.7231550796063816e-06,
      "loss": 0.82672489,
      "num_input_tokens_seen": 34236840,
      "step": 1617,
      "time_per_iteration": 2.7260186672210693
    },
    {
      "auxiliary_loss_clip": 0.01334483,
      "auxiliary_loss_mlp": 0.01035505,
      "balance_loss_clip": 1.06733775,
      "balance_loss_mlp": 1.02643895,
      "epoch": 0.1945529970540492,
      "flos": 15846625077120.0,
      "grad_norm": 2.6935295279768154,
      "language_loss": 0.65229195,
      "learning_rate": 3.722759522664266e-06,
      "loss": 0.67599177,
      "num_input_tokens_seen": 34254140,
      "step": 1618,
      "time_per_iteration": 2.6493189334869385
    },
    {
      "auxiliary_loss_clip": 0.01427362,
      "auxiliary_loss_mlp": 0.01032593,
      "balance_loss_clip": 1.05990553,
      "balance_loss_mlp": 1.02409327,
      "epoch": 0.19467323994468827,
      "flos": 19314985403520.0,
      "grad_norm": 2.960070663531769,
      "language_loss": 0.82022285,
      "learning_rate": 3.7223637043809016e-06,
      "loss": 0.84482241,
      "num_input_tokens_seen": 34273120,
      "step": 1619,
      "time_per_iteration": 2.6997365951538086
    },
    {
      "auxiliary_loss_clip": 0.01380111,
      "auxiliary_loss_mlp": 0.01039721,
      "balance_loss_clip": 1.06456208,
      "balance_loss_mlp": 1.03097677,
      "epoch": 0.19479348283532735,
      "flos": 24133227770880.0,
      "grad_norm": 2.626150399216743,
      "language_loss": 0.86686182,
      "learning_rate": 3.7219676248163322e-06,
      "loss": 0.89106017,
      "num_input_tokens_seen": 34290285,
      "step": 1620,
      "time_per_iteration": 2.724510669708252
    },
    {
      "auxiliary_loss_clip": 0.0128569,
      "auxiliary_loss_mlp": 0.01045906,
      "balance_loss_clip": 1.06960034,
      "balance_loss_mlp": 1.03691792,
      "epoch": 0.19491372572596646,
      "flos": 25775638174080.0,
      "grad_norm": 2.076605864202758,
      "language_loss": 0.93719983,
      "learning_rate": 3.721571284030643e-06,
      "loss": 0.9605158,
      "num_input_tokens_seen": 34310095,
      "step": 1621,
      "time_per_iteration": 2.6683905124664307
    },
    {
      "auxiliary_loss_clip": 0.01279396,
      "auxiliary_loss_mlp": 0.01030326,
      "balance_loss_clip": 1.06515706,
      "balance_loss_mlp": 1.02140903,
      "epoch": 0.19503396861660555,
      "flos": 19645220067840.0,
      "grad_norm": 2.1842344338348005,
      "language_loss": 0.79052651,
      "learning_rate": 3.7211746820839587e-06,
      "loss": 0.81362373,
      "num_input_tokens_seen": 34327190,
      "step": 1622,
      "time_per_iteration": 2.7129218578338623
    },
    {
      "auxiliary_loss_clip": 0.01508868,
      "auxiliary_loss_mlp": 0.0103688,
      "balance_loss_clip": 1.05419111,
      "balance_loss_mlp": 1.02774239,
      "epoch": 0.19515421150724463,
      "flos": 21033023892480.0,
      "grad_norm": 1.7772317286182924,
      "language_loss": 0.80885893,
      "learning_rate": 3.7207778190364437e-06,
      "loss": 0.83431643,
      "num_input_tokens_seen": 34345615,
      "step": 1623,
      "time_per_iteration": 2.8700013160705566
    },
    {
      "auxiliary_loss_clip": 0.01462349,
      "auxiliary_loss_mlp": 0.01035747,
      "balance_loss_clip": 1.05476546,
      "balance_loss_mlp": 1.02653193,
      "epoch": 0.1952744543978837,
      "flos": 32961255143040.0,
      "grad_norm": 1.6583199020224215,
      "language_loss": 0.74294996,
      "learning_rate": 3.720380694948302e-06,
      "loss": 0.76793087,
      "num_input_tokens_seen": 34368500,
      "step": 1624,
      "time_per_iteration": 2.9747817516326904
    },
    {
      "auxiliary_loss_clip": 0.01354666,
      "auxiliary_loss_mlp": 0.01009032,
      "balance_loss_clip": 1.04805779,
      "balance_loss_mlp": 1.00704098,
      "epoch": 0.19539469728852282,
      "flos": 64044312030720.0,
      "grad_norm": 1.0347040211516259,
      "language_loss": 0.71264768,
      "learning_rate": 3.719983309879777e-06,
      "loss": 0.73628461,
      "num_input_tokens_seen": 34428280,
      "step": 1625,
      "time_per_iteration": 4.256620407104492
    },
    {
      "auxiliary_loss_clip": 0.0136844,
      "auxiliary_loss_mlp": 0.01032764,
      "balance_loss_clip": 1.06002927,
      "balance_loss_mlp": 1.02412772,
      "epoch": 0.1955149401791619,
      "flos": 13370908078080.0,
      "grad_norm": 1.9409172921653,
      "language_loss": 0.77587306,
      "learning_rate": 3.719585663891151e-06,
      "loss": 0.79988509,
      "num_input_tokens_seen": 34445815,
      "step": 1626,
      "time_per_iteration": 2.721174955368042
    },
    {
      "auxiliary_loss_clip": 0.01423042,
      "auxiliary_loss_mlp": 0.01041042,
      "balance_loss_clip": 1.06337428,
      "balance_loss_mlp": 1.03182101,
      "epoch": 0.195635183069801,
      "flos": 18728887184640.0,
      "grad_norm": 2.4068151477579645,
      "language_loss": 0.78923512,
      "learning_rate": 3.719187757042747e-06,
      "loss": 0.81387591,
      "num_input_tokens_seen": 34463635,
      "step": 1627,
      "time_per_iteration": 2.728510856628418
    },
    {
      "auxiliary_loss_clip": 0.01238639,
      "auxiliary_loss_mlp": 0.01007779,
      "balance_loss_clip": 1.04216862,
      "balance_loss_mlp": 1.00589585,
      "epoch": 0.1957554259604401,
      "flos": 69313952615040.0,
      "grad_norm": 0.7315007079120539,
      "language_loss": 0.54955041,
      "learning_rate": 3.7187895893949275e-06,
      "loss": 0.57201457,
      "num_input_tokens_seen": 34530105,
      "step": 1628,
      "time_per_iteration": 3.32606840133667
    },
    {
      "auxiliary_loss_clip": 0.0141532,
      "auxiliary_loss_mlp": 0.0103375,
      "balance_loss_clip": 1.05637193,
      "balance_loss_mlp": 1.02519691,
      "epoch": 0.19587566885107918,
      "flos": 21069257736960.0,
      "grad_norm": 2.995131503901059,
      "language_loss": 0.76030356,
      "learning_rate": 3.7183911610080937e-06,
      "loss": 0.78479421,
      "num_input_tokens_seen": 34546970,
      "step": 1629,
      "time_per_iteration": 3.6623995304107666
    },
    {
      "auxiliary_loss_clip": 0.01305366,
      "auxiliary_loss_mlp": 0.01035309,
      "balance_loss_clip": 1.06120396,
      "balance_loss_mlp": 1.02604651,
      "epoch": 0.19599591174171827,
      "flos": 22194661731840.0,
      "grad_norm": 4.1566697818360225,
      "language_loss": 0.75111079,
      "learning_rate": 3.7179924719426872e-06,
      "loss": 0.77451754,
      "num_input_tokens_seen": 34564865,
      "step": 1630,
      "time_per_iteration": 2.7885754108428955
    },
    {
      "auxiliary_loss_clip": 0.01278838,
      "auxiliary_loss_mlp": 0.01041541,
      "balance_loss_clip": 1.06625915,
      "balance_loss_mlp": 1.03202832,
      "epoch": 0.19611615463235738,
      "flos": 23768375374080.0,
      "grad_norm": 2.9022533482444013,
      "language_loss": 0.76047027,
      "learning_rate": 3.7175935222591885e-06,
      "loss": 0.783674,
      "num_input_tokens_seen": 34584165,
      "step": 1631,
      "time_per_iteration": 3.607504367828369
    },
    {
      "auxiliary_loss_clip": 0.013311,
      "auxiliary_loss_mlp": 0.01041831,
      "balance_loss_clip": 1.06975675,
      "balance_loss_mlp": 1.03293777,
      "epoch": 0.19623639752299646,
      "flos": 28618218731520.0,
      "grad_norm": 2.0646939994658036,
      "language_loss": 0.74516457,
      "learning_rate": 3.717194312018118e-06,
      "loss": 0.7688939,
      "num_input_tokens_seen": 34603150,
      "step": 1632,
      "time_per_iteration": 2.7685353755950928
    },
    {
      "auxiliary_loss_clip": 0.01277834,
      "auxiliary_loss_mlp": 0.01037987,
      "balance_loss_clip": 1.06596112,
      "balance_loss_mlp": 1.02944529,
      "epoch": 0.19635664041363554,
      "flos": 21032700670080.0,
      "grad_norm": 2.747004545600834,
      "language_loss": 0.76131326,
      "learning_rate": 3.716794841280036e-06,
      "loss": 0.78447139,
      "num_input_tokens_seen": 34621855,
      "step": 1633,
      "time_per_iteration": 3.582960605621338
    },
    {
      "auxiliary_loss_clip": 0.01282731,
      "auxiliary_loss_mlp": 0.01033988,
      "balance_loss_clip": 1.06494069,
      "balance_loss_mlp": 1.02458835,
      "epoch": 0.19647688330427462,
      "flos": 18879748306560.0,
      "grad_norm": 2.0224364333353804,
      "language_loss": 0.77362412,
      "learning_rate": 3.7163951101055407e-06,
      "loss": 0.79679132,
      "num_input_tokens_seen": 34639915,
      "step": 1634,
      "time_per_iteration": 2.652224063873291
    },
    {
      "auxiliary_loss_clip": 0.0132899,
      "auxiliary_loss_mlp": 0.01036812,
      "balance_loss_clip": 1.06788564,
      "balance_loss_mlp": 1.027812,
      "epoch": 0.19659712619491373,
      "flos": 24242503921920.0,
      "grad_norm": 2.713023112800421,
      "language_loss": 0.7873801,
      "learning_rate": 3.715995118555273e-06,
      "loss": 0.81103814,
      "num_input_tokens_seen": 34659890,
      "step": 1635,
      "time_per_iteration": 2.7692902088165283
    },
    {
      "auxiliary_loss_clip": 0.01355697,
      "auxiliary_loss_mlp": 0.01032775,
      "balance_loss_clip": 1.06057656,
      "balance_loss_mlp": 1.02292252,
      "epoch": 0.19671736908555282,
      "flos": 24717422568960.0,
      "grad_norm": 2.20729684441871,
      "language_loss": 0.8589884,
      "learning_rate": 3.71559486668991e-06,
      "loss": 0.88287312,
      "num_input_tokens_seen": 34678750,
      "step": 1636,
      "time_per_iteration": 2.9372501373291016
    },
    {
      "auxiliary_loss_clip": 0.01282083,
      "auxiliary_loss_mlp": 0.02578211,
      "balance_loss_clip": 1.06735468,
      "balance_loss_mlp": 0.99987519,
      "epoch": 0.1968376119761919,
      "flos": 23842279607040.0,
      "grad_norm": 1.62807709885512,
      "language_loss": 0.77539194,
      "learning_rate": 3.715194354570169e-06,
      "loss": 0.81399488,
      "num_input_tokens_seen": 34698755,
      "step": 1637,
      "time_per_iteration": 2.7777137756347656
    },
    {
      "auxiliary_loss_clip": 0.0127829,
      "auxiliary_loss_mlp": 0.01038615,
      "balance_loss_clip": 1.07143891,
      "balance_loss_mlp": 1.02998996,
      "epoch": 0.196957854866831,
      "flos": 18113917409280.0,
      "grad_norm": 1.9818268899689517,
      "language_loss": 0.83549643,
      "learning_rate": 3.714793582256809e-06,
      "loss": 0.85866547,
      "num_input_tokens_seen": 34715820,
      "step": 1638,
      "time_per_iteration": 2.720048189163208
    },
    {
      "auxiliary_loss_clip": 0.01225942,
      "auxiliary_loss_mlp": 0.01036275,
      "balance_loss_clip": 1.06679404,
      "balance_loss_mlp": 1.02684557,
      "epoch": 0.1970780977574701,
      "flos": 21653129312640.0,
      "grad_norm": 2.946475094882912,
      "language_loss": 0.84997672,
      "learning_rate": 3.7143925498106253e-06,
      "loss": 0.87259889,
      "num_input_tokens_seen": 34734360,
      "step": 1639,
      "time_per_iteration": 2.6627140045166016
    },
    {
      "auxiliary_loss_clip": 0.01325471,
      "auxiliary_loss_mlp": 0.01040324,
      "balance_loss_clip": 1.05932939,
      "balance_loss_mlp": 1.03091812,
      "epoch": 0.19719834064810918,
      "flos": 20811813984000.0,
      "grad_norm": 1.991505543512618,
      "language_loss": 0.79373944,
      "learning_rate": 3.7139912572924558e-06,
      "loss": 0.81739742,
      "num_input_tokens_seen": 34753390,
      "step": 1640,
      "time_per_iteration": 2.6914827823638916
    },
    {
      "auxiliary_loss_clip": 0.01273705,
      "auxiliary_loss_mlp": 0.01029837,
      "balance_loss_clip": 1.06144691,
      "balance_loss_mlp": 1.02092612,
      "epoch": 0.19731858353874826,
      "flos": 23434800744960.0,
      "grad_norm": 2.5315872197725526,
      "language_loss": 0.8065201,
      "learning_rate": 3.7135897047631744e-06,
      "loss": 0.82955551,
      "num_input_tokens_seen": 34771275,
      "step": 1641,
      "time_per_iteration": 2.6427090167999268
    },
    {
      "auxiliary_loss_clip": 0.01334248,
      "auxiliary_loss_mlp": 0.01030876,
      "balance_loss_clip": 1.06787193,
      "balance_loss_mlp": 1.02139902,
      "epoch": 0.19743882642938737,
      "flos": 23988184652160.0,
      "grad_norm": 2.9806369794888865,
      "language_loss": 0.76329261,
      "learning_rate": 3.713187892283698e-06,
      "loss": 0.78694391,
      "num_input_tokens_seen": 34790885,
      "step": 1642,
      "time_per_iteration": 2.693169593811035
    },
    {
      "auxiliary_loss_clip": 0.01420425,
      "auxiliary_loss_mlp": 0.01036055,
      "balance_loss_clip": 1.05612993,
      "balance_loss_mlp": 1.02741218,
      "epoch": 0.19755906932002645,
      "flos": 15004340081280.0,
      "grad_norm": 3.154642741200047,
      "language_loss": 0.87062871,
      "learning_rate": 3.71278581991498e-06,
      "loss": 0.89519358,
      "num_input_tokens_seen": 34806745,
      "step": 1643,
      "time_per_iteration": 2.7449069023132324
    },
    {
      "auxiliary_loss_clip": 0.01384026,
      "auxiliary_loss_mlp": 0.02582457,
      "balance_loss_clip": 1.07087326,
      "balance_loss_mlp": 0.99984628,
      "epoch": 0.19767931221066554,
      "flos": 19494466686720.0,
      "grad_norm": 1.7908206521950611,
      "language_loss": 0.78996474,
      "learning_rate": 3.712383487718015e-06,
      "loss": 0.82962954,
      "num_input_tokens_seen": 34824985,
      "step": 1644,
      "time_per_iteration": 2.752018928527832
    },
    {
      "auxiliary_loss_clip": 0.01390973,
      "auxiliary_loss_mlp": 0.01031379,
      "balance_loss_clip": 1.05664515,
      "balance_loss_mlp": 1.02343392,
      "epoch": 0.19779955510130465,
      "flos": 25737895958400.0,
      "grad_norm": 1.9011143634136871,
      "language_loss": 0.87350762,
      "learning_rate": 3.7119808957538365e-06,
      "loss": 0.89773113,
      "num_input_tokens_seen": 34843980,
      "step": 1645,
      "time_per_iteration": 2.755253791809082
    },
    {
      "auxiliary_loss_clip": 0.01321177,
      "auxiliary_loss_mlp": 0.01032107,
      "balance_loss_clip": 1.05944324,
      "balance_loss_mlp": 1.02311814,
      "epoch": 0.19791979799194373,
      "flos": 20777699041920.0,
      "grad_norm": 12.864956822226443,
      "language_loss": 0.80397975,
      "learning_rate": 3.711578044083517e-06,
      "loss": 0.82751256,
      "num_input_tokens_seen": 34860780,
      "step": 1646,
      "time_per_iteration": 2.693596601486206
    },
    {
      "auxiliary_loss_clip": 0.01319741,
      "auxiliary_loss_mlp": 0.01038244,
      "balance_loss_clip": 1.05999136,
      "balance_loss_mlp": 1.02958381,
      "epoch": 0.1980400408825828,
      "flos": 25589010084480.0,
      "grad_norm": 2.434915707221235,
      "language_loss": 0.74726009,
      "learning_rate": 3.7111749327681698e-06,
      "loss": 0.77083993,
      "num_input_tokens_seen": 34880815,
      "step": 1647,
      "time_per_iteration": 2.7016055583953857
    },
    {
      "auxiliary_loss_clip": 0.01281165,
      "auxiliary_loss_mlp": 0.01032551,
      "balance_loss_clip": 1.0677743,
      "balance_loss_mlp": 1.02434909,
      "epoch": 0.1981602837732219,
      "flos": 23513840622720.0,
      "grad_norm": 1.9499021584300493,
      "language_loss": 0.86662006,
      "learning_rate": 3.7107715618689455e-06,
      "loss": 0.88975728,
      "num_input_tokens_seen": 34899790,
      "step": 1648,
      "time_per_iteration": 2.6840555667877197
    },
    {
      "auxiliary_loss_clip": 0.01272329,
      "auxiliary_loss_mlp": 0.0103058,
      "balance_loss_clip": 1.06497228,
      "balance_loss_mlp": 1.02189565,
      "epoch": 0.198280526663861,
      "flos": 23185365724800.0,
      "grad_norm": 1.7629934197173565,
      "language_loss": 0.83423412,
      "learning_rate": 3.710367931447035e-06,
      "loss": 0.85726321,
      "num_input_tokens_seen": 34921570,
      "step": 1649,
      "time_per_iteration": 2.709613084793091
    },
    {
      "auxiliary_loss_clip": 0.01285358,
      "auxiliary_loss_mlp": 0.01038143,
      "balance_loss_clip": 1.06688917,
      "balance_loss_mlp": 1.02912486,
      "epoch": 0.1984007695545001,
      "flos": 21689470897920.0,
      "grad_norm": 2.9420838386285064,
      "language_loss": 0.86822867,
      "learning_rate": 3.70996404156367e-06,
      "loss": 0.89146376,
      "num_input_tokens_seen": 34941205,
      "step": 1650,
      "time_per_iteration": 2.597578287124634
    },
    {
      "auxiliary_loss_clip": 0.0141286,
      "auxiliary_loss_mlp": 0.01032898,
      "balance_loss_clip": 1.05712652,
      "balance_loss_mlp": 1.02388549,
      "epoch": 0.19852101244513917,
      "flos": 36064008887040.0,
      "grad_norm": 1.7857287795391952,
      "language_loss": 0.72997165,
      "learning_rate": 3.7095598922801187e-06,
      "loss": 0.75442922,
      "num_input_tokens_seen": 34963280,
      "step": 1651,
      "time_per_iteration": 3.8049657344818115
    },
    {
      "auxiliary_loss_clip": 0.01226515,
      "auxiliary_loss_mlp": 0.01034985,
      "balance_loss_clip": 1.06663799,
      "balance_loss_mlp": 1.02659845,
      "epoch": 0.19864125533577828,
      "flos": 23105894883840.0,
      "grad_norm": 3.9779927920367473,
      "language_loss": 0.76350474,
      "learning_rate": 3.7091554836576914e-06,
      "loss": 0.7861197,
      "num_input_tokens_seen": 34979955,
      "step": 1652,
      "time_per_iteration": 2.592480182647705
    },
    {
      "auxiliary_loss_clip": 0.01267206,
      "auxiliary_loss_mlp": 0.02574943,
      "balance_loss_clip": 1.05986881,
      "balance_loss_mlp": 0.99982131,
      "epoch": 0.19876149822641737,
      "flos": 24608505553920.0,
      "grad_norm": 1.8463627967460794,
      "language_loss": 0.83127427,
      "learning_rate": 3.708750815757736e-06,
      "loss": 0.86969572,
      "num_input_tokens_seen": 35000725,
      "step": 1653,
      "time_per_iteration": 2.7091140747070312
    },
    {
      "auxiliary_loss_clip": 0.01274152,
      "auxiliary_loss_mlp": 0.01036041,
      "balance_loss_clip": 1.06216621,
      "balance_loss_mlp": 1.02693319,
      "epoch": 0.19888174111705645,
      "flos": 32196645308160.0,
      "grad_norm": 4.111203368968098,
      "language_loss": 0.73352361,
      "learning_rate": 3.7083458886416407e-06,
      "loss": 0.75662553,
      "num_input_tokens_seen": 35019920,
      "step": 1654,
      "time_per_iteration": 2.7219035625457764
    },
    {
      "auxiliary_loss_clip": 0.01397142,
      "auxiliary_loss_mlp": 0.01036578,
      "balance_loss_clip": 1.05861056,
      "balance_loss_mlp": 1.02748227,
      "epoch": 0.19900198400769553,
      "flos": 24608469640320.0,
      "grad_norm": 4.82672370187564,
      "language_loss": 0.88096631,
      "learning_rate": 3.707940702370832e-06,
      "loss": 0.90530348,
      "num_input_tokens_seen": 35040765,
      "step": 1655,
      "time_per_iteration": 3.7567856311798096
    },
    {
      "auxiliary_loss_clip": 0.01183619,
      "auxiliary_loss_mlp": 0.01000082,
      "balance_loss_clip": 1.04266131,
      "balance_loss_mlp": 0.99844855,
      "epoch": 0.19912222689833464,
      "flos": 67915805673600.0,
      "grad_norm": 0.7545459574257879,
      "language_loss": 0.58276856,
      "learning_rate": 3.707535257006777e-06,
      "loss": 0.60460556,
      "num_input_tokens_seen": 35106390,
      "step": 1656,
      "time_per_iteration": 3.3170125484466553
    },
    {
      "auxiliary_loss_clip": 0.01334399,
      "auxiliary_loss_mlp": 0.01036105,
      "balance_loss_clip": 1.06816506,
      "balance_loss_mlp": 1.02637768,
      "epoch": 0.19924246978897373,
      "flos": 15742340916480.0,
      "grad_norm": 5.52228309801078,
      "language_loss": 0.88618791,
      "learning_rate": 3.707129552610981e-06,
      "loss": 0.90989298,
      "num_input_tokens_seen": 35125040,
      "step": 1657,
      "time_per_iteration": 3.5554146766662598
    },
    {
      "auxiliary_loss_clip": 0.01315483,
      "auxiliary_loss_mlp": 0.01032999,
      "balance_loss_clip": 1.05961525,
      "balance_loss_mlp": 1.02472031,
      "epoch": 0.1993627126796128,
      "flos": 17566566986880.0,
      "grad_norm": 2.0146780058519314,
      "language_loss": 0.73835731,
      "learning_rate": 3.70672358924499e-06,
      "loss": 0.76184213,
      "num_input_tokens_seen": 35144280,
      "step": 1658,
      "time_per_iteration": 2.643749713897705
    },
    {
      "auxiliary_loss_clip": 0.01375914,
      "auxiliary_loss_mlp": 0.01035854,
      "balance_loss_clip": 1.06568003,
      "balance_loss_mlp": 1.02759886,
      "epoch": 0.19948295557025192,
      "flos": 40843826680320.0,
      "grad_norm": 2.017569368247558,
      "language_loss": 0.78778106,
      "learning_rate": 3.706317366970386e-06,
      "loss": 0.81189871,
      "num_input_tokens_seen": 35165280,
      "step": 1659,
      "time_per_iteration": 3.826295852661133
    },
    {
      "auxiliary_loss_clip": 0.01226937,
      "auxiliary_loss_mlp": 0.02579213,
      "balance_loss_clip": 1.06432962,
      "balance_loss_mlp": 0.99989778,
      "epoch": 0.199603198460891,
      "flos": 25082418620160.0,
      "grad_norm": 2.0891626123938574,
      "language_loss": 0.83409524,
      "learning_rate": 3.705910885848795e-06,
      "loss": 0.87215674,
      "num_input_tokens_seen": 35183655,
      "step": 1660,
      "time_per_iteration": 2.634070634841919
    },
    {
      "auxiliary_loss_clip": 0.01275128,
      "auxiliary_loss_mlp": 0.01036516,
      "balance_loss_clip": 1.06553149,
      "balance_loss_mlp": 1.02777171,
      "epoch": 0.19972344135153008,
      "flos": 20084120352000.0,
      "grad_norm": 2.3362508821818038,
      "language_loss": 0.84930742,
      "learning_rate": 3.705504145941879e-06,
      "loss": 0.87242383,
      "num_input_tokens_seen": 35201825,
      "step": 1661,
      "time_per_iteration": 2.6389496326446533
    },
    {
      "auxiliary_loss_clip": 0.01222216,
      "auxiliary_loss_mlp": 0.0103209,
      "balance_loss_clip": 1.06341696,
      "balance_loss_mlp": 1.02309525,
      "epoch": 0.1998436842421692,
      "flos": 23727472761600.0,
      "grad_norm": 3.1470355402084804,
      "language_loss": 0.78875494,
      "learning_rate": 3.7050971473113403e-06,
      "loss": 0.81129801,
      "num_input_tokens_seen": 35221600,
      "step": 1662,
      "time_per_iteration": 2.6208863258361816
    },
    {
      "auxiliary_loss_clip": 0.01272177,
      "auxiliary_loss_mlp": 0.02573699,
      "balance_loss_clip": 1.06165528,
      "balance_loss_mlp": 0.99982476,
      "epoch": 0.19996392713280828,
      "flos": 36102361633920.0,
      "grad_norm": 2.0991158106972394,
      "language_loss": 0.79944628,
      "learning_rate": 3.7046898900189196e-06,
      "loss": 0.83790505,
      "num_input_tokens_seen": 35245935,
      "step": 1663,
      "time_per_iteration": 2.7715420722961426
    },
    {
      "auxiliary_loss_clip": 0.01377963,
      "auxiliary_loss_mlp": 0.01039795,
      "balance_loss_clip": 1.0650878,
      "balance_loss_mlp": 1.03024614,
      "epoch": 0.20008417002344736,
      "flos": 23657662679040.0,
      "grad_norm": 2.9491536421111877,
      "language_loss": 0.82873875,
      "learning_rate": 3.704282374126398e-06,
      "loss": 0.8529163,
      "num_input_tokens_seen": 35265615,
      "step": 1664,
      "time_per_iteration": 2.7438595294952393
    },
    {
      "auxiliary_loss_clip": 0.01377481,
      "auxiliary_loss_mlp": 0.01033123,
      "balance_loss_clip": 1.06274319,
      "balance_loss_mlp": 1.02404535,
      "epoch": 0.20020441291408644,
      "flos": 21872076664320.0,
      "grad_norm": 1.7475760122731514,
      "language_loss": 0.87413543,
      "learning_rate": 3.7038745996955954e-06,
      "loss": 0.89824146,
      "num_input_tokens_seen": 35284960,
      "step": 1665,
      "time_per_iteration": 2.8300352096557617
    },
    {
      "auxiliary_loss_clip": 0.01304695,
      "auxiliary_loss_mlp": 0.0103899,
      "balance_loss_clip": 1.06222439,
      "balance_loss_mlp": 1.02949524,
      "epoch": 0.20032465580472555,
      "flos": 23179691376000.0,
      "grad_norm": 3.4241071825257094,
      "language_loss": 0.72360092,
      "learning_rate": 3.703466566788371e-06,
      "loss": 0.74703783,
      "num_input_tokens_seen": 35304090,
      "step": 1666,
      "time_per_iteration": 2.846301794052124
    },
    {
      "auxiliary_loss_clip": 0.01328142,
      "auxiliary_loss_mlp": 0.0104111,
      "balance_loss_clip": 1.06756091,
      "balance_loss_mlp": 1.03160346,
      "epoch": 0.20044489869536464,
      "flos": 23873521461120.0,
      "grad_norm": 2.008750459134699,
      "language_loss": 0.74827456,
      "learning_rate": 3.703058275466622e-06,
      "loss": 0.77196711,
      "num_input_tokens_seen": 35323325,
      "step": 1667,
      "time_per_iteration": 2.814880609512329
    },
    {
      "auxiliary_loss_clip": 0.01320041,
      "auxiliary_loss_mlp": 0.01031392,
      "balance_loss_clip": 1.05873942,
      "balance_loss_mlp": 1.02314234,
      "epoch": 0.20056514158600372,
      "flos": 21945226711680.0,
      "grad_norm": 7.347942489737876,
      "language_loss": 0.777843,
      "learning_rate": 3.7026497257922877e-06,
      "loss": 0.80135733,
      "num_input_tokens_seen": 35343635,
      "step": 1668,
      "time_per_iteration": 2.750514030456543
    },
    {
      "auxiliary_loss_clip": 0.01422254,
      "auxiliary_loss_mlp": 0.01034763,
      "balance_loss_clip": 1.05843413,
      "balance_loss_mlp": 1.02586436,
      "epoch": 0.20068538447664283,
      "flos": 23879159896320.0,
      "grad_norm": 1.7411298520704428,
      "language_loss": 0.85087669,
      "learning_rate": 3.7022409178273436e-06,
      "loss": 0.87544692,
      "num_input_tokens_seen": 35364615,
      "step": 1669,
      "time_per_iteration": 2.7990002632141113
    },
    {
      "auxiliary_loss_clip": 0.01268626,
      "auxiliary_loss_mlp": 0.01034228,
      "balance_loss_clip": 1.05951357,
      "balance_loss_mlp": 1.02538848,
      "epoch": 0.2008056273672819,
      "flos": 18442823270400.0,
      "grad_norm": 2.9427474532139666,
      "language_loss": 0.78642505,
      "learning_rate": 3.7018318516338054e-06,
      "loss": 0.80945355,
      "num_input_tokens_seen": 35383775,
      "step": 1670,
      "time_per_iteration": 2.627643346786499
    },
    {
      "auxiliary_loss_clip": 0.0128135,
      "auxiliary_loss_mlp": 0.01028256,
      "balance_loss_clip": 1.06601119,
      "balance_loss_mlp": 1.02003026,
      "epoch": 0.200925870257921,
      "flos": 23659530186240.0,
      "grad_norm": 2.5854287198896504,
      "language_loss": 0.81203669,
      "learning_rate": 3.7014225272737284e-06,
      "loss": 0.83513272,
      "num_input_tokens_seen": 35403000,
      "step": 1671,
      "time_per_iteration": 2.6971592903137207
    },
    {
      "auxiliary_loss_clip": 0.01267941,
      "auxiliary_loss_mlp": 0.01033244,
      "balance_loss_clip": 1.06192994,
      "balance_loss_mlp": 1.0241127,
      "epoch": 0.20104611314856008,
      "flos": 16217115909120.0,
      "grad_norm": 2.8196039912423387,
      "language_loss": 0.74118567,
      "learning_rate": 3.701012944809207e-06,
      "loss": 0.76419753,
      "num_input_tokens_seen": 35420115,
      "step": 1672,
      "time_per_iteration": 2.65588641166687
    },
    {
      "auxiliary_loss_clip": 0.01325154,
      "auxiliary_loss_mlp": 0.02577064,
      "balance_loss_clip": 1.06522584,
      "balance_loss_mlp": 0.99983704,
      "epoch": 0.2011663560391992,
      "flos": 21397373498880.0,
      "grad_norm": 2.3434051998572745,
      "language_loss": 0.78958309,
      "learning_rate": 3.700603104302374e-06,
      "loss": 0.82860529,
      "num_input_tokens_seen": 35439925,
      "step": 1673,
      "time_per_iteration": 2.6895415782928467
    },
    {
      "auxiliary_loss_clip": 0.01264681,
      "auxiliary_loss_mlp": 0.01001912,
      "balance_loss_clip": 1.0291481,
      "balance_loss_mlp": 1.00010014,
      "epoch": 0.20128659892983827,
      "flos": 62229459409920.0,
      "grad_norm": 0.9200876188886893,
      "language_loss": 0.55956268,
      "learning_rate": 3.7001930058154027e-06,
      "loss": 0.58222866,
      "num_input_tokens_seen": 35504885,
      "step": 1674,
      "time_per_iteration": 3.3157219886779785
    },
    {
      "auxiliary_loss_clip": 0.01370544,
      "auxiliary_loss_mlp": 0.01041009,
      "balance_loss_clip": 1.05983293,
      "balance_loss_mlp": 1.03156161,
      "epoch": 0.20140684182047736,
      "flos": 28438737448320.0,
      "grad_norm": 2.8489277070641745,
      "language_loss": 0.79951513,
      "learning_rate": 3.6997826494105037e-06,
      "loss": 0.82363057,
      "num_input_tokens_seen": 35525330,
      "step": 1675,
      "time_per_iteration": 2.75981068611145
    },
    {
      "auxiliary_loss_clip": 0.0132397,
      "auxiliary_loss_mlp": 0.01031985,
      "balance_loss_clip": 1.06301785,
      "balance_loss_mlp": 1.02215636,
      "epoch": 0.20152708471111647,
      "flos": 28074064619520.0,
      "grad_norm": 4.850010425315004,
      "language_loss": 0.69830894,
      "learning_rate": 3.6993720351499286e-06,
      "loss": 0.72186852,
      "num_input_tokens_seen": 35546455,
      "step": 1676,
      "time_per_iteration": 2.8245737552642822
    },
    {
      "auxiliary_loss_clip": 0.01331477,
      "auxiliary_loss_mlp": 0.01033736,
      "balance_loss_clip": 1.07280171,
      "balance_loss_mlp": 1.02454519,
      "epoch": 0.20164732760175555,
      "flos": 23549751244800.0,
      "grad_norm": 2.0451117016223592,
      "language_loss": 0.76975119,
      "learning_rate": 3.6989611630959666e-06,
      "loss": 0.79340327,
      "num_input_tokens_seen": 35565010,
      "step": 1677,
      "time_per_iteration": 3.7269303798675537
    },
    {
      "auxiliary_loss_clip": 0.01181521,
      "auxiliary_loss_mlp": 0.01004492,
      "balance_loss_clip": 1.04058337,
      "balance_loss_mlp": 1.00270355,
      "epoch": 0.20176757049239463,
      "flos": 71100616037760.0,
      "grad_norm": 0.7100113937801776,
      "language_loss": 0.58250046,
      "learning_rate": 3.6985500333109474e-06,
      "loss": 0.60436058,
      "num_input_tokens_seen": 35633340,
      "step": 1678,
      "time_per_iteration": 3.284301280975342
    },
    {
      "auxiliary_loss_clip": 0.01365833,
      "auxiliary_loss_mlp": 0.01032169,
      "balance_loss_clip": 1.05786407,
      "balance_loss_mlp": 1.0231986,
      "epoch": 0.20188781338303372,
      "flos": 21430159637760.0,
      "grad_norm": 3.604574505055815,
      "language_loss": 0.76035565,
      "learning_rate": 3.6981386458572385e-06,
      "loss": 0.78433567,
      "num_input_tokens_seen": 35651315,
      "step": 1679,
      "time_per_iteration": 3.665034294128418
    },
    {
      "auxiliary_loss_clip": 0.01371209,
      "auxiliary_loss_mlp": 0.01032261,
      "balance_loss_clip": 1.05997419,
      "balance_loss_mlp": 1.02144873,
      "epoch": 0.20200805627367283,
      "flos": 11546215130880.0,
      "grad_norm": 2.2914695103364973,
      "language_loss": 0.76098937,
      "learning_rate": 3.6977270007972468e-06,
      "loss": 0.78502405,
      "num_input_tokens_seen": 35668850,
      "step": 1680,
      "time_per_iteration": 2.7360188961029053
    },
    {
      "auxiliary_loss_clip": 0.01327373,
      "auxiliary_loss_mlp": 0.01045641,
      "balance_loss_clip": 1.06296182,
      "balance_loss_mlp": 1.0368607,
      "epoch": 0.2021282991643119,
      "flos": 28545391906560.0,
      "grad_norm": 2.2790244707231264,
      "language_loss": 0.72214818,
      "learning_rate": 3.6973150981934196e-06,
      "loss": 0.74587834,
      "num_input_tokens_seen": 35690080,
      "step": 1681,
      "time_per_iteration": 2.723375082015991
    },
    {
      "auxiliary_loss_clip": 0.01229956,
      "auxiliary_loss_mlp": 0.01042086,
      "balance_loss_clip": 1.06610656,
      "balance_loss_mlp": 1.03337789,
      "epoch": 0.202248542054951,
      "flos": 17923446564480.0,
      "grad_norm": 6.85945107841545,
      "language_loss": 0.83658177,
      "learning_rate": 3.6969029381082415e-06,
      "loss": 0.85930216,
      "num_input_tokens_seen": 35706075,
      "step": 1682,
      "time_per_iteration": 3.5132176876068115
    },
    {
      "auxiliary_loss_clip": 0.01244998,
      "auxiliary_loss_mlp": 0.01034252,
      "balance_loss_clip": 1.06225336,
      "balance_loss_mlp": 1.02570462,
      "epoch": 0.2023687849455901,
      "flos": 19864634296320.0,
      "grad_norm": 1.691394996584591,
      "language_loss": 0.79480517,
      "learning_rate": 3.696490520604237e-06,
      "loss": 0.81759763,
      "num_input_tokens_seen": 35724765,
      "step": 1683,
      "time_per_iteration": 2.688344955444336
    },
    {
      "auxiliary_loss_clip": 0.0127087,
      "auxiliary_loss_mlp": 0.01030676,
      "balance_loss_clip": 1.06498289,
      "balance_loss_mlp": 1.02289701,
      "epoch": 0.20248902783622919,
      "flos": 22564721600640.0,
      "grad_norm": 2.070584048190453,
      "language_loss": 0.80574334,
      "learning_rate": 3.696077845743968e-06,
      "loss": 0.82875878,
      "num_input_tokens_seen": 35744355,
      "step": 1684,
      "time_per_iteration": 2.696791887283325
    },
    {
      "auxiliary_loss_clip": 0.01225769,
      "auxiliary_loss_mlp": 0.01030843,
      "balance_loss_clip": 1.06374061,
      "balance_loss_mlp": 1.02058482,
      "epoch": 0.20260927072686827,
      "flos": 22709728805760.0,
      "grad_norm": 3.246539925304922,
      "language_loss": 0.73417181,
      "learning_rate": 3.69566491359004e-06,
      "loss": 0.75673795,
      "num_input_tokens_seen": 35761000,
      "step": 1685,
      "time_per_iteration": 3.5936810970306396
    },
    {
      "auxiliary_loss_clip": 0.01321669,
      "auxiliary_loss_mlp": 0.01034412,
      "balance_loss_clip": 1.05793643,
      "balance_loss_mlp": 1.02516687,
      "epoch": 0.20272951361750738,
      "flos": 51023998650240.0,
      "grad_norm": 1.9953262378128434,
      "language_loss": 0.69378853,
      "learning_rate": 3.695251724205092e-06,
      "loss": 0.71734929,
      "num_input_tokens_seen": 35785360,
      "step": 1686,
      "time_per_iteration": 2.927419662475586
    },
    {
      "auxiliary_loss_clip": 0.01225291,
      "auxiliary_loss_mlp": 0.01038524,
      "balance_loss_clip": 1.0660789,
      "balance_loss_mlp": 1.02948213,
      "epoch": 0.20284975650814646,
      "flos": 26578133879040.0,
      "grad_norm": 2.082580894639418,
      "language_loss": 0.86496037,
      "learning_rate": 3.6948382776518054e-06,
      "loss": 0.88759851,
      "num_input_tokens_seen": 35806065,
      "step": 1687,
      "time_per_iteration": 2.7014248371124268
    },
    {
      "auxiliary_loss_clip": 0.01381268,
      "auxiliary_loss_mlp": 0.0103289,
      "balance_loss_clip": 1.05974865,
      "balance_loss_mlp": 1.02362156,
      "epoch": 0.20296999939878554,
      "flos": 16034222833920.0,
      "grad_norm": 2.629379662004305,
      "language_loss": 0.79582012,
      "learning_rate": 3.6944245739929e-06,
      "loss": 0.81996167,
      "num_input_tokens_seen": 35822225,
      "step": 1688,
      "time_per_iteration": 2.804232120513916
    },
    {
      "auxiliary_loss_clip": 0.01280102,
      "auxiliary_loss_mlp": 0.01033204,
      "balance_loss_clip": 1.06789589,
      "balance_loss_mlp": 1.02369118,
      "epoch": 0.20309024228942463,
      "flos": 19203374868480.0,
      "grad_norm": 2.2709526950643717,
      "language_loss": 0.71940595,
      "learning_rate": 3.6940106132911332e-06,
      "loss": 0.74253905,
      "num_input_tokens_seen": 35839410,
      "step": 1689,
      "time_per_iteration": 2.668020009994507
    },
    {
      "auxiliary_loss_clip": 0.012764,
      "auxiliary_loss_mlp": 0.01034824,
      "balance_loss_clip": 1.06584299,
      "balance_loss_mlp": 1.02663982,
      "epoch": 0.20321048518006374,
      "flos": 22821087945600.0,
      "grad_norm": 2.0071278942120307,
      "language_loss": 0.88753408,
      "learning_rate": 3.6935963956093037e-06,
      "loss": 0.91064632,
      "num_input_tokens_seen": 35859495,
      "step": 1690,
      "time_per_iteration": 2.625235080718994
    },
    {
      "auxiliary_loss_clip": 0.01267848,
      "auxiliary_loss_mlp": 0.0103717,
      "balance_loss_clip": 1.06277204,
      "balance_loss_mlp": 1.02881968,
      "epoch": 0.20333072807070282,
      "flos": 19096397187840.0,
      "grad_norm": 2.4890665604988675,
      "language_loss": 0.6918695,
      "learning_rate": 3.6931819210102474e-06,
      "loss": 0.71491969,
      "num_input_tokens_seen": 35878890,
      "step": 1691,
      "time_per_iteration": 2.699068546295166
    },
    {
      "auxiliary_loss_clip": 0.01226903,
      "auxiliary_loss_mlp": 0.01032456,
      "balance_loss_clip": 1.06556296,
      "balance_loss_mlp": 1.02343166,
      "epoch": 0.2034509709613419,
      "flos": 18180962144640.0,
      "grad_norm": 2.184465144649324,
      "language_loss": 0.8461051,
      "learning_rate": 3.6927671895568402e-06,
      "loss": 0.86869866,
      "num_input_tokens_seen": 35897950,
      "step": 1692,
      "time_per_iteration": 2.5956993103027344
    },
    {
      "auxiliary_loss_clip": 0.01230696,
      "auxiliary_loss_mlp": 0.01030919,
      "balance_loss_clip": 1.06898785,
      "balance_loss_mlp": 1.02206206,
      "epoch": 0.20357121385198101,
      "flos": 22923899648640.0,
      "grad_norm": 2.021867251389478,
      "language_loss": 0.87210697,
      "learning_rate": 3.692352201311996e-06,
      "loss": 0.89472318,
      "num_input_tokens_seen": 35916800,
      "step": 1693,
      "time_per_iteration": 2.6266164779663086
    },
    {
      "auxiliary_loss_clip": 0.01370027,
      "auxiliary_loss_mlp": 0.01032275,
      "balance_loss_clip": 1.06009388,
      "balance_loss_mlp": 1.02337027,
      "epoch": 0.2036914567426201,
      "flos": 20922131629440.0,
      "grad_norm": 1.8460726947749244,
      "language_loss": 0.76840031,
      "learning_rate": 3.6919369563386687e-06,
      "loss": 0.79242337,
      "num_input_tokens_seen": 35936600,
      "step": 1694,
      "time_per_iteration": 2.729376792907715
    },
    {
      "auxiliary_loss_clip": 0.01319847,
      "auxiliary_loss_mlp": 0.01032456,
      "balance_loss_clip": 1.06499648,
      "balance_loss_mlp": 1.02411771,
      "epoch": 0.20381169963325918,
      "flos": 15519155760000.0,
      "grad_norm": 2.304111998423511,
      "language_loss": 0.7890687,
      "learning_rate": 3.69152145469985e-06,
      "loss": 0.81259179,
      "num_input_tokens_seen": 35953645,
      "step": 1695,
      "time_per_iteration": 2.678624153137207
    },
    {
      "auxiliary_loss_clip": 0.01423939,
      "auxiliary_loss_mlp": 0.01034151,
      "balance_loss_clip": 1.05842352,
      "balance_loss_mlp": 1.02434587,
      "epoch": 0.20393194252389826,
      "flos": 28833143760000.0,
      "grad_norm": 3.46805242552658,
      "language_loss": 0.82432044,
      "learning_rate": 3.691105696458572e-06,
      "loss": 0.84890127,
      "num_input_tokens_seen": 35970940,
      "step": 1696,
      "time_per_iteration": 2.7951457500457764
    },
    {
      "auxiliary_loss_clip": 0.01229361,
      "auxiliary_loss_mlp": 0.01029436,
      "balance_loss_clip": 1.07042587,
      "balance_loss_mlp": 1.02061415,
      "epoch": 0.20405218541453737,
      "flos": 22488554810880.0,
      "grad_norm": 3.35183288827725,
      "language_loss": 0.68442309,
      "learning_rate": 3.690689681677904e-06,
      "loss": 0.70701104,
      "num_input_tokens_seen": 35989410,
      "step": 1697,
      "time_per_iteration": 2.6266632080078125
    },
    {
      "auxiliary_loss_clip": 0.013245,
      "auxiliary_loss_mlp": 0.01034465,
      "balance_loss_clip": 1.06032741,
      "balance_loss_mlp": 1.02598894,
      "epoch": 0.20417242830517646,
      "flos": 25374408278400.0,
      "grad_norm": 1.7559516687438865,
      "language_loss": 0.88647115,
      "learning_rate": 3.690273410420956e-06,
      "loss": 0.91006076,
      "num_input_tokens_seen": 36009175,
      "step": 1698,
      "time_per_iteration": 2.749053955078125
    },
    {
      "auxiliary_loss_clip": 0.01271157,
      "auxiliary_loss_mlp": 0.01033254,
      "balance_loss_clip": 1.06364775,
      "balance_loss_mlp": 1.02458191,
      "epoch": 0.20429267119581554,
      "flos": 14793078240000.0,
      "grad_norm": 2.5971573943267963,
      "language_loss": 0.76430291,
      "learning_rate": 3.689856882750875e-06,
      "loss": 0.78734702,
      "num_input_tokens_seen": 36024375,
      "step": 1699,
      "time_per_iteration": 2.6545238494873047
    },
    {
      "auxiliary_loss_clip": 0.01270598,
      "auxiliary_loss_mlp": 0.01037213,
      "balance_loss_clip": 1.0669558,
      "balance_loss_mlp": 1.02877331,
      "epoch": 0.20441291408645465,
      "flos": 17781851151360.0,
      "grad_norm": 1.7647510006711382,
      "language_loss": 0.7874831,
      "learning_rate": 3.6894400987308486e-06,
      "loss": 0.81056118,
      "num_input_tokens_seen": 36041895,
      "step": 1700,
      "time_per_iteration": 2.6769585609436035
    },
    {
      "auxiliary_loss_clip": 0.01277441,
      "auxiliary_loss_mlp": 0.01030548,
      "balance_loss_clip": 1.06326032,
      "balance_loss_mlp": 1.02172017,
      "epoch": 0.20453315697709373,
      "flos": 16435668211200.0,
      "grad_norm": 2.9430100475385808,
      "language_loss": 0.85330898,
      "learning_rate": 3.6890230584241024e-06,
      "loss": 0.87638885,
      "num_input_tokens_seen": 36058825,
      "step": 1701,
      "time_per_iteration": 2.62534761428833
    },
    {
      "auxiliary_loss_clip": 0.01130801,
      "auxiliary_loss_mlp": 0.01002604,
      "balance_loss_clip": 1.04127932,
      "balance_loss_mlp": 1.00095928,
      "epoch": 0.20465339986773282,
      "flos": 66713085653760.0,
      "grad_norm": 1.0804683706377751,
      "language_loss": 0.66401911,
      "learning_rate": 3.6886057618939016e-06,
      "loss": 0.68535316,
      "num_input_tokens_seen": 36121645,
      "step": 1702,
      "time_per_iteration": 3.2759385108947754
    },
    {
      "auxiliary_loss_clip": 0.01298187,
      "auxiliary_loss_mlp": 0.01034362,
      "balance_loss_clip": 1.06118989,
      "balance_loss_mlp": 1.02530789,
      "epoch": 0.2047736427583719,
      "flos": 41974114924800.0,
      "grad_norm": 2.0316128916545564,
      "language_loss": 0.6887942,
      "learning_rate": 3.6881882092035492e-06,
      "loss": 0.71211964,
      "num_input_tokens_seen": 36143030,
      "step": 1703,
      "time_per_iteration": 3.7965002059936523
    },
    {
      "auxiliary_loss_clip": 0.0129342,
      "auxiliary_loss_mlp": 0.02528347,
      "balance_loss_clip": 1.03952289,
      "balance_loss_mlp": 0.9996891,
      "epoch": 0.204893885649011,
      "flos": 69940878641280.0,
      "grad_norm": 0.9393771280061165,
      "language_loss": 0.61222589,
      "learning_rate": 3.6877704004163873e-06,
      "loss": 0.65044355,
      "num_input_tokens_seen": 36203435,
      "step": 1704,
      "time_per_iteration": 3.385798454284668
    },
    {
      "auxiliary_loss_clip": 0.01227521,
      "auxiliary_loss_mlp": 0.01033135,
      "balance_loss_clip": 1.06591773,
      "balance_loss_mlp": 1.02388477,
      "epoch": 0.2050141285396501,
      "flos": 22200012858240.0,
      "grad_norm": 1.8395011068794145,
      "language_loss": 0.77852625,
      "learning_rate": 3.6873523355957984e-06,
      "loss": 0.8011328,
      "num_input_tokens_seen": 36222435,
      "step": 1705,
      "time_per_iteration": 3.649529218673706
    },
    {
      "auxiliary_loss_clip": 0.0112819,
      "auxiliary_loss_mlp": 0.01000645,
      "balance_loss_clip": 1.03906333,
      "balance_loss_mlp": 0.99896395,
      "epoch": 0.20513437143028918,
      "flos": 46283721730560.0,
      "grad_norm": 1.0372227072027458,
      "language_loss": 0.64085948,
      "learning_rate": 3.686934014805201e-06,
      "loss": 0.66214776,
      "num_input_tokens_seen": 36273065,
      "step": 1706,
      "time_per_iteration": 3.050250291824341
    },
    {
      "auxiliary_loss_clip": 0.01275895,
      "auxiliary_loss_mlp": 0.01042837,
      "balance_loss_clip": 1.06679106,
      "balance_loss_mlp": 1.03373551,
      "epoch": 0.20525461432092829,
      "flos": 21904324099200.0,
      "grad_norm": 1.8270061935013222,
      "language_loss": 0.81094134,
      "learning_rate": 3.6865154381080552e-06,
      "loss": 0.83412862,
      "num_input_tokens_seen": 36293750,
      "step": 1707,
      "time_per_iteration": 2.6549713611602783
    },
    {
      "auxiliary_loss_clip": 0.01519603,
      "auxiliary_loss_mlp": 0.01033156,
      "balance_loss_clip": 1.05395532,
      "balance_loss_mlp": 1.02441764,
      "epoch": 0.20537485721156737,
      "flos": 21214264942080.0,
      "grad_norm": 2.5054879899293216,
      "language_loss": 0.82816321,
      "learning_rate": 3.6860966055678585e-06,
      "loss": 0.8536908,
      "num_input_tokens_seen": 36310105,
      "step": 1708,
      "time_per_iteration": 3.7794735431671143
    },
    {
      "auxiliary_loss_clip": 0.01275493,
      "auxiliary_loss_mlp": 0.01031655,
      "balance_loss_clip": 1.06605613,
      "balance_loss_mlp": 1.02210593,
      "epoch": 0.20549510010220645,
      "flos": 20191205773440.0,
      "grad_norm": 2.1401690789310157,
      "language_loss": 0.8653003,
      "learning_rate": 3.685677517248147e-06,
      "loss": 0.88837171,
      "num_input_tokens_seen": 36328995,
      "step": 1709,
      "time_per_iteration": 2.682203531265259
    },
    {
      "auxiliary_loss_clip": 0.01320501,
      "auxiliary_loss_mlp": 0.0257455,
      "balance_loss_clip": 1.06568813,
      "balance_loss_mlp": 0.99987012,
      "epoch": 0.20561534299284553,
      "flos": 17016702612480.0,
      "grad_norm": 9.79805664118204,
      "language_loss": 0.80551362,
      "learning_rate": 3.6852581732124967e-06,
      "loss": 0.84446412,
      "num_input_tokens_seen": 36346340,
      "step": 1710,
      "time_per_iteration": 2.67453670501709
    },
    {
      "auxiliary_loss_clip": 0.01275807,
      "auxiliary_loss_mlp": 0.01033037,
      "balance_loss_clip": 1.06626153,
      "balance_loss_mlp": 1.02371526,
      "epoch": 0.20573558588348465,
      "flos": 22890467064960.0,
      "grad_norm": 1.8610296955554615,
      "language_loss": 0.76307273,
      "learning_rate": 3.6848385735245213e-06,
      "loss": 0.78616118,
      "num_input_tokens_seen": 36365430,
      "step": 1711,
      "time_per_iteration": 3.618485927581787
    },
    {
      "auxiliary_loss_clip": 0.012624,
      "auxiliary_loss_mlp": 0.01036655,
      "balance_loss_clip": 1.0578469,
      "balance_loss_mlp": 1.02794111,
      "epoch": 0.20585582877412373,
      "flos": 24643123286400.0,
      "grad_norm": 1.9909103560763126,
      "language_loss": 0.861525,
      "learning_rate": 3.6844187182478734e-06,
      "loss": 0.88451558,
      "num_input_tokens_seen": 36386285,
      "step": 1712,
      "time_per_iteration": 2.7200582027435303
    },
    {
      "auxiliary_loss_clip": 0.01317113,
      "auxiliary_loss_mlp": 0.01030743,
      "balance_loss_clip": 1.05878687,
      "balance_loss_mlp": 1.02161729,
      "epoch": 0.2059760716647628,
      "flos": 24206952435840.0,
      "grad_norm": 1.7870360130062108,
      "language_loss": 0.75278431,
      "learning_rate": 3.683998607446246e-06,
      "loss": 0.77626288,
      "num_input_tokens_seen": 36404935,
      "step": 1713,
      "time_per_iteration": 2.7512638568878174
    },
    {
      "auxiliary_loss_clip": 0.01279654,
      "auxiliary_loss_mlp": 0.0103416,
      "balance_loss_clip": 1.06811404,
      "balance_loss_mlp": 1.02565992,
      "epoch": 0.20609631455540192,
      "flos": 20229522606720.0,
      "grad_norm": 2.493863199302456,
      "language_loss": 0.75440609,
      "learning_rate": 3.6835782411833686e-06,
      "loss": 0.77754414,
      "num_input_tokens_seen": 36424455,
      "step": 1714,
      "time_per_iteration": 2.6189229488372803
    },
    {
      "auxiliary_loss_clip": 0.0136037,
      "auxiliary_loss_mlp": 0.01034588,
      "balance_loss_clip": 1.05821443,
      "balance_loss_mlp": 1.02603459,
      "epoch": 0.206216557446041,
      "flos": 19864957518720.0,
      "grad_norm": 1.9286203832992679,
      "language_loss": 0.74052942,
      "learning_rate": 3.68315761952301e-06,
      "loss": 0.76447904,
      "num_input_tokens_seen": 36441685,
      "step": 1715,
      "time_per_iteration": 2.770328998565674
    },
    {
      "auxiliary_loss_clip": 0.01226084,
      "auxiliary_loss_mlp": 0.01037958,
      "balance_loss_clip": 1.06686676,
      "balance_loss_mlp": 1.02930927,
      "epoch": 0.2063368003366801,
      "flos": 24096311568000.0,
      "grad_norm": 2.0262083934546036,
      "language_loss": 0.83388329,
      "learning_rate": 3.6827367425289797e-06,
      "loss": 0.85652369,
      "num_input_tokens_seen": 36461460,
      "step": 1716,
      "time_per_iteration": 2.6359527111053467
    },
    {
      "auxiliary_loss_clip": 0.0131973,
      "auxiliary_loss_mlp": 0.01030916,
      "balance_loss_clip": 1.06310081,
      "balance_loss_mlp": 1.02182651,
      "epoch": 0.2064570432273192,
      "flos": 20340163474560.0,
      "grad_norm": 2.643337097838561,
      "language_loss": 0.72670609,
      "learning_rate": 3.6823156102651225e-06,
      "loss": 0.75021255,
      "num_input_tokens_seen": 36479615,
      "step": 1717,
      "time_per_iteration": 2.7509500980377197
    },
    {
      "auxiliary_loss_clip": 0.01458158,
      "auxiliary_loss_mlp": 0.0103724,
      "balance_loss_clip": 1.05575335,
      "balance_loss_mlp": 1.02832305,
      "epoch": 0.20657728611795828,
      "flos": 20520363029760.0,
      "grad_norm": 2.222896996465956,
      "language_loss": 0.70807743,
      "learning_rate": 3.6818942227953257e-06,
      "loss": 0.73303139,
      "num_input_tokens_seen": 36500160,
      "step": 1718,
      "time_per_iteration": 2.8121814727783203
    },
    {
      "auxiliary_loss_clip": 0.01375539,
      "auxiliary_loss_mlp": 0.01036572,
      "balance_loss_clip": 1.0619061,
      "balance_loss_mlp": 1.02808428,
      "epoch": 0.20669752900859736,
      "flos": 21799285752960.0,
      "grad_norm": 2.1305746952922076,
      "language_loss": 0.68848747,
      "learning_rate": 3.681472580183512e-06,
      "loss": 0.71260858,
      "num_input_tokens_seen": 36518810,
      "step": 1719,
      "time_per_iteration": 2.779588222503662
    },
    {
      "auxiliary_loss_clip": 0.01270087,
      "auxiliary_loss_mlp": 0.01034506,
      "balance_loss_clip": 1.06375289,
      "balance_loss_mlp": 1.02615559,
      "epoch": 0.20681777189923645,
      "flos": 15122020014720.0,
      "grad_norm": 2.0407202710539676,
      "language_loss": 0.86124218,
      "learning_rate": 3.6810506824936455e-06,
      "loss": 0.88428807,
      "num_input_tokens_seen": 36536890,
      "step": 1720,
      "time_per_iteration": 2.6859185695648193
    },
    {
      "auxiliary_loss_clip": 0.01223899,
      "auxiliary_loss_mlp": 0.01004349,
      "balance_loss_clip": 1.03211451,
      "balance_loss_mlp": 1.00276375,
      "epoch": 0.20693801478987556,
      "flos": 56481021509760.0,
      "grad_norm": 1.0253389557418644,
      "language_loss": 0.62522817,
      "learning_rate": 3.680628529789726e-06,
      "loss": 0.64751065,
      "num_input_tokens_seen": 36589300,
      "step": 1721,
      "time_per_iteration": 3.207810878753662
    },
    {
      "auxiliary_loss_clip": 0.01230421,
      "auxiliary_loss_mlp": 0.01038388,
      "balance_loss_clip": 1.06694508,
      "balance_loss_mlp": 1.02863717,
      "epoch": 0.20705825768051464,
      "flos": 21614201948160.0,
      "grad_norm": 2.7459276978122915,
      "language_loss": 0.862643,
      "learning_rate": 3.680206122135796e-06,
      "loss": 0.88533109,
      "num_input_tokens_seen": 36609905,
      "step": 1722,
      "time_per_iteration": 2.6467299461364746
    },
    {
      "auxiliary_loss_clip": 0.01428142,
      "auxiliary_loss_mlp": 0.01040387,
      "balance_loss_clip": 1.06271458,
      "balance_loss_mlp": 1.03179204,
      "epoch": 0.20717850057115372,
      "flos": 25848895962240.0,
      "grad_norm": 1.9247565899619996,
      "language_loss": 0.78343666,
      "learning_rate": 3.6797834595959323e-06,
      "loss": 0.80812204,
      "num_input_tokens_seen": 36629805,
      "step": 1723,
      "time_per_iteration": 2.878404140472412
    },
    {
      "auxiliary_loss_clip": 0.01416573,
      "auxiliary_loss_mlp": 0.01036331,
      "balance_loss_clip": 1.05715394,
      "balance_loss_mlp": 1.02639461,
      "epoch": 0.20729874346179283,
      "flos": 29130807767040.0,
      "grad_norm": 2.7373245652356264,
      "language_loss": 0.7773127,
      "learning_rate": 3.679360542234254e-06,
      "loss": 0.80184174,
      "num_input_tokens_seen": 36649150,
      "step": 1724,
      "time_per_iteration": 2.8337326049804688
    },
    {
      "auxiliary_loss_clip": 0.01315276,
      "auxiliary_loss_mlp": 0.02576445,
      "balance_loss_clip": 1.05540848,
      "balance_loss_mlp": 0.99999416,
      "epoch": 0.20741898635243192,
      "flos": 29023363209600.0,
      "grad_norm": 2.299169074848696,
      "language_loss": 0.72686625,
      "learning_rate": 3.678937370114916e-06,
      "loss": 0.76578343,
      "num_input_tokens_seen": 36668955,
      "step": 1725,
      "time_per_iteration": 2.7990612983703613
    },
    {
      "auxiliary_loss_clip": 0.01317617,
      "auxiliary_loss_mlp": 0.01031434,
      "balance_loss_clip": 1.06241095,
      "balance_loss_mlp": 1.02370334,
      "epoch": 0.207539229243071,
      "flos": 15559447841280.0,
      "grad_norm": 2.037022919483095,
      "language_loss": 0.78655446,
      "learning_rate": 3.678513943302114e-06,
      "loss": 0.810045,
      "num_input_tokens_seen": 36685730,
      "step": 1726,
      "time_per_iteration": 2.8272628784179688
    },
    {
      "auxiliary_loss_clip": 0.01222643,
      "auxiliary_loss_mlp": 0.01036626,
      "balance_loss_clip": 1.06505609,
      "balance_loss_mlp": 1.02749503,
      "epoch": 0.20765947213371008,
      "flos": 20521081301760.0,
      "grad_norm": 3.7254250067592234,
      "language_loss": 0.85263479,
      "learning_rate": 3.678090261860082e-06,
      "loss": 0.87522739,
      "num_input_tokens_seen": 36705460,
      "step": 1727,
      "time_per_iteration": 2.671719551086426
    },
    {
      "auxiliary_loss_clip": 0.01372209,
      "auxiliary_loss_mlp": 0.01035283,
      "balance_loss_clip": 1.05554938,
      "balance_loss_mlp": 1.02622271,
      "epoch": 0.2077797150243492,
      "flos": 19354415558400.0,
      "grad_norm": 2.9632718390182693,
      "language_loss": 0.77425575,
      "learning_rate": 3.6776663258530906e-06,
      "loss": 0.79833066,
      "num_input_tokens_seen": 36724110,
      "step": 1728,
      "time_per_iteration": 2.771486759185791
    },
    {
      "auxiliary_loss_clip": 0.01275482,
      "auxiliary_loss_mlp": 0.01039287,
      "balance_loss_clip": 1.06305742,
      "balance_loss_mlp": 1.03100252,
      "epoch": 0.20789995791498828,
      "flos": 21829952989440.0,
      "grad_norm": 2.471891837246355,
      "language_loss": 0.7140522,
      "learning_rate": 3.6772421353454516e-06,
      "loss": 0.73719996,
      "num_input_tokens_seen": 36742705,
      "step": 1729,
      "time_per_iteration": 3.7716257572174072
    },
    {
      "auxiliary_loss_clip": 0.01272886,
      "auxiliary_loss_mlp": 0.01035307,
      "balance_loss_clip": 1.06518805,
      "balance_loss_mlp": 1.02635396,
      "epoch": 0.20802020080562736,
      "flos": 23148844571520.0,
      "grad_norm": 1.8708959961133214,
      "language_loss": 0.88950652,
      "learning_rate": 3.6768176904015153e-06,
      "loss": 0.91258848,
      "num_input_tokens_seen": 36762510,
      "step": 1730,
      "time_per_iteration": 2.638502836227417
    },
    {
      "auxiliary_loss_clip": 0.0127689,
      "auxiliary_loss_mlp": 0.0103397,
      "balance_loss_clip": 1.06461203,
      "balance_loss_mlp": 1.02439129,
      "epoch": 0.20814044369626647,
      "flos": 23072677781760.0,
      "grad_norm": 2.4868299715985174,
      "language_loss": 0.6011585,
      "learning_rate": 3.6763929910856674e-06,
      "loss": 0.6242671,
      "num_input_tokens_seen": 36780960,
      "step": 1731,
      "time_per_iteration": 3.643238067626953
    },
    {
      "auxiliary_loss_clip": 0.01273704,
      "auxiliary_loss_mlp": 0.01039885,
      "balance_loss_clip": 1.06446779,
      "balance_loss_mlp": 1.03055739,
      "epoch": 0.20826068658690555,
      "flos": 19608016556160.0,
      "grad_norm": 2.6329506134499754,
      "language_loss": 0.77624631,
      "learning_rate": 3.6759680374623365e-06,
      "loss": 0.79938215,
      "num_input_tokens_seen": 36798875,
      "step": 1732,
      "time_per_iteration": 2.7076990604400635
    },
    {
      "auxiliary_loss_clip": 0.01226388,
      "auxiliary_loss_mlp": 0.01028146,
      "balance_loss_clip": 1.06867456,
      "balance_loss_mlp": 1.0195868,
      "epoch": 0.20838092947754464,
      "flos": 25374049142400.0,
      "grad_norm": 2.3900421590955125,
      "language_loss": 0.75257939,
      "learning_rate": 3.675542829595986e-06,
      "loss": 0.77512479,
      "num_input_tokens_seen": 36818540,
      "step": 1733,
      "time_per_iteration": 2.628584861755371
    },
    {
      "auxiliary_loss_clip": 0.01320551,
      "auxiliary_loss_mlp": 0.01033545,
      "balance_loss_clip": 1.06103384,
      "balance_loss_mlp": 1.02508092,
      "epoch": 0.20850117236818372,
      "flos": 24061729749120.0,
      "grad_norm": 1.881764362917191,
      "language_loss": 0.79538709,
      "learning_rate": 3.6751173675511213e-06,
      "loss": 0.81892806,
      "num_input_tokens_seen": 36840585,
      "step": 1734,
      "time_per_iteration": 3.8192715644836426
    },
    {
      "auxiliary_loss_clip": 0.01322181,
      "auxiliary_loss_mlp": 0.01036236,
      "balance_loss_clip": 1.05705404,
      "balance_loss_mlp": 1.02705646,
      "epoch": 0.20862141525882283,
      "flos": 20077799558400.0,
      "grad_norm": 2.1222476218305077,
      "language_loss": 0.8762387,
      "learning_rate": 3.674691651392283e-06,
      "loss": 0.89982289,
      "num_input_tokens_seen": 36858255,
      "step": 1735,
      "time_per_iteration": 2.744683265686035
    },
    {
      "auxiliary_loss_clip": 0.01328806,
      "auxiliary_loss_mlp": 0.0103925,
      "balance_loss_clip": 1.06422114,
      "balance_loss_mlp": 1.0298506,
      "epoch": 0.2087416581494619,
      "flos": 39015183237120.0,
      "grad_norm": 2.194188202382927,
      "language_loss": 0.75737637,
      "learning_rate": 3.674265681184053e-06,
      "loss": 0.78105694,
      "num_input_tokens_seen": 36881515,
      "step": 1736,
      "time_per_iteration": 2.864872932434082
    },
    {
      "auxiliary_loss_clip": 0.01320968,
      "auxiliary_loss_mlp": 0.01032233,
      "balance_loss_clip": 1.05824804,
      "balance_loss_mlp": 1.02269077,
      "epoch": 0.208861901040101,
      "flos": 26101994169600.0,
      "grad_norm": 2.5555526631259307,
      "language_loss": 0.86441398,
      "learning_rate": 3.6738394569910504e-06,
      "loss": 0.88794601,
      "num_input_tokens_seen": 36902055,
      "step": 1737,
      "time_per_iteration": 3.6651058197021484
    },
    {
      "auxiliary_loss_clip": 0.01280282,
      "auxiliary_loss_mlp": 0.01036092,
      "balance_loss_clip": 1.06819701,
      "balance_loss_mlp": 1.02647805,
      "epoch": 0.2089821439307401,
      "flos": 28398732675840.0,
      "grad_norm": 1.9846856410028844,
      "language_loss": 0.82845634,
      "learning_rate": 3.6734129788779333e-06,
      "loss": 0.85162008,
      "num_input_tokens_seen": 36921230,
      "step": 1738,
      "time_per_iteration": 2.665830612182617
    },
    {
      "auxiliary_loss_clip": 0.01293219,
      "auxiliary_loss_mlp": 0.0103398,
      "balance_loss_clip": 1.06160712,
      "balance_loss_mlp": 1.02554011,
      "epoch": 0.2091023868213792,
      "flos": 21069616872960.0,
      "grad_norm": 2.014672139354798,
      "language_loss": 0.90324295,
      "learning_rate": 3.6729862469093976e-06,
      "loss": 0.92651498,
      "num_input_tokens_seen": 36940325,
      "step": 1739,
      "time_per_iteration": 2.7729225158691406
    },
    {
      "auxiliary_loss_clip": 0.01308765,
      "auxiliary_loss_mlp": 0.01030164,
      "balance_loss_clip": 1.0561111,
      "balance_loss_mlp": 1.02099729,
      "epoch": 0.20922262971201827,
      "flos": 22455481363200.0,
      "grad_norm": 2.3547151240556716,
      "language_loss": 0.83040273,
      "learning_rate": 3.6725592611501782e-06,
      "loss": 0.85379201,
      "num_input_tokens_seen": 36959000,
      "step": 1740,
      "time_per_iteration": 2.695065975189209
    },
    {
      "auxiliary_loss_clip": 0.01273416,
      "auxiliary_loss_mlp": 0.01030318,
      "balance_loss_clip": 1.06140423,
      "balance_loss_mlp": 1.02129424,
      "epoch": 0.20934287260265738,
      "flos": 27852244179840.0,
      "grad_norm": 2.35073843927977,
      "language_loss": 0.76103365,
      "learning_rate": 3.6721320216650496e-06,
      "loss": 0.78407097,
      "num_input_tokens_seen": 36979615,
      "step": 1741,
      "time_per_iteration": 2.767033576965332
    },
    {
      "auxiliary_loss_clip": 0.01316229,
      "auxiliary_loss_mlp": 0.01028846,
      "balance_loss_clip": 1.06263232,
      "balance_loss_mlp": 1.01997137,
      "epoch": 0.20946311549329646,
      "flos": 16435309075200.0,
      "grad_norm": 2.9019244404021993,
      "language_loss": 0.8373152,
      "learning_rate": 3.6717045285188215e-06,
      "loss": 0.86076593,
      "num_input_tokens_seen": 36997310,
      "step": 1742,
      "time_per_iteration": 2.7059032917022705
    },
    {
      "auxiliary_loss_clip": 0.01409121,
      "auxiliary_loss_mlp": 0.01042718,
      "balance_loss_clip": 1.05169153,
      "balance_loss_mlp": 1.03334212,
      "epoch": 0.20958335838393555,
      "flos": 22492720788480.0,
      "grad_norm": 2.311275593555789,
      "language_loss": 0.87058568,
      "learning_rate": 3.671276781776346e-06,
      "loss": 0.89510405,
      "num_input_tokens_seen": 37015965,
      "step": 1743,
      "time_per_iteration": 2.754453420639038
    },
    {
      "auxiliary_loss_clip": 0.01300288,
      "auxiliary_loss_mlp": 0.01038066,
      "balance_loss_clip": 1.055583,
      "balance_loss_mlp": 1.02933383,
      "epoch": 0.20970360127457463,
      "flos": 25224768218880.0,
      "grad_norm": 3.253123060543526,
      "language_loss": 0.67245382,
      "learning_rate": 3.6708487815025128e-06,
      "loss": 0.69583738,
      "num_input_tokens_seen": 37036545,
      "step": 1744,
      "time_per_iteration": 2.7577831745147705
    },
    {
      "auxiliary_loss_clip": 0.01370625,
      "auxiliary_loss_mlp": 0.0103339,
      "balance_loss_clip": 1.0582931,
      "balance_loss_mlp": 1.02372193,
      "epoch": 0.20982384416521374,
      "flos": 18479164855680.0,
      "grad_norm": 2.385000999865213,
      "language_loss": 0.74370337,
      "learning_rate": 3.6704205277622463e-06,
      "loss": 0.76774347,
      "num_input_tokens_seen": 37054985,
      "step": 1745,
      "time_per_iteration": 2.7095906734466553
    },
    {
      "auxiliary_loss_clip": 0.01326247,
      "auxiliary_loss_mlp": 0.01032229,
      "balance_loss_clip": 1.05931377,
      "balance_loss_mlp": 1.02260232,
      "epoch": 0.20994408705585282,
      "flos": 25373546352000.0,
      "grad_norm": 1.8693245127873834,
      "language_loss": 0.80210394,
      "learning_rate": 3.6699920206205146e-06,
      "loss": 0.82568872,
      "num_input_tokens_seen": 37075725,
      "step": 1746,
      "time_per_iteration": 2.788283586502075
    },
    {
      "auxiliary_loss_clip": 0.01273124,
      "auxiliary_loss_mlp": 0.01038257,
      "balance_loss_clip": 1.0612787,
      "balance_loss_mlp": 1.02982283,
      "epoch": 0.2100643299464919,
      "flos": 21320955313920.0,
      "grad_norm": 3.0883696773192018,
      "language_loss": 0.81937522,
      "learning_rate": 3.669563260142321e-06,
      "loss": 0.842489,
      "num_input_tokens_seen": 37094615,
      "step": 1747,
      "time_per_iteration": 2.6362459659576416
    },
    {
      "auxiliary_loss_clip": 0.01318426,
      "auxiliary_loss_mlp": 0.01028119,
      "balance_loss_clip": 1.06335521,
      "balance_loss_mlp": 1.01949978,
      "epoch": 0.21018457283713102,
      "flos": 19354379644800.0,
      "grad_norm": 4.399219998747569,
      "language_loss": 0.84299892,
      "learning_rate": 3.6691342463927083e-06,
      "loss": 0.86646438,
      "num_input_tokens_seen": 37113610,
      "step": 1748,
      "time_per_iteration": 2.739823818206787
    },
    {
      "auxiliary_loss_clip": 0.01375696,
      "auxiliary_loss_mlp": 0.01040669,
      "balance_loss_clip": 1.05984259,
      "balance_loss_mlp": 1.03117371,
      "epoch": 0.2103048157277701,
      "flos": 28330035914880.0,
      "grad_norm": 2.022125697180014,
      "language_loss": 0.81959039,
      "learning_rate": 3.668704979436758e-06,
      "loss": 0.84375405,
      "num_input_tokens_seen": 37133705,
      "step": 1749,
      "time_per_iteration": 2.7498390674591064
    },
    {
      "auxiliary_loss_clip": 0.01319126,
      "auxiliary_loss_mlp": 0.01035904,
      "balance_loss_clip": 1.05865085,
      "balance_loss_mlp": 1.02682626,
      "epoch": 0.21042505861840918,
      "flos": 17457290835840.0,
      "grad_norm": 2.950291104536783,
      "language_loss": 0.78632414,
      "learning_rate": 3.668275459339588e-06,
      "loss": 0.80987442,
      "num_input_tokens_seen": 37152185,
      "step": 1750,
      "time_per_iteration": 2.7106075286865234
    },
    {
      "auxiliary_loss_clip": 0.01224472,
      "auxiliary_loss_mlp": 0.01031767,
      "balance_loss_clip": 1.06659126,
      "balance_loss_mlp": 1.02351809,
      "epoch": 0.21054530150904827,
      "flos": 14209817195520.0,
      "grad_norm": 1.9582133532635921,
      "language_loss": 0.80320483,
      "learning_rate": 3.667845686166358e-06,
      "loss": 0.82576722,
      "num_input_tokens_seen": 37169110,
      "step": 1751,
      "time_per_iteration": 2.623673439025879
    },
    {
      "auxiliary_loss_clip": 0.01361072,
      "auxiliary_loss_mlp": 0.01035274,
      "balance_loss_clip": 1.056499,
      "balance_loss_mlp": 1.02604079,
      "epoch": 0.21066554439968738,
      "flos": 18618210403200.0,
      "grad_norm": 5.166637474835836,
      "language_loss": 0.86176252,
      "learning_rate": 3.6674156599822634e-06,
      "loss": 0.88572598,
      "num_input_tokens_seen": 37184905,
      "step": 1752,
      "time_per_iteration": 2.6811368465423584
    },
    {
      "auxiliary_loss_clip": 0.01422186,
      "auxiliary_loss_mlp": 0.01039946,
      "balance_loss_clip": 1.0534488,
      "balance_loss_mlp": 1.03048062,
      "epoch": 0.21078578729032646,
      "flos": 23658883741440.0,
      "grad_norm": 2.527886186663741,
      "language_loss": 0.81879407,
      "learning_rate": 3.666985380852539e-06,
      "loss": 0.84341538,
      "num_input_tokens_seen": 37203910,
      "step": 1753,
      "time_per_iteration": 2.7897238731384277
    },
    {
      "auxiliary_loss_clip": 0.01322127,
      "auxiliary_loss_mlp": 0.01035222,
      "balance_loss_clip": 1.06275153,
      "balance_loss_mlp": 1.02513671,
      "epoch": 0.21090603018096554,
      "flos": 29346379240320.0,
      "grad_norm": 2.7505427351975666,
      "language_loss": 0.74410188,
      "learning_rate": 3.6665548488424576e-06,
      "loss": 0.76767534,
      "num_input_tokens_seen": 37222670,
      "step": 1754,
      "time_per_iteration": 2.7297680377960205
    },
    {
      "auxiliary_loss_clip": 0.0122621,
      "auxiliary_loss_mlp": 0.01032277,
      "balance_loss_clip": 1.06643772,
      "balance_loss_mlp": 1.02313972,
      "epoch": 0.21102627307160465,
      "flos": 23261245205760.0,
      "grad_norm": 1.7734979551404007,
      "language_loss": 0.88196695,
      "learning_rate": 3.6661240640173307e-06,
      "loss": 0.9045518,
      "num_input_tokens_seen": 37244140,
      "step": 1755,
      "time_per_iteration": 3.6571662425994873
    },
    {
      "auxiliary_loss_clip": 0.01201777,
      "auxiliary_loss_mlp": 0.01005858,
      "balance_loss_clip": 1.03724074,
      "balance_loss_mlp": 1.00436199,
      "epoch": 0.21114651596224374,
      "flos": 54633454577280.0,
      "grad_norm": 0.8568274907729799,
      "language_loss": 0.57889402,
      "learning_rate": 3.6656930264425085e-06,
      "loss": 0.60097039,
      "num_input_tokens_seen": 37308185,
      "step": 1756,
      "time_per_iteration": 3.3091979026794434
    },
    {
      "auxiliary_loss_clip": 0.01226057,
      "auxiliary_loss_mlp": 0.01030088,
      "balance_loss_clip": 1.06609023,
      "balance_loss_mlp": 1.0210458,
      "epoch": 0.21126675885288282,
      "flos": 21543314457600.0,
      "grad_norm": 2.367687893141832,
      "language_loss": 0.75797862,
      "learning_rate": 3.665261736183378e-06,
      "loss": 0.78054011,
      "num_input_tokens_seen": 37328220,
      "step": 1757,
      "time_per_iteration": 3.5261120796203613
    },
    {
      "auxiliary_loss_clip": 0.0138645,
      "auxiliary_loss_mlp": 0.01031441,
      "balance_loss_clip": 1.0698204,
      "balance_loss_mlp": 1.02267945,
      "epoch": 0.2113870017435219,
      "flos": 10961876678400.0,
      "grad_norm": 2.49880337742844,
      "language_loss": 0.88979357,
      "learning_rate": 3.664830193305366e-06,
      "loss": 0.91397244,
      "num_input_tokens_seen": 37345995,
      "step": 1758,
      "time_per_iteration": 2.8925654888153076
    },
    {
      "auxiliary_loss_clip": 0.01369972,
      "auxiliary_loss_mlp": 0.01039115,
      "balance_loss_clip": 1.05845833,
      "balance_loss_mlp": 1.03024006,
      "epoch": 0.211507244634161,
      "flos": 16653825463680.0,
      "grad_norm": 2.8754511319670812,
      "language_loss": 0.77223563,
      "learning_rate": 3.6643983978739373e-06,
      "loss": 0.79632652,
      "num_input_tokens_seen": 37362610,
      "step": 1759,
      "time_per_iteration": 2.6272199153900146
    },
    {
      "auxiliary_loss_clip": 0.01322306,
      "auxiliary_loss_mlp": 0.01039777,
      "balance_loss_clip": 1.06404722,
      "balance_loss_mlp": 1.03074741,
      "epoch": 0.2116274875248001,
      "flos": 20954091755520.0,
      "grad_norm": 1.8574461690617623,
      "language_loss": 0.82494563,
      "learning_rate": 3.663966349954596e-06,
      "loss": 0.84856641,
      "num_input_tokens_seen": 37382790,
      "step": 1760,
      "time_per_iteration": 3.5971145629882812
    },
    {
      "auxiliary_loss_clip": 0.01178462,
      "auxiliary_loss_mlp": 0.01004776,
      "balance_loss_clip": 1.034729,
      "balance_loss_mlp": 1.00329232,
      "epoch": 0.21174773041543918,
      "flos": 68196949424640.0,
      "grad_norm": 0.7875044742188703,
      "language_loss": 0.59662235,
      "learning_rate": 3.6635340496128816e-06,
      "loss": 0.61845481,
      "num_input_tokens_seen": 37439720,
      "step": 1761,
      "time_per_iteration": 3.1651768684387207
    },
    {
      "auxiliary_loss_clip": 0.01422838,
      "auxiliary_loss_mlp": 0.01035138,
      "balance_loss_clip": 1.05956697,
      "balance_loss_mlp": 1.02633429,
      "epoch": 0.2118679733060783,
      "flos": 20668315150080.0,
      "grad_norm": 1.673745513944039,
      "language_loss": 0.92819566,
      "learning_rate": 3.6631014969143747e-06,
      "loss": 0.95277542,
      "num_input_tokens_seen": 37459410,
      "step": 1762,
      "time_per_iteration": 2.8077399730682373
    },
    {
      "auxiliary_loss_clip": 0.01274521,
      "auxiliary_loss_mlp": 0.01041595,
      "balance_loss_clip": 1.06449914,
      "balance_loss_mlp": 1.03196263,
      "epoch": 0.21198821619671737,
      "flos": 23223431162880.0,
      "grad_norm": 1.8820962280805233,
      "language_loss": 0.88953674,
      "learning_rate": 3.662668691924693e-06,
      "loss": 0.91269779,
      "num_input_tokens_seen": 37480460,
      "step": 1763,
      "time_per_iteration": 3.620624303817749
    },
    {
      "auxiliary_loss_clip": 0.01375487,
      "auxiliary_loss_mlp": 0.01038924,
      "balance_loss_clip": 1.05902386,
      "balance_loss_mlp": 1.0291307,
      "epoch": 0.21210845908735645,
      "flos": 24498547044480.0,
      "grad_norm": 2.263020172934962,
      "language_loss": 0.71367347,
      "learning_rate": 3.6622356347094927e-06,
      "loss": 0.73781765,
      "num_input_tokens_seen": 37502025,
      "step": 1764,
      "time_per_iteration": 2.842888116836548
    },
    {
      "auxiliary_loss_clip": 0.01303834,
      "auxiliary_loss_mlp": 0.01041219,
      "balance_loss_clip": 1.06034708,
      "balance_loss_mlp": 1.03154516,
      "epoch": 0.21222870197799554,
      "flos": 27089789160960.0,
      "grad_norm": 2.1624631395346507,
      "language_loss": 0.78514129,
      "learning_rate": 3.6618023253344684e-06,
      "loss": 0.80859184,
      "num_input_tokens_seen": 37520885,
      "step": 1765,
      "time_per_iteration": 2.7139394283294678
    },
    {
      "auxiliary_loss_clip": 0.01275103,
      "auxiliary_loss_mlp": 0.01040112,
      "balance_loss_clip": 1.0625062,
      "balance_loss_mlp": 1.02979493,
      "epoch": 0.21234894486863465,
      "flos": 16873850223360.0,
      "grad_norm": 1.7301384679887115,
      "language_loss": 0.83933461,
      "learning_rate": 3.6613687638653527e-06,
      "loss": 0.86248678,
      "num_input_tokens_seen": 37539055,
      "step": 1766,
      "time_per_iteration": 2.7071187496185303
    },
    {
      "auxiliary_loss_clip": 0.01322323,
      "auxiliary_loss_mlp": 0.01033067,
      "balance_loss_clip": 1.06804407,
      "balance_loss_mlp": 1.02425098,
      "epoch": 0.21246918775927373,
      "flos": 23474949171840.0,
      "grad_norm": 3.004519927478719,
      "language_loss": 0.7764371,
      "learning_rate": 3.660934950367916e-06,
      "loss": 0.79999101,
      "num_input_tokens_seen": 37558300,
      "step": 1767,
      "time_per_iteration": 2.7264814376831055
    },
    {
      "auxiliary_loss_clip": 0.01276578,
      "auxiliary_loss_mlp": 0.01034168,
      "balance_loss_clip": 1.06478167,
      "balance_loss_mlp": 1.0248636,
      "epoch": 0.21258943064991281,
      "flos": 22382295402240.0,
      "grad_norm": 1.7269809657744628,
      "language_loss": 0.8302356,
      "learning_rate": 3.660500884907968e-06,
      "loss": 0.85334313,
      "num_input_tokens_seen": 37579040,
      "step": 1768,
      "time_per_iteration": 2.719905376434326
    },
    {
      "auxiliary_loss_clip": 0.01260619,
      "auxiliary_loss_mlp": 0.01004029,
      "balance_loss_clip": 1.02463698,
      "balance_loss_mlp": 1.00212145,
      "epoch": 0.21270967354055192,
      "flos": 59440168679040.0,
      "grad_norm": 0.8289654695877348,
      "language_loss": 0.59951389,
      "learning_rate": 3.660066567551356e-06,
      "loss": 0.62216038,
      "num_input_tokens_seen": 37639185,
      "step": 1769,
      "time_per_iteration": 3.2155396938323975
    },
    {
      "auxiliary_loss_clip": 0.01276763,
      "auxiliary_loss_mlp": 0.02580387,
      "balance_loss_clip": 1.06637144,
      "balance_loss_mlp": 1.00010633,
      "epoch": 0.212829916431191,
      "flos": 21544032729600.0,
      "grad_norm": 6.515510965562395,
      "language_loss": 0.84543657,
      "learning_rate": 3.6596319983639657e-06,
      "loss": 0.88400805,
      "num_input_tokens_seen": 37657765,
      "step": 1770,
      "time_per_iteration": 2.7066287994384766
    },
    {
      "auxiliary_loss_clip": 0.01371283,
      "auxiliary_loss_mlp": 0.02580169,
      "balance_loss_clip": 1.06236625,
      "balance_loss_mlp": 1.00003147,
      "epoch": 0.2129501593218301,
      "flos": 28987739896320.0,
      "grad_norm": 1.681172293775958,
      "language_loss": 0.86436832,
      "learning_rate": 3.6591971774117214e-06,
      "loss": 0.90388286,
      "num_input_tokens_seen": 37680740,
      "step": 1771,
      "time_per_iteration": 2.80973219871521
    },
    {
      "auxiliary_loss_clip": 0.01280346,
      "auxiliary_loss_mlp": 0.01033758,
      "balance_loss_clip": 1.06702852,
      "balance_loss_mlp": 1.02479959,
      "epoch": 0.2130704022124692,
      "flos": 18806993308800.0,
      "grad_norm": 2.1846579218978452,
      "language_loss": 0.80594134,
      "learning_rate": 3.6587621047605833e-06,
      "loss": 0.82908237,
      "num_input_tokens_seen": 37697910,
      "step": 1772,
      "time_per_iteration": 2.6686975955963135
    },
    {
      "auxiliary_loss_clip": 0.01274324,
      "auxiliary_loss_mlp": 0.01029977,
      "balance_loss_clip": 1.06544495,
      "balance_loss_mlp": 1.02082753,
      "epoch": 0.21319064510310828,
      "flos": 13918150759680.0,
      "grad_norm": 2.2534928364758273,
      "language_loss": 0.86625165,
      "learning_rate": 3.6583267804765542e-06,
      "loss": 0.88929462,
      "num_input_tokens_seen": 37712245,
      "step": 1773,
      "time_per_iteration": 2.6781694889068604
    },
    {
      "auxiliary_loss_clip": 0.01267512,
      "auxiliary_loss_mlp": 0.01030817,
      "balance_loss_clip": 1.06133509,
      "balance_loss_mlp": 1.02188253,
      "epoch": 0.21331088799374737,
      "flos": 20959694277120.0,
      "grad_norm": 2.3494531725510535,
      "language_loss": 0.85925657,
      "learning_rate": 3.6578912046256702e-06,
      "loss": 0.88223982,
      "num_input_tokens_seen": 37730765,
      "step": 1774,
      "time_per_iteration": 2.6669938564300537
    },
    {
      "auxiliary_loss_clip": 0.01369505,
      "auxiliary_loss_mlp": 0.0104284,
      "balance_loss_clip": 1.05793381,
      "balance_loss_mlp": 1.03374469,
      "epoch": 0.21343113088438645,
      "flos": 18624638937600.0,
      "grad_norm": 2.547695201039595,
      "language_loss": 0.76169783,
      "learning_rate": 3.6574553772740083e-06,
      "loss": 0.78582132,
      "num_input_tokens_seen": 37748695,
      "step": 1775,
      "time_per_iteration": 2.7556629180908203
    },
    {
      "auxiliary_loss_clip": 0.01177854,
      "auxiliary_loss_mlp": 0.0101472,
      "balance_loss_clip": 1.0407443,
      "balance_loss_mlp": 1.01326525,
      "epoch": 0.21355137377502556,
      "flos": 67413128791680.0,
      "grad_norm": 0.8609051581600449,
      "language_loss": 0.61823809,
      "learning_rate": 3.657019298487684e-06,
      "loss": 0.64016384,
      "num_input_tokens_seen": 37813705,
      "step": 1776,
      "time_per_iteration": 3.246567726135254
    },
    {
      "auxiliary_loss_clip": 0.0120199,
      "auxiliary_loss_mlp": 0.02581356,
      "balance_loss_clip": 1.06397784,
      "balance_loss_mlp": 1.00004864,
      "epoch": 0.21367161666566464,
      "flos": 34532095697280.0,
      "grad_norm": 2.0677815715548196,
      "language_loss": 0.83927566,
      "learning_rate": 3.6565829683328495e-06,
      "loss": 0.87710917,
      "num_input_tokens_seen": 37836330,
      "step": 1777,
      "time_per_iteration": 2.7719578742980957
    },
    {
      "auxiliary_loss_clip": 0.01278678,
      "auxiliary_loss_mlp": 0.01040513,
      "balance_loss_clip": 1.07033849,
      "balance_loss_mlp": 1.03157246,
      "epoch": 0.21379185955630373,
      "flos": 18989347680000.0,
      "grad_norm": 2.072539695510629,
      "language_loss": 0.86121464,
      "learning_rate": 3.6561463868756965e-06,
      "loss": 0.88440657,
      "num_input_tokens_seen": 37855030,
      "step": 1778,
      "time_per_iteration": 2.6473135948181152
    },
    {
      "auxiliary_loss_clip": 0.01271631,
      "auxiliary_loss_mlp": 0.01036237,
      "balance_loss_clip": 1.06641829,
      "balance_loss_mlp": 1.02670574,
      "epoch": 0.21391210244694284,
      "flos": 28218497207040.0,
      "grad_norm": 2.5916173600739283,
      "language_loss": 0.78083807,
      "learning_rate": 3.655709554182452e-06,
      "loss": 0.80391675,
      "num_input_tokens_seen": 37875370,
      "step": 1779,
      "time_per_iteration": 2.7059826850891113
    },
    {
      "auxiliary_loss_clip": 0.01277951,
      "auxiliary_loss_mlp": 0.01036481,
      "balance_loss_clip": 1.06457853,
      "balance_loss_mlp": 1.02692032,
      "epoch": 0.21403234533758192,
      "flos": 17455064192640.0,
      "grad_norm": 2.151908928040915,
      "language_loss": 0.84718746,
      "learning_rate": 3.6552724703193855e-06,
      "loss": 0.87033176,
      "num_input_tokens_seen": 37892560,
      "step": 1780,
      "time_per_iteration": 2.6796207427978516
    },
    {
      "auxiliary_loss_clip": 0.0132744,
      "auxiliary_loss_mlp": 0.01001632,
      "balance_loss_clip": 1.02319264,
      "balance_loss_mlp": 1.00009429,
      "epoch": 0.214152588228221,
      "flos": 51637606686720.0,
      "grad_norm": 0.7921044272079738,
      "language_loss": 0.55883908,
      "learning_rate": 3.654835135352801e-06,
      "loss": 0.58212984,
      "num_input_tokens_seen": 37947370,
      "step": 1781,
      "time_per_iteration": 4.1659018993377686
    },
    {
      "auxiliary_loss_clip": 0.01425297,
      "auxiliary_loss_mlp": 0.01031328,
      "balance_loss_clip": 1.05423832,
      "balance_loss_mlp": 1.02214921,
      "epoch": 0.21427283111886009,
      "flos": 19496154625920.0,
      "grad_norm": 3.9191011373234037,
      "language_loss": 0.87221694,
      "learning_rate": 3.654397549349043e-06,
      "loss": 0.89678323,
      "num_input_tokens_seen": 37964745,
      "step": 1782,
      "time_per_iteration": 2.755384922027588
    },
    {
      "auxiliary_loss_clip": 0.01318316,
      "auxiliary_loss_mlp": 0.01032081,
      "balance_loss_clip": 1.0649972,
      "balance_loss_mlp": 1.02341413,
      "epoch": 0.2143930740094992,
      "flos": 20084802710400.0,
      "grad_norm": 2.989746319018865,
      "language_loss": 0.75754344,
      "learning_rate": 3.653959712374491e-06,
      "loss": 0.7810474,
      "num_input_tokens_seen": 37982850,
      "step": 1783,
      "time_per_iteration": 3.627288818359375
    },
    {
      "auxiliary_loss_clip": 0.01373825,
      "auxiliary_loss_mlp": 0.01040906,
      "balance_loss_clip": 1.06853032,
      "balance_loss_mlp": 1.03058839,
      "epoch": 0.21451331690013828,
      "flos": 21798603394560.0,
      "grad_norm": 2.0498277978259023,
      "language_loss": 0.83139491,
      "learning_rate": 3.6535216244955663e-06,
      "loss": 0.85554218,
      "num_input_tokens_seen": 38002745,
      "step": 1784,
      "time_per_iteration": 2.7062625885009766
    },
    {
      "auxiliary_loss_clip": 0.01319471,
      "auxiliary_loss_mlp": 0.01031265,
      "balance_loss_clip": 1.06164813,
      "balance_loss_mlp": 1.02239037,
      "epoch": 0.21463355979077736,
      "flos": 32853882412800.0,
      "grad_norm": 1.8729542045869425,
      "language_loss": 0.70876324,
      "learning_rate": 3.653083285778726e-06,
      "loss": 0.7322706,
      "num_input_tokens_seen": 38024115,
      "step": 1785,
      "time_per_iteration": 2.827552556991577
    },
    {
      "auxiliary_loss_clip": 0.01279897,
      "auxiliary_loss_mlp": 0.01038334,
      "balance_loss_clip": 1.06548309,
      "balance_loss_mlp": 1.02904749,
      "epoch": 0.21475380268141647,
      "flos": 21543817248000.0,
      "grad_norm": 4.430154575242656,
      "language_loss": 0.81321549,
      "learning_rate": 3.6526446962904653e-06,
      "loss": 0.83639783,
      "num_input_tokens_seen": 38042830,
      "step": 1786,
      "time_per_iteration": 3.536332368850708
    },
    {
      "auxiliary_loss_clip": 0.01270112,
      "auxiliary_loss_mlp": 0.01036712,
      "balance_loss_clip": 1.06536269,
      "balance_loss_mlp": 1.02787876,
      "epoch": 0.21487404557205556,
      "flos": 32159082660480.0,
      "grad_norm": 1.5936039418731456,
      "language_loss": 0.74281693,
      "learning_rate": 3.652205856097318e-06,
      "loss": 0.76588523,
      "num_input_tokens_seen": 38066015,
      "step": 1787,
      "time_per_iteration": 2.7212398052215576
    },
    {
      "auxiliary_loss_clip": 0.01385551,
      "auxiliary_loss_mlp": 0.02581464,
      "balance_loss_clip": 1.06244278,
      "balance_loss_mlp": 1.00004101,
      "epoch": 0.21499428846269464,
      "flos": 12673091583360.0,
      "grad_norm": 2.096367683624753,
      "language_loss": 0.79048938,
      "learning_rate": 3.651766765265856e-06,
      "loss": 0.83015954,
      "num_input_tokens_seen": 38083025,
      "step": 1788,
      "time_per_iteration": 2.6626946926116943
    },
    {
      "auxiliary_loss_clip": 0.01317496,
      "auxiliary_loss_mlp": 0.01036262,
      "balance_loss_clip": 1.05823255,
      "balance_loss_mlp": 1.02668929,
      "epoch": 0.21511453135333372,
      "flos": 23471573293440.0,
      "grad_norm": 2.7871301142360396,
      "language_loss": 0.81330216,
      "learning_rate": 3.65132742386269e-06,
      "loss": 0.83683974,
      "num_input_tokens_seen": 38098245,
      "step": 1789,
      "time_per_iteration": 2.685608148574829
    },
    {
      "auxiliary_loss_clip": 0.01223384,
      "auxiliary_loss_mlp": 0.010395,
      "balance_loss_clip": 1.06453812,
      "balance_loss_mlp": 1.02989173,
      "epoch": 0.21523477424397283,
      "flos": 26943560893440.0,
      "grad_norm": 1.8271341856337657,
      "language_loss": 0.84700918,
      "learning_rate": 3.6508878319544656e-06,
      "loss": 0.86963803,
      "num_input_tokens_seen": 38118460,
      "step": 1790,
      "time_per_iteration": 3.5692155361175537
    },
    {
      "auxiliary_loss_clip": 0.01317693,
      "auxiliary_loss_mlp": 0.0105159,
      "balance_loss_clip": 1.06213307,
      "balance_loss_mlp": 1.04167211,
      "epoch": 0.21535501713461191,
      "flos": 18916161719040.0,
      "grad_norm": 3.496962010025046,
      "language_loss": 0.81780076,
      "learning_rate": 3.65044798960787e-06,
      "loss": 0.84149361,
      "num_input_tokens_seen": 38136800,
      "step": 1791,
      "time_per_iteration": 2.6545181274414062
    },
    {
      "auxiliary_loss_clip": 0.01361306,
      "auxiliary_loss_mlp": 0.01030397,
      "balance_loss_clip": 1.0561043,
      "balance_loss_mlp": 1.0217483,
      "epoch": 0.215475260025251,
      "flos": 17895113712000.0,
      "grad_norm": 1.9567917602834946,
      "language_loss": 0.7835598,
      "learning_rate": 3.650007896889627e-06,
      "loss": 0.80747688,
      "num_input_tokens_seen": 38155380,
      "step": 1792,
      "time_per_iteration": 2.758579969406128
    },
    {
      "auxiliary_loss_clip": 0.0122394,
      "auxiliary_loss_mlp": 0.01038767,
      "balance_loss_clip": 1.06836724,
      "balance_loss_mlp": 1.02967739,
      "epoch": 0.2155955029158901,
      "flos": 16654292340480.0,
      "grad_norm": 2.0858369676241,
      "language_loss": 0.8107267,
      "learning_rate": 3.6495675538664974e-06,
      "loss": 0.83335376,
      "num_input_tokens_seen": 38174395,
      "step": 1793,
      "time_per_iteration": 2.549421548843384
    },
    {
      "auxiliary_loss_clip": 0.01322159,
      "auxiliary_loss_mlp": 0.01033342,
      "balance_loss_clip": 1.05783868,
      "balance_loss_mlp": 1.02453816,
      "epoch": 0.2157157458065292,
      "flos": 23621213352960.0,
      "grad_norm": 1.8043421653627878,
      "language_loss": 0.82531834,
      "learning_rate": 3.649126960605282e-06,
      "loss": 0.84887332,
      "num_input_tokens_seen": 38195380,
      "step": 1794,
      "time_per_iteration": 2.7396738529205322
    },
    {
      "auxiliary_loss_clip": 0.01327114,
      "auxiliary_loss_mlp": 0.01036224,
      "balance_loss_clip": 1.06694138,
      "balance_loss_mlp": 1.02718806,
      "epoch": 0.21583598869716827,
      "flos": 22127078292480.0,
      "grad_norm": 3.8218788601248934,
      "language_loss": 0.83437389,
      "learning_rate": 3.6486861171728174e-06,
      "loss": 0.85800731,
      "num_input_tokens_seen": 38213775,
      "step": 1795,
      "time_per_iteration": 2.8259174823760986
    },
    {
      "auxiliary_loss_clip": 0.01375823,
      "auxiliary_loss_mlp": 0.01036619,
      "balance_loss_clip": 1.05755234,
      "balance_loss_mlp": 1.02748787,
      "epoch": 0.21595623158780738,
      "flos": 23441229279360.0,
      "grad_norm": 2.4148424152212438,
      "language_loss": 0.78591049,
      "learning_rate": 3.6482450236359803e-06,
      "loss": 0.81003493,
      "num_input_tokens_seen": 38235630,
      "step": 1796,
      "time_per_iteration": 2.803398609161377
    },
    {
      "auxiliary_loss_clip": 0.01271812,
      "auxiliary_loss_mlp": 0.0102835,
      "balance_loss_clip": 1.06546664,
      "balance_loss_mlp": 1.01981449,
      "epoch": 0.21607647447844647,
      "flos": 26906501036160.0,
      "grad_norm": 2.561215370409979,
      "language_loss": 0.77874839,
      "learning_rate": 3.647803680061683e-06,
      "loss": 0.80175,
      "num_input_tokens_seen": 38256045,
      "step": 1797,
      "time_per_iteration": 2.722527027130127
    },
    {
      "auxiliary_loss_clip": 0.01321405,
      "auxiliary_loss_mlp": 0.01033037,
      "balance_loss_clip": 1.06302953,
      "balance_loss_mlp": 1.02364278,
      "epoch": 0.21619671736908555,
      "flos": 14495378319360.0,
      "grad_norm": 4.040781519588101,
      "language_loss": 0.74851084,
      "learning_rate": 3.6473620865168776e-06,
      "loss": 0.77205527,
      "num_input_tokens_seen": 38272915,
      "step": 1798,
      "time_per_iteration": 2.6588895320892334
    },
    {
      "auxiliary_loss_clip": 0.01323789,
      "auxiliary_loss_mlp": 0.01037547,
      "balance_loss_clip": 1.06502819,
      "balance_loss_mlp": 1.02877951,
      "epoch": 0.21631696025972463,
      "flos": 17931096161280.0,
      "grad_norm": 1.8637174041370437,
      "language_loss": 0.81647867,
      "learning_rate": 3.646920243068554e-06,
      "loss": 0.84009206,
      "num_input_tokens_seen": 38290810,
      "step": 1799,
      "time_per_iteration": 2.698209762573242
    },
    {
      "auxiliary_loss_clip": 0.01311413,
      "auxiliary_loss_mlp": 0.01025272,
      "balance_loss_clip": 1.0587709,
      "balance_loss_mlp": 1.01662982,
      "epoch": 0.21643720315036374,
      "flos": 24462385027200.0,
      "grad_norm": 6.690382426527499,
      "language_loss": 0.74545413,
      "learning_rate": 3.6464781497837384e-06,
      "loss": 0.768821,
      "num_input_tokens_seen": 38312785,
      "step": 1800,
      "time_per_iteration": 2.7022581100463867
    },
    {
      "auxiliary_loss_clip": 0.0132416,
      "auxiliary_loss_mlp": 0.01037102,
      "balance_loss_clip": 1.05836141,
      "balance_loss_mlp": 1.02804792,
      "epoch": 0.21655744604100283,
      "flos": 28474432588800.0,
      "grad_norm": 1.8000329786039209,
      "language_loss": 0.7290144,
      "learning_rate": 3.6460358067294965e-06,
      "loss": 0.75262702,
      "num_input_tokens_seen": 38334015,
      "step": 1801,
      "time_per_iteration": 2.779560089111328
    },
    {
      "auxiliary_loss_clip": 0.01224774,
      "auxiliary_loss_mlp": 0.01031283,
      "balance_loss_clip": 1.06364369,
      "balance_loss_mlp": 1.02151382,
      "epoch": 0.2166776889316419,
      "flos": 20152960767360.0,
      "grad_norm": 2.121907900622631,
      "language_loss": 0.77874351,
      "learning_rate": 3.645593213972932e-06,
      "loss": 0.8013041,
      "num_input_tokens_seen": 38352920,
      "step": 1802,
      "time_per_iteration": 2.623281478881836
    },
    {
      "auxiliary_loss_clip": 0.01270112,
      "auxiliary_loss_mlp": 0.01044163,
      "balance_loss_clip": 1.06137514,
      "balance_loss_mlp": 1.0345602,
      "epoch": 0.21679793182228102,
      "flos": 15193482122880.0,
      "grad_norm": 2.1023726878575846,
      "language_loss": 0.80270457,
      "learning_rate": 3.6451503715811852e-06,
      "loss": 0.82584739,
      "num_input_tokens_seen": 38371230,
      "step": 1803,
      "time_per_iteration": 2.6429011821746826
    },
    {
      "auxiliary_loss_clip": 0.01321374,
      "auxiliary_loss_mlp": 0.01034021,
      "balance_loss_clip": 1.06406641,
      "balance_loss_mlp": 1.02505672,
      "epoch": 0.2169181747129201,
      "flos": 17384464010880.0,
      "grad_norm": 1.9040539132475822,
      "language_loss": 0.80028355,
      "learning_rate": 3.6447072796214345e-06,
      "loss": 0.82383746,
      "num_input_tokens_seen": 38389795,
      "step": 1804,
      "time_per_iteration": 2.6334779262542725
    },
    {
      "auxiliary_loss_clip": 0.01329649,
      "auxiliary_loss_mlp": 0.01001847,
      "balance_loss_clip": 1.02841902,
      "balance_loss_mlp": 1.00020206,
      "epoch": 0.21703841760355919,
      "flos": 58760955429120.0,
      "grad_norm": 0.9049969286103443,
      "language_loss": 0.63119602,
      "learning_rate": 3.644263938160898e-06,
      "loss": 0.65451097,
      "num_input_tokens_seen": 38445760,
      "step": 1805,
      "time_per_iteration": 3.2207343578338623
    },
    {
      "auxiliary_loss_clip": 0.01373416,
      "auxiliary_loss_mlp": 0.01040969,
      "balance_loss_clip": 1.05969465,
      "balance_loss_mlp": 1.03090167,
      "epoch": 0.21715866049419827,
      "flos": 22418457419520.0,
      "grad_norm": 2.2940225935224445,
      "language_loss": 0.71886533,
      "learning_rate": 3.6438203472668293e-06,
      "loss": 0.74300915,
      "num_input_tokens_seen": 38465405,
      "step": 1806,
      "time_per_iteration": 3.667471408843994
    },
    {
      "auxiliary_loss_clip": 0.01325371,
      "auxiliary_loss_mlp": 0.01038249,
      "balance_loss_clip": 1.06387711,
      "balance_loss_mlp": 1.0295825,
      "epoch": 0.21727890338483738,
      "flos": 17237732952960.0,
      "grad_norm": 3.761815281131092,
      "language_loss": 0.82234895,
      "learning_rate": 3.6433765070065206e-06,
      "loss": 0.84598511,
      "num_input_tokens_seen": 38483195,
      "step": 1807,
      "time_per_iteration": 2.7338051795959473
    },
    {
      "auxiliary_loss_clip": 0.01224926,
      "auxiliary_loss_mlp": 0.0103414,
      "balance_loss_clip": 1.06554794,
      "balance_loss_mlp": 1.02522922,
      "epoch": 0.21739914627547646,
      "flos": 13434792416640.0,
      "grad_norm": 3.1218814819774416,
      "language_loss": 0.87782538,
      "learning_rate": 3.6429324174473025e-06,
      "loss": 0.90041608,
      "num_input_tokens_seen": 38496735,
      "step": 1808,
      "time_per_iteration": 2.6756317615509033
    },
    {
      "auxiliary_loss_clip": 0.01273087,
      "auxiliary_loss_mlp": 0.01034979,
      "balance_loss_clip": 1.06221628,
      "balance_loss_mlp": 1.02589536,
      "epoch": 0.21751938916611555,
      "flos": 20959514709120.0,
      "grad_norm": 3.0161572708815245,
      "language_loss": 0.84781528,
      "learning_rate": 3.6424880786565425e-06,
      "loss": 0.87089598,
      "num_input_tokens_seen": 38512880,
      "step": 1809,
      "time_per_iteration": 2.6617579460144043
    },
    {
      "auxiliary_loss_clip": 0.01412873,
      "auxiliary_loss_mlp": 0.01034733,
      "balance_loss_clip": 1.05963922,
      "balance_loss_mlp": 1.02527356,
      "epoch": 0.21763963205675466,
      "flos": 27599936071680.0,
      "grad_norm": 2.8105043503279843,
      "language_loss": 0.79709983,
      "learning_rate": 3.6420434907016482e-06,
      "loss": 0.82157588,
      "num_input_tokens_seen": 38532570,
      "step": 1810,
      "time_per_iteration": 3.701144218444824
    },
    {
      "auxiliary_loss_clip": 0.01274401,
      "auxiliary_loss_mlp": 0.01034123,
      "balance_loss_clip": 1.06681943,
      "balance_loss_mlp": 1.0250628,
      "epoch": 0.21775987494739374,
      "flos": 21430411032960.0,
      "grad_norm": 3.5978555649867676,
      "language_loss": 0.81395048,
      "learning_rate": 3.6415986536500606e-06,
      "loss": 0.83703572,
      "num_input_tokens_seen": 38550900,
      "step": 1811,
      "time_per_iteration": 2.6591572761535645
    },
    {
      "auxiliary_loss_clip": 0.01412531,
      "auxiliary_loss_mlp": 0.01037765,
      "balance_loss_clip": 1.06119049,
      "balance_loss_mlp": 1.02909303,
      "epoch": 0.21788011783803282,
      "flos": 18332972501760.0,
      "grad_norm": 2.342628866402087,
      "language_loss": 0.80443585,
      "learning_rate": 3.641153567569263e-06,
      "loss": 0.82893884,
      "num_input_tokens_seen": 38569215,
      "step": 1812,
      "time_per_iteration": 3.694887161254883
    },
    {
      "auxiliary_loss_clip": 0.01266325,
      "auxiliary_loss_mlp": 0.01035373,
      "balance_loss_clip": 1.0613656,
      "balance_loss_mlp": 1.02620554,
      "epoch": 0.2180003607286719,
      "flos": 30262748037120.0,
      "grad_norm": 2.5794797298738183,
      "language_loss": 0.95532787,
      "learning_rate": 3.640708232526774e-06,
      "loss": 0.9783448,
      "num_input_tokens_seen": 38587870,
      "step": 1813,
      "time_per_iteration": 2.736747980117798
    },
    {
      "auxiliary_loss_clip": 0.0146356,
      "auxiliary_loss_mlp": 0.01040572,
      "balance_loss_clip": 1.04798591,
      "balance_loss_mlp": 1.03146994,
      "epoch": 0.21812060361931102,
      "flos": 25480272637440.0,
      "grad_norm": 2.3396340931039066,
      "language_loss": 0.78917521,
      "learning_rate": 3.6402626485901504e-06,
      "loss": 0.81421655,
      "num_input_tokens_seen": 38606965,
      "step": 1814,
      "time_per_iteration": 2.797799825668335
    },
    {
      "auxiliary_loss_clip": 0.01265921,
      "auxiliary_loss_mlp": 0.01038536,
      "balance_loss_clip": 1.06284189,
      "balance_loss_mlp": 1.0298574,
      "epoch": 0.2182408465099501,
      "flos": 21908166854400.0,
      "grad_norm": 2.3710009202224533,
      "language_loss": 0.78312647,
      "learning_rate": 3.639816815826988e-06,
      "loss": 0.806171,
      "num_input_tokens_seen": 38626290,
      "step": 1815,
      "time_per_iteration": 2.7242581844329834
    },
    {
      "auxiliary_loss_clip": 0.01317931,
      "auxiliary_loss_mlp": 0.01036168,
      "balance_loss_clip": 1.06298518,
      "balance_loss_mlp": 1.02760911,
      "epoch": 0.21836108940058918,
      "flos": 23657339456640.0,
      "grad_norm": 1.8858278050058819,
      "language_loss": 0.7815882,
      "learning_rate": 3.6393707343049176e-06,
      "loss": 0.80512917,
      "num_input_tokens_seen": 38646620,
      "step": 1816,
      "time_per_iteration": 3.6603691577911377
    },
    {
      "auxiliary_loss_clip": 0.01274608,
      "auxiliary_loss_mlp": 0.01038893,
      "balance_loss_clip": 1.0620513,
      "balance_loss_mlp": 1.02955294,
      "epoch": 0.2184813322912283,
      "flos": 24681009156480.0,
      "grad_norm": 2.5102095147729004,
      "language_loss": 0.735116,
      "learning_rate": 3.6389244040916104e-06,
      "loss": 0.75825107,
      "num_input_tokens_seen": 38665695,
      "step": 1817,
      "time_per_iteration": 2.7004237174987793
    },
    {
      "auxiliary_loss_clip": 0.01316257,
      "auxiliary_loss_mlp": 0.02579937,
      "balance_loss_clip": 1.05892324,
      "balance_loss_mlp": 1.00002372,
      "epoch": 0.21860157518186737,
      "flos": 26574650259840.0,
      "grad_norm": 2.223379214500999,
      "language_loss": 0.79566258,
      "learning_rate": 3.6384778252547747e-06,
      "loss": 0.83462453,
      "num_input_tokens_seen": 38681575,
      "step": 1818,
      "time_per_iteration": 2.7978434562683105
    },
    {
      "auxiliary_loss_clip": 0.01314885,
      "auxiliary_loss_mlp": 0.02580902,
      "balance_loss_clip": 1.0615567,
      "balance_loss_mlp": 1.00004733,
      "epoch": 0.21872181807250646,
      "flos": 20886292834560.0,
      "grad_norm": 2.285920552183603,
      "language_loss": 0.78202534,
      "learning_rate": 3.638030997862155e-06,
      "loss": 0.82098323,
      "num_input_tokens_seen": 38700510,
      "step": 1819,
      "time_per_iteration": 2.671473741531372
    },
    {
      "auxiliary_loss_clip": 0.01216162,
      "auxiliary_loss_mlp": 0.01008649,
      "balance_loss_clip": 1.03184485,
      "balance_loss_mlp": 1.00694418,
      "epoch": 0.21884206096314554,
      "flos": 61209452897280.0,
      "grad_norm": 0.7564528170685185,
      "language_loss": 0.59432018,
      "learning_rate": 3.6375839219815356e-06,
      "loss": 0.61656833,
      "num_input_tokens_seen": 38758310,
      "step": 1820,
      "time_per_iteration": 3.1525533199310303
    },
    {
      "auxiliary_loss_clip": 0.01222009,
      "auxiliary_loss_mlp": 0.01032649,
      "balance_loss_clip": 1.06364989,
      "balance_loss_mlp": 1.02298164,
      "epoch": 0.21896230385378465,
      "flos": 23473835850240.0,
      "grad_norm": 2.0161550397297696,
      "language_loss": 0.82601744,
      "learning_rate": 3.6371365976807375e-06,
      "loss": 0.84856403,
      "num_input_tokens_seen": 38778705,
      "step": 1821,
      "time_per_iteration": 2.6714985370635986
    },
    {
      "auxiliary_loss_clip": 0.01416506,
      "auxiliary_loss_mlp": 0.01041943,
      "balance_loss_clip": 1.06102824,
      "balance_loss_mlp": 1.0325433,
      "epoch": 0.21908254674442373,
      "flos": 25081915829760.0,
      "grad_norm": 1.6779311393265888,
      "language_loss": 0.83610618,
      "learning_rate": 3.6366890250276185e-06,
      "loss": 0.86069071,
      "num_input_tokens_seen": 38799660,
      "step": 1822,
      "time_per_iteration": 2.777338743209839
    },
    {
      "auxiliary_loss_clip": 0.01221653,
      "auxiliary_loss_mlp": 0.01033858,
      "balance_loss_clip": 1.06375146,
      "balance_loss_mlp": 1.02465546,
      "epoch": 0.21920278963506282,
      "flos": 23513768795520.0,
      "grad_norm": 2.387016392903714,
      "language_loss": 0.89780402,
      "learning_rate": 3.6362412040900764e-06,
      "loss": 0.92035913,
      "num_input_tokens_seen": 38819450,
      "step": 1823,
      "time_per_iteration": 2.6886239051818848
    },
    {
      "auxiliary_loss_clip": 0.0127732,
      "auxiliary_loss_mlp": 0.01036313,
      "balance_loss_clip": 1.06378341,
      "balance_loss_mlp": 1.02676427,
      "epoch": 0.21932303252570193,
      "flos": 29242238734080.0,
      "grad_norm": 1.864789241091381,
      "language_loss": 0.80332911,
      "learning_rate": 3.635793134936044e-06,
      "loss": 0.82646543,
      "num_input_tokens_seen": 38840460,
      "step": 1824,
      "time_per_iteration": 2.703505039215088
    },
    {
      "auxiliary_loss_clip": 0.01268723,
      "auxiliary_loss_mlp": 0.01034521,
      "balance_loss_clip": 1.06213379,
      "balance_loss_mlp": 1.02583683,
      "epoch": 0.219443275416341,
      "flos": 20806857907200.0,
      "grad_norm": 1.8725702383660616,
      "language_loss": 0.72995639,
      "learning_rate": 3.635344817633494e-06,
      "loss": 0.75298882,
      "num_input_tokens_seen": 38859775,
      "step": 1825,
      "time_per_iteration": 2.6812338829040527
    },
    {
      "auxiliary_loss_clip": 0.01269274,
      "auxiliary_loss_mlp": 0.01029747,
      "balance_loss_clip": 1.06391335,
      "balance_loss_mlp": 1.02088928,
      "epoch": 0.2195635183069801,
      "flos": 14501555458560.0,
      "grad_norm": 2.0191732481649467,
      "language_loss": 0.75573444,
      "learning_rate": 3.634896252250436e-06,
      "loss": 0.77872461,
      "num_input_tokens_seen": 38876540,
      "step": 1826,
      "time_per_iteration": 2.7083253860473633
    },
    {
      "auxiliary_loss_clip": 0.01225352,
      "auxiliary_loss_mlp": 0.01037689,
      "balance_loss_clip": 1.06650662,
      "balance_loss_mlp": 1.0278002,
      "epoch": 0.2196837611976192,
      "flos": 24243473589120.0,
      "grad_norm": 1.7698607574185932,
      "language_loss": 0.82270294,
      "learning_rate": 3.6344474388549157e-06,
      "loss": 0.8453334,
      "num_input_tokens_seen": 38896195,
      "step": 1827,
      "time_per_iteration": 2.6956613063812256
    },
    {
      "auxiliary_loss_clip": 0.01274283,
      "auxiliary_loss_mlp": 0.01030954,
      "balance_loss_clip": 1.06459033,
      "balance_loss_mlp": 1.02148294,
      "epoch": 0.2198040040882583,
      "flos": 18074523168000.0,
      "grad_norm": 2.3342130709076074,
      "language_loss": 0.80255377,
      "learning_rate": 3.6339983775150183e-06,
      "loss": 0.82560611,
      "num_input_tokens_seen": 38912755,
      "step": 1828,
      "time_per_iteration": 2.625328302383423
    },
    {
      "auxiliary_loss_clip": 0.01269461,
      "auxiliary_loss_mlp": 0.01029305,
      "balance_loss_clip": 1.06438029,
      "balance_loss_mlp": 1.02045393,
      "epoch": 0.21992424697889737,
      "flos": 17784185535360.0,
      "grad_norm": 2.7970604649599644,
      "language_loss": 0.84078288,
      "learning_rate": 3.6335490682988664e-06,
      "loss": 0.8637706,
      "num_input_tokens_seen": 38928365,
      "step": 1829,
      "time_per_iteration": 2.6860878467559814
    },
    {
      "auxiliary_loss_clip": 0.01469119,
      "auxiliary_loss_mlp": 0.01037673,
      "balance_loss_clip": 1.05670798,
      "balance_loss_mlp": 1.02802289,
      "epoch": 0.22004448986953645,
      "flos": 17638495971840.0,
      "grad_norm": 2.9914652258743852,
      "language_loss": 0.82970464,
      "learning_rate": 3.63309951127462e-06,
      "loss": 0.85477257,
      "num_input_tokens_seen": 38945275,
      "step": 1830,
      "time_per_iteration": 2.7630081176757812
    },
    {
      "auxiliary_loss_clip": 0.01371227,
      "auxiliary_loss_mlp": 0.01033191,
      "balance_loss_clip": 1.0628891,
      "balance_loss_mlp": 1.02401781,
      "epoch": 0.22016473276017556,
      "flos": 22275533203200.0,
      "grad_norm": 2.1403053935800065,
      "language_loss": 0.75179166,
      "learning_rate": 3.6326497065104757e-06,
      "loss": 0.77583587,
      "num_input_tokens_seen": 38965740,
      "step": 1831,
      "time_per_iteration": 2.7554643154144287
    },
    {
      "auxiliary_loss_clip": 0.01279803,
      "auxiliary_loss_mlp": 0.01035738,
      "balance_loss_clip": 1.06540871,
      "balance_loss_mlp": 1.02693462,
      "epoch": 0.22028497565081465,
      "flos": 25556259859200.0,
      "grad_norm": 2.3441794960813147,
      "language_loss": 0.78161693,
      "learning_rate": 3.6321996540746697e-06,
      "loss": 0.80477238,
      "num_input_tokens_seen": 38984815,
      "step": 1832,
      "time_per_iteration": 3.67094349861145
    },
    {
      "auxiliary_loss_clip": 0.01370664,
      "auxiliary_loss_mlp": 0.01027243,
      "balance_loss_clip": 1.06219411,
      "balance_loss_mlp": 1.0185349,
      "epoch": 0.22040521854145373,
      "flos": 36247332925440.0,
      "grad_norm": 1.8522488254730218,
      "language_loss": 0.80658162,
      "learning_rate": 3.6317493540354733e-06,
      "loss": 0.83056068,
      "num_input_tokens_seen": 39008230,
      "step": 1833,
      "time_per_iteration": 2.827941656112671
    },
    {
      "auxiliary_loss_clip": 0.01269258,
      "auxiliary_loss_mlp": 0.0103112,
      "balance_loss_clip": 1.06122804,
      "balance_loss_mlp": 1.02160096,
      "epoch": 0.22052546143209284,
      "flos": 11838420270720.0,
      "grad_norm": 3.363740250902558,
      "language_loss": 0.7705164,
      "learning_rate": 3.6312988064611976e-06,
      "loss": 0.79352015,
      "num_input_tokens_seen": 39026540,
      "step": 1834,
      "time_per_iteration": 2.70715069770813
    },
    {
      "auxiliary_loss_clip": 0.01369863,
      "auxiliary_loss_mlp": 0.01029318,
      "balance_loss_clip": 1.05544591,
      "balance_loss_mlp": 1.02060425,
      "epoch": 0.22064570432273192,
      "flos": 24209250906240.0,
      "grad_norm": 1.7508629114528453,
      "language_loss": 0.81353599,
      "learning_rate": 3.6308480114201896e-06,
      "loss": 0.83752781,
      "num_input_tokens_seen": 39048460,
      "step": 1835,
      "time_per_iteration": 3.5960114002227783
    },
    {
      "auxiliary_loss_clip": 0.01226517,
      "auxiliary_loss_mlp": 0.01045347,
      "balance_loss_clip": 1.06802845,
      "balance_loss_mlp": 1.0354526,
      "epoch": 0.220765947213371,
      "flos": 17931347556480.0,
      "grad_norm": 1.9589345506656368,
      "language_loss": 0.76258695,
      "learning_rate": 3.630396968980835e-06,
      "loss": 0.78530562,
      "num_input_tokens_seen": 39066335,
      "step": 1836,
      "time_per_iteration": 2.6338889598846436
    },
    {
      "auxiliary_loss_clip": 0.01322303,
      "auxiliary_loss_mlp": 0.0103645,
      "balance_loss_clip": 1.06048012,
      "balance_loss_mlp": 1.02738392,
      "epoch": 0.2208861901040101,
      "flos": 26757040544640.0,
      "grad_norm": 3.20191906944085,
      "language_loss": 0.83128929,
      "learning_rate": 3.6299456792115575e-06,
      "loss": 0.85487688,
      "num_input_tokens_seen": 39087590,
      "step": 1837,
      "time_per_iteration": 2.823408365249634
    },
    {
      "auxiliary_loss_clip": 0.01426012,
      "auxiliary_loss_mlp": 0.0103245,
      "balance_loss_clip": 1.04846036,
      "balance_loss_mlp": 1.02321124,
      "epoch": 0.2210064329946492,
      "flos": 17817977255040.0,
      "grad_norm": 2.7530071133951695,
      "language_loss": 0.80904317,
      "learning_rate": 3.629494142180815e-06,
      "loss": 0.83362782,
      "num_input_tokens_seen": 39106335,
      "step": 1838,
      "time_per_iteration": 3.7246651649475098
    },
    {
      "auxiliary_loss_clip": 0.01223889,
      "auxiliary_loss_mlp": 0.01032127,
      "balance_loss_clip": 1.06633472,
      "balance_loss_mlp": 1.02261984,
      "epoch": 0.22112667588528828,
      "flos": 17967401832960.0,
      "grad_norm": 2.1913291336338014,
      "language_loss": 0.85129225,
      "learning_rate": 3.6290423579571075e-06,
      "loss": 0.87385237,
      "num_input_tokens_seen": 39122875,
      "step": 1839,
      "time_per_iteration": 2.6867613792419434
    },
    {
      "auxiliary_loss_clip": 0.0126897,
      "auxiliary_loss_mlp": 0.0103319,
      "balance_loss_clip": 1.06325495,
      "balance_loss_mlp": 1.02430928,
      "epoch": 0.22124691877592736,
      "flos": 18369206346240.0,
      "grad_norm": 1.696262651886668,
      "language_loss": 0.80342662,
      "learning_rate": 3.6285903266089694e-06,
      "loss": 0.82644826,
      "num_input_tokens_seen": 39142150,
      "step": 1840,
      "time_per_iteration": 2.6539618968963623
    },
    {
      "auxiliary_loss_clip": 0.01325917,
      "auxiliary_loss_mlp": 0.01037076,
      "balance_loss_clip": 1.06407619,
      "balance_loss_mlp": 1.027843,
      "epoch": 0.22136716166656648,
      "flos": 20813286441600.0,
      "grad_norm": 1.962005155059415,
      "language_loss": 0.77145129,
      "learning_rate": 3.628138048204974e-06,
      "loss": 0.79508126,
      "num_input_tokens_seen": 39162835,
      "step": 1841,
      "time_per_iteration": 2.684108257293701
    },
    {
      "auxiliary_loss_clip": 0.01418558,
      "auxiliary_loss_mlp": 0.01033344,
      "balance_loss_clip": 1.05913782,
      "balance_loss_mlp": 1.02258527,
      "epoch": 0.22148740455720556,
      "flos": 17675699483520.0,
      "grad_norm": 3.6966878753413464,
      "language_loss": 0.76377082,
      "learning_rate": 3.6276855228137304e-06,
      "loss": 0.78828979,
      "num_input_tokens_seen": 39181040,
      "step": 1842,
      "time_per_iteration": 3.703460693359375
    },
    {
      "auxiliary_loss_clip": 0.01222313,
      "auxiliary_loss_mlp": 0.02581577,
      "balance_loss_clip": 1.06520689,
      "balance_loss_mlp": 0.99998903,
      "epoch": 0.22160764744784464,
      "flos": 21726710323200.0,
      "grad_norm": 2.286886766137286,
      "language_loss": 0.81913334,
      "learning_rate": 3.6272327505038874e-06,
      "loss": 0.85717225,
      "num_input_tokens_seen": 39197505,
      "step": 1843,
      "time_per_iteration": 2.640068531036377
    },
    {
      "auxiliary_loss_clip": 0.01347407,
      "auxiliary_loss_mlp": 0.0102958,
      "balance_loss_clip": 1.0582788,
      "balance_loss_mlp": 1.02045465,
      "epoch": 0.22172789033848372,
      "flos": 23764712186880.0,
      "grad_norm": 3.0225514934789377,
      "language_loss": 0.78342295,
      "learning_rate": 3.626779731344131e-06,
      "loss": 0.80719286,
      "num_input_tokens_seen": 39217295,
      "step": 1844,
      "time_per_iteration": 2.803255558013916
    },
    {
      "auxiliary_loss_clip": 0.01218379,
      "auxiliary_loss_mlp": 0.01033887,
      "balance_loss_clip": 1.06287909,
      "balance_loss_mlp": 1.02512467,
      "epoch": 0.22184813322912283,
      "flos": 16982300361600.0,
      "grad_norm": 2.16371920075099,
      "language_loss": 0.85393989,
      "learning_rate": 3.6263264654031814e-06,
      "loss": 0.87646258,
      "num_input_tokens_seen": 39234195,
      "step": 1845,
      "time_per_iteration": 2.5834646224975586
    },
    {
      "auxiliary_loss_clip": 0.01268515,
      "auxiliary_loss_mlp": 0.01000562,
      "balance_loss_clip": 1.02855587,
      "balance_loss_mlp": 0.99905455,
      "epoch": 0.22196837611976192,
      "flos": 61823740314240.0,
      "grad_norm": 0.6932171599192984,
      "language_loss": 0.59074008,
      "learning_rate": 3.6258729527498008e-06,
      "loss": 0.61343086,
      "num_input_tokens_seen": 39295040,
      "step": 1846,
      "time_per_iteration": 3.2964723110198975
    },
    {
      "auxiliary_loss_clip": 0.01327028,
      "auxiliary_loss_mlp": 0.01034212,
      "balance_loss_clip": 1.06356025,
      "balance_loss_mlp": 1.02463913,
      "epoch": 0.222088619010401,
      "flos": 25558019625600.0,
      "grad_norm": 2.5143963170106027,
      "language_loss": 0.6476115,
      "learning_rate": 3.6254191934527854e-06,
      "loss": 0.67122394,
      "num_input_tokens_seen": 39314395,
      "step": 1847,
      "time_per_iteration": 2.7232534885406494
    },
    {
      "auxiliary_loss_clip": 0.01379144,
      "auxiliary_loss_mlp": 0.01034664,
      "balance_loss_clip": 1.06766224,
      "balance_loss_mlp": 1.02559829,
      "epoch": 0.2222088619010401,
      "flos": 19318612677120.0,
      "grad_norm": 2.5656449517258086,
      "language_loss": 0.64896661,
      "learning_rate": 3.6249651875809715e-06,
      "loss": 0.67310464,
      "num_input_tokens_seen": 39334275,
      "step": 1848,
      "time_per_iteration": 2.7850353717803955
    },
    {
      "auxiliary_loss_clip": 0.01316132,
      "auxiliary_loss_mlp": 0.01033504,
      "balance_loss_clip": 1.06382978,
      "balance_loss_mlp": 1.02548718,
      "epoch": 0.2223291047916792,
      "flos": 19099342103040.0,
      "grad_norm": 2.3796383790200477,
      "language_loss": 0.8915965,
      "learning_rate": 3.62451093520323e-06,
      "loss": 0.91509283,
      "num_input_tokens_seen": 39352180,
      "step": 1849,
      "time_per_iteration": 2.6955721378326416
    },
    {
      "auxiliary_loss_clip": 0.01420753,
      "auxiliary_loss_mlp": 0.01037859,
      "balance_loss_clip": 1.05746925,
      "balance_loss_mlp": 1.02904391,
      "epoch": 0.22244934768231828,
      "flos": 20850418126080.0,
      "grad_norm": 3.026755904331516,
      "language_loss": 0.90478647,
      "learning_rate": 3.6240564363884714e-06,
      "loss": 0.92937261,
      "num_input_tokens_seen": 39372125,
      "step": 1850,
      "time_per_iteration": 2.731663942337036
    },
    {
      "auxiliary_loss_clip": 0.01279598,
      "auxiliary_loss_mlp": 0.01035225,
      "balance_loss_clip": 1.06434083,
      "balance_loss_mlp": 1.0252353,
      "epoch": 0.2225695905729574,
      "flos": 15632921111040.0,
      "grad_norm": 2.556576510665642,
      "language_loss": 0.70534098,
      "learning_rate": 3.623601691205643e-06,
      "loss": 0.72848928,
      "num_input_tokens_seen": 39391200,
      "step": 1851,
      "time_per_iteration": 2.677340030670166
    },
    {
      "auxiliary_loss_clip": 0.01270425,
      "auxiliary_loss_mlp": 0.01033046,
      "balance_loss_clip": 1.06165981,
      "balance_loss_mlp": 1.02423072,
      "epoch": 0.22268983346359647,
      "flos": 25373582265600.0,
      "grad_norm": 2.738875591268448,
      "language_loss": 0.81741476,
      "learning_rate": 3.623146699723729e-06,
      "loss": 0.84044945,
      "num_input_tokens_seen": 39410660,
      "step": 1852,
      "time_per_iteration": 2.656926155090332
    },
    {
      "auxiliary_loss_clip": 0.0131753,
      "auxiliary_loss_mlp": 0.01039355,
      "balance_loss_clip": 1.06518447,
      "balance_loss_mlp": 1.03019929,
      "epoch": 0.22281007635423555,
      "flos": 13261452359040.0,
      "grad_norm": 1.8387116214174863,
      "language_loss": 0.77685106,
      "learning_rate": 3.6226914620117507e-06,
      "loss": 0.80041993,
      "num_input_tokens_seen": 39429280,
      "step": 1853,
      "time_per_iteration": 2.6997530460357666
    },
    {
      "auxiliary_loss_clip": 0.01288104,
      "auxiliary_loss_mlp": 0.01029154,
      "balance_loss_clip": 1.05038667,
      "balance_loss_mlp": 1.02056468,
      "epoch": 0.22293031924487464,
      "flos": 15340536403200.0,
      "grad_norm": 2.4870314769279425,
      "language_loss": 0.8087306,
      "learning_rate": 3.622235978138768e-06,
      "loss": 0.83190322,
      "num_input_tokens_seen": 39446905,
      "step": 1854,
      "time_per_iteration": 2.731600284576416
    },
    {
      "auxiliary_loss_clip": 0.01271268,
      "auxiliary_loss_mlp": 0.01031622,
      "balance_loss_clip": 1.06600428,
      "balance_loss_mlp": 1.02274144,
      "epoch": 0.22305056213551375,
      "flos": 22564649773440.0,
      "grad_norm": 1.8860092181182444,
      "language_loss": 0.8156479,
      "learning_rate": 3.621780248173877e-06,
      "loss": 0.83867681,
      "num_input_tokens_seen": 39465105,
      "step": 1855,
      "time_per_iteration": 2.7144312858581543
    },
    {
      "auxiliary_loss_clip": 0.01171349,
      "auxiliary_loss_mlp": 0.01000355,
      "balance_loss_clip": 1.03004885,
      "balance_loss_mlp": 0.99881679,
      "epoch": 0.22317080502615283,
      "flos": 64880419887360.0,
      "grad_norm": 0.8316270463594376,
      "language_loss": 0.61016047,
      "learning_rate": 3.6213242721862125e-06,
      "loss": 0.63187754,
      "num_input_tokens_seen": 39523560,
      "step": 1856,
      "time_per_iteration": 3.2136545181274414
    },
    {
      "auxiliary_loss_clip": 0.01317244,
      "auxiliary_loss_mlp": 0.01036428,
      "balance_loss_clip": 1.06292582,
      "balance_loss_mlp": 1.02690911,
      "epoch": 0.2232910479167919,
      "flos": 25775997310080.0,
      "grad_norm": 1.699694044303913,
      "language_loss": 0.75320333,
      "learning_rate": 3.620868050244945e-06,
      "loss": 0.77674007,
      "num_input_tokens_seen": 39544040,
      "step": 1857,
      "time_per_iteration": 2.7618048191070557
    },
    {
      "auxiliary_loss_clip": 0.01321739,
      "auxiliary_loss_mlp": 0.01032043,
      "balance_loss_clip": 1.06208038,
      "balance_loss_mlp": 1.02226138,
      "epoch": 0.22341129080743102,
      "flos": 23251799928960.0,
      "grad_norm": 1.9547039535291633,
      "language_loss": 0.77669793,
      "learning_rate": 3.6204115824192817e-06,
      "loss": 0.80023575,
      "num_input_tokens_seen": 39561515,
      "step": 1858,
      "time_per_iteration": 3.595716953277588
    },
    {
      "auxiliary_loss_clip": 0.01313205,
      "auxiliary_loss_mlp": 0.01031488,
      "balance_loss_clip": 1.05587721,
      "balance_loss_mlp": 1.02262509,
      "epoch": 0.2235315336980701,
      "flos": 21214552250880.0,
      "grad_norm": 2.5876381746343995,
      "language_loss": 0.76561755,
      "learning_rate": 3.619954868778471e-06,
      "loss": 0.78906453,
      "num_input_tokens_seen": 39578210,
      "step": 1859,
      "time_per_iteration": 2.6792781352996826
    },
    {
      "auxiliary_loss_clip": 0.01319507,
      "auxiliary_loss_mlp": 0.01043254,
      "balance_loss_clip": 1.05817735,
      "balance_loss_mlp": 1.03422976,
      "epoch": 0.2236517765887092,
      "flos": 19901945548800.0,
      "grad_norm": 1.908814384156504,
      "language_loss": 0.83008993,
      "learning_rate": 3.6194979093917944e-06,
      "loss": 0.85371745,
      "num_input_tokens_seen": 39597625,
      "step": 1860,
      "time_per_iteration": 3.656123399734497
    },
    {
      "auxiliary_loss_clip": 0.01316235,
      "auxiliary_loss_mlp": 0.01031795,
      "balance_loss_clip": 1.05902982,
      "balance_loss_mlp": 1.02294922,
      "epoch": 0.22377201947934827,
      "flos": 23214847812480.0,
      "grad_norm": 2.888739745645323,
      "language_loss": 0.86972153,
      "learning_rate": 3.6190407043285724e-06,
      "loss": 0.89320183,
      "num_input_tokens_seen": 39615360,
      "step": 1861,
      "time_per_iteration": 2.675870656967163
    },
    {
      "auxiliary_loss_clip": 0.01227475,
      "auxiliary_loss_mlp": 0.01035521,
      "balance_loss_clip": 1.06779146,
      "balance_loss_mlp": 1.02618694,
      "epoch": 0.22389226236998738,
      "flos": 26794244056320.0,
      "grad_norm": 1.9823733642289447,
      "language_loss": 0.75730932,
      "learning_rate": 3.618583253658163e-06,
      "loss": 0.77993935,
      "num_input_tokens_seen": 39635460,
      "step": 1862,
      "time_per_iteration": 2.6585733890533447
    },
    {
      "auxiliary_loss_clip": 0.01423567,
      "auxiliary_loss_mlp": 0.02580019,
      "balance_loss_clip": 1.06051755,
      "balance_loss_mlp": 0.99990785,
      "epoch": 0.22401250526062647,
      "flos": 24170359455360.0,
      "grad_norm": 8.498673202335121,
      "language_loss": 0.86513865,
      "learning_rate": 3.618125557449961e-06,
      "loss": 0.90517449,
      "num_input_tokens_seen": 39653515,
      "step": 1863,
      "time_per_iteration": 3.713798761367798
    },
    {
      "auxiliary_loss_clip": 0.012702,
      "auxiliary_loss_mlp": 0.01028086,
      "balance_loss_clip": 1.0642978,
      "balance_loss_mlp": 1.01983643,
      "epoch": 0.22413274815126555,
      "flos": 16759761649920.0,
      "grad_norm": 2.091846162269351,
      "language_loss": 0.83193767,
      "learning_rate": 3.6176676157733983e-06,
      "loss": 0.85492051,
      "num_input_tokens_seen": 39668525,
      "step": 1864,
      "time_per_iteration": 2.6170475482940674
    },
    {
      "auxiliary_loss_clip": 0.01367701,
      "auxiliary_loss_mlp": 0.01035713,
      "balance_loss_clip": 1.0594089,
      "balance_loss_mlp": 1.02679074,
      "epoch": 0.22425299104190466,
      "flos": 21360205900800.0,
      "grad_norm": 2.763568460897376,
      "language_loss": 0.76400989,
      "learning_rate": 3.6172094286979443e-06,
      "loss": 0.7880441,
      "num_input_tokens_seen": 39685895,
      "step": 1865,
      "time_per_iteration": 2.7104973793029785
    },
    {
      "auxiliary_loss_clip": 0.01323299,
      "auxiliary_loss_mlp": 0.0102961,
      "balance_loss_clip": 1.06223083,
      "balance_loss_mlp": 1.02079415,
      "epoch": 0.22437323393254374,
      "flos": 32165547108480.0,
      "grad_norm": 1.80730714345949,
      "language_loss": 0.81481832,
      "learning_rate": 3.6167509962931064e-06,
      "loss": 0.83834743,
      "num_input_tokens_seen": 39711595,
      "step": 1866,
      "time_per_iteration": 2.7885870933532715
    },
    {
      "auxiliary_loss_clip": 0.01426716,
      "auxiliary_loss_mlp": 0.01032979,
      "balance_loss_clip": 1.06423235,
      "balance_loss_mlp": 1.02391338,
      "epoch": 0.22449347682318282,
      "flos": 18002809664640.0,
      "grad_norm": 4.971600565844564,
      "language_loss": 0.77124548,
      "learning_rate": 3.6162923186284276e-06,
      "loss": 0.79584241,
      "num_input_tokens_seen": 39727555,
      "step": 1867,
      "time_per_iteration": 3.689216375350952
    },
    {
      "auxiliary_loss_clip": 0.01319084,
      "auxiliary_loss_mlp": 0.01032336,
      "balance_loss_clip": 1.05751073,
      "balance_loss_mlp": 1.02381861,
      "epoch": 0.2246137197138219,
      "flos": 18697286194560.0,
      "grad_norm": 2.608587945673555,
      "language_loss": 0.85960579,
      "learning_rate": 3.6158333957734888e-06,
      "loss": 0.88312,
      "num_input_tokens_seen": 39746145,
      "step": 1868,
      "time_per_iteration": 2.6634373664855957
    },
    {
      "auxiliary_loss_clip": 0.01374505,
      "auxiliary_loss_mlp": 0.010315,
      "balance_loss_clip": 1.05964208,
      "balance_loss_mlp": 1.02297103,
      "epoch": 0.22473396260446102,
      "flos": 15590653781760.0,
      "grad_norm": 2.169609644411522,
      "language_loss": 0.83120674,
      "learning_rate": 3.6153742277979088e-06,
      "loss": 0.85526681,
      "num_input_tokens_seen": 39763575,
      "step": 1869,
      "time_per_iteration": 2.689760446548462
    },
    {
      "auxiliary_loss_clip": 0.01316822,
      "auxiliary_loss_mlp": 0.01037052,
      "balance_loss_clip": 1.05990291,
      "balance_loss_mlp": 1.02871335,
      "epoch": 0.2248542054951001,
      "flos": 14465501182080.0,
      "grad_norm": 2.4737512980458116,
      "language_loss": 0.78222299,
      "learning_rate": 3.6149148147713434e-06,
      "loss": 0.80576175,
      "num_input_tokens_seen": 39781810,
      "step": 1870,
      "time_per_iteration": 2.6986377239227295
    },
    {
      "auxiliary_loss_clip": 0.01280478,
      "auxiliary_loss_mlp": 0.01040224,
      "balance_loss_clip": 1.07044697,
      "balance_loss_mlp": 1.0320642,
      "epoch": 0.22497444838573918,
      "flos": 19243882431360.0,
      "grad_norm": 2.960107293850705,
      "language_loss": 0.86944038,
      "learning_rate": 3.614455156763484e-06,
      "loss": 0.89264739,
      "num_input_tokens_seen": 39800115,
      "step": 1871,
      "time_per_iteration": 2.6707797050476074
    },
    {
      "auxiliary_loss_clip": 0.01418687,
      "auxiliary_loss_mlp": 0.01029862,
      "balance_loss_clip": 1.05591345,
      "balance_loss_mlp": 1.0213027,
      "epoch": 0.2250946912763783,
      "flos": 16910299549440.0,
      "grad_norm": 2.224286838864007,
      "language_loss": 0.7122336,
      "learning_rate": 3.613995253844061e-06,
      "loss": 0.73671907,
      "num_input_tokens_seen": 39817795,
      "step": 1872,
      "time_per_iteration": 2.8312413692474365
    },
    {
      "auxiliary_loss_clip": 0.01275853,
      "auxiliary_loss_mlp": 0.01039833,
      "balance_loss_clip": 1.0692122,
      "balance_loss_mlp": 1.03011799,
      "epoch": 0.22521493416701738,
      "flos": 24681368292480.0,
      "grad_norm": 9.120987066373251,
      "language_loss": 0.80703247,
      "learning_rate": 3.6135351060828414e-06,
      "loss": 0.83018923,
      "num_input_tokens_seen": 39838270,
      "step": 1873,
      "time_per_iteration": 2.7239787578582764
    },
    {
      "auxiliary_loss_clip": 0.01228214,
      "auxiliary_loss_mlp": 0.01034711,
      "balance_loss_clip": 1.06929827,
      "balance_loss_mlp": 1.02521658,
      "epoch": 0.22533517705765646,
      "flos": 17821963664640.0,
      "grad_norm": 1.996554383918932,
      "language_loss": 0.69767499,
      "learning_rate": 3.6130747135496285e-06,
      "loss": 0.72030425,
      "num_input_tokens_seen": 39857270,
      "step": 1874,
      "time_per_iteration": 2.6796834468841553
    },
    {
      "auxiliary_loss_clip": 0.01221459,
      "auxiliary_loss_mlp": 0.01035503,
      "balance_loss_clip": 1.0647819,
      "balance_loss_mlp": 1.02675319,
      "epoch": 0.22545541994829554,
      "flos": 33691390899840.0,
      "grad_norm": 2.079552866384218,
      "language_loss": 0.66292602,
      "learning_rate": 3.6126140763142646e-06,
      "loss": 0.68549562,
      "num_input_tokens_seen": 39882300,
      "step": 1875,
      "time_per_iteration": 2.739823579788208
    },
    {
      "auxiliary_loss_clip": 0.01223835,
      "auxiliary_loss_mlp": 0.01038969,
      "balance_loss_clip": 1.06625664,
      "balance_loss_mlp": 1.03024292,
      "epoch": 0.22557566283893465,
      "flos": 19171594310400.0,
      "grad_norm": 7.872005608953601,
      "language_loss": 0.85812747,
      "learning_rate": 3.6121531944466275e-06,
      "loss": 0.88075542,
      "num_input_tokens_seen": 39899625,
      "step": 1876,
      "time_per_iteration": 2.6168830394744873
    },
    {
      "auxiliary_loss_clip": 0.01268277,
      "auxiliary_loss_mlp": 0.0102474,
      "balance_loss_clip": 1.06307912,
      "balance_loss_mlp": 1.01590037,
      "epoch": 0.22569590572957374,
      "flos": 20773281669120.0,
      "grad_norm": 2.5878705896651417,
      "language_loss": 0.78105211,
      "learning_rate": 3.611692068016633e-06,
      "loss": 0.80398232,
      "num_input_tokens_seen": 39915955,
      "step": 1877,
      "time_per_iteration": 2.7410264015197754
    },
    {
      "auxiliary_loss_clip": 0.01372222,
      "auxiliary_loss_mlp": 0.01041258,
      "balance_loss_clip": 1.05798399,
      "balance_loss_mlp": 1.03114319,
      "epoch": 0.22581614862021282,
      "flos": 18442715529600.0,
      "grad_norm": 2.3765302702046314,
      "language_loss": 0.75533044,
      "learning_rate": 3.611230697094233e-06,
      "loss": 0.77946526,
      "num_input_tokens_seen": 39932655,
      "step": 1878,
      "time_per_iteration": 2.707064628601074
    },
    {
      "auxiliary_loss_clip": 0.01325473,
      "auxiliary_loss_mlp": 0.01033619,
      "balance_loss_clip": 1.06293571,
      "balance_loss_mlp": 1.0247972,
      "epoch": 0.22593639151085193,
      "flos": 20048389297920.0,
      "grad_norm": 2.192801456587915,
      "language_loss": 0.87583548,
      "learning_rate": 3.6107690817494173e-06,
      "loss": 0.89942634,
      "num_input_tokens_seen": 39952875,
      "step": 1879,
      "time_per_iteration": 2.7143213748931885
    },
    {
      "auxiliary_loss_clip": 0.01409743,
      "auxiliary_loss_mlp": 0.01034254,
      "balance_loss_clip": 1.05304372,
      "balance_loss_mlp": 1.02531898,
      "epoch": 0.226056634401491,
      "flos": 13115116350720.0,
      "grad_norm": 3.3640416079368523,
      "language_loss": 0.71066928,
      "learning_rate": 3.6103072220522117e-06,
      "loss": 0.73510927,
      "num_input_tokens_seen": 39968405,
      "step": 1880,
      "time_per_iteration": 2.757096529006958
    },
    {
      "auxiliary_loss_clip": 0.01372335,
      "auxiliary_loss_mlp": 0.01032565,
      "balance_loss_clip": 1.05819595,
      "balance_loss_mlp": 1.02436352,
      "epoch": 0.2261768772921301,
      "flos": 18988378012800.0,
      "grad_norm": 2.496030038775358,
      "language_loss": 0.9186008,
      "learning_rate": 3.609845118072682e-06,
      "loss": 0.94264978,
      "num_input_tokens_seen": 39987075,
      "step": 1881,
      "time_per_iteration": 2.7222440242767334
    },
    {
      "auxiliary_loss_clip": 0.01275011,
      "auxiliary_loss_mlp": 0.02577282,
      "balance_loss_clip": 1.06329823,
      "balance_loss_mlp": 1.00004017,
      "epoch": 0.2262971201827692,
      "flos": 19974054101760.0,
      "grad_norm": 3.765272620739279,
      "language_loss": 0.79860616,
      "learning_rate": 3.6093827698809276e-06,
      "loss": 0.83712912,
      "num_input_tokens_seen": 40006175,
      "step": 1882,
      "time_per_iteration": 2.667694330215454
    },
    {
      "auxiliary_loss_clip": 0.0126834,
      "auxiliary_loss_mlp": 0.01037119,
      "balance_loss_clip": 1.05982089,
      "balance_loss_mlp": 1.02872109,
      "epoch": 0.2264173630734083,
      "flos": 16654543735680.0,
      "grad_norm": 2.247304075911097,
      "language_loss": 0.84504592,
      "learning_rate": 3.6089201775470864e-06,
      "loss": 0.86810052,
      "num_input_tokens_seen": 40021630,
      "step": 1883,
      "time_per_iteration": 3.621570587158203
    },
    {
      "auxiliary_loss_clip": 0.01357114,
      "auxiliary_loss_mlp": 0.01029989,
      "balance_loss_clip": 1.05574155,
      "balance_loss_mlp": 1.02106047,
      "epoch": 0.22653760596404737,
      "flos": 24389809597440.0,
      "grad_norm": 1.4723351130395,
      "language_loss": 0.77404165,
      "learning_rate": 3.6084573411413334e-06,
      "loss": 0.7979126,
      "num_input_tokens_seen": 40041025,
      "step": 1884,
      "time_per_iteration": 2.779315710067749
    },
    {
      "auxiliary_loss_clip": 0.01366764,
      "auxiliary_loss_mlp": 0.01031629,
      "balance_loss_clip": 1.05836153,
      "balance_loss_mlp": 1.02199674,
      "epoch": 0.22665784885468646,
      "flos": 18332541538560.0,
      "grad_norm": 1.9399862537407804,
      "language_loss": 0.8090809,
      "learning_rate": 3.607994260733881e-06,
      "loss": 0.83306485,
      "num_input_tokens_seen": 40060265,
      "step": 1885,
      "time_per_iteration": 2.6707701683044434
    },
    {
      "auxiliary_loss_clip": 0.01258592,
      "auxiliary_loss_mlp": 0.01028746,
      "balance_loss_clip": 1.05752087,
      "balance_loss_mlp": 1.02106321,
      "epoch": 0.22677809174532557,
      "flos": 24058102475520.0,
      "grad_norm": 1.6824783798474794,
      "language_loss": 0.74632347,
      "learning_rate": 3.6075309363949776e-06,
      "loss": 0.76919687,
      "num_input_tokens_seen": 40079435,
      "step": 1886,
      "time_per_iteration": 3.563429594039917
    },
    {
      "auxiliary_loss_clip": 0.01221824,
      "auxiliary_loss_mlp": 0.01034796,
      "balance_loss_clip": 1.06466126,
      "balance_loss_mlp": 1.02586675,
      "epoch": 0.22689833463596465,
      "flos": 20374242503040.0,
      "grad_norm": 1.9379266225566665,
      "language_loss": 0.81295007,
      "learning_rate": 3.6070673681949094e-06,
      "loss": 0.83551633,
      "num_input_tokens_seen": 40097800,
      "step": 1887,
      "time_per_iteration": 2.6333870887756348
    },
    {
      "auxiliary_loss_clip": 0.01320123,
      "auxiliary_loss_mlp": 0.02572672,
      "balance_loss_clip": 1.06190348,
      "balance_loss_mlp": 1.00002217,
      "epoch": 0.22701857752660373,
      "flos": 30120398438400.0,
      "grad_norm": 1.7518672529011046,
      "language_loss": 0.81588238,
      "learning_rate": 3.606603556203999e-06,
      "loss": 0.85481036,
      "num_input_tokens_seen": 40122745,
      "step": 1888,
      "time_per_iteration": 2.794067144393921
    },
    {
      "auxiliary_loss_clip": 0.01268978,
      "auxiliary_loss_mlp": 0.01036988,
      "balance_loss_clip": 1.06021714,
      "balance_loss_mlp": 1.02866077,
      "epoch": 0.22713882041724284,
      "flos": 22492182084480.0,
      "grad_norm": 2.1923081740506305,
      "language_loss": 0.83778906,
      "learning_rate": 3.6061395004926066e-06,
      "loss": 0.86084872,
      "num_input_tokens_seen": 40141680,
      "step": 1889,
      "time_per_iteration": 3.576988935470581
    },
    {
      "auxiliary_loss_clip": 0.01219981,
      "auxiliary_loss_mlp": 0.01032394,
      "balance_loss_clip": 1.06331861,
      "balance_loss_mlp": 1.02312541,
      "epoch": 0.22725906330788193,
      "flos": 20521548178560.0,
      "grad_norm": 2.8146734521628973,
      "language_loss": 0.84979779,
      "learning_rate": 3.605675201131129e-06,
      "loss": 0.87232149,
      "num_input_tokens_seen": 40160140,
      "step": 1890,
      "time_per_iteration": 2.58349347114563
    },
    {
      "auxiliary_loss_clip": 0.01196077,
      "auxiliary_loss_mlp": 0.01027653,
      "balance_loss_clip": 1.06601608,
      "balance_loss_mlp": 1.01912928,
      "epoch": 0.227379306198521,
      "flos": 18989922297600.0,
      "grad_norm": 2.1206896244923934,
      "language_loss": 0.7986542,
      "learning_rate": 3.60521065819e-06,
      "loss": 0.8208915,
      "num_input_tokens_seen": 40177450,
      "step": 1891,
      "time_per_iteration": 2.6721315383911133
    },
    {
      "auxiliary_loss_clip": 0.01319885,
      "auxiliary_loss_mlp": 0.01038562,
      "balance_loss_clip": 1.05863976,
      "balance_loss_mlp": 1.02986002,
      "epoch": 0.2274995490891601,
      "flos": 21798351999360.0,
      "grad_norm": 1.866319222661721,
      "language_loss": 0.87831426,
      "learning_rate": 3.60474587173969e-06,
      "loss": 0.90189874,
      "num_input_tokens_seen": 40195935,
      "step": 1892,
      "time_per_iteration": 2.716698169708252
    },
    {
      "auxiliary_loss_clip": 0.01267578,
      "auxiliary_loss_mlp": 0.01035471,
      "balance_loss_clip": 1.06370807,
      "balance_loss_mlp": 1.0267868,
      "epoch": 0.2276197919797992,
      "flos": 19058654972160.0,
      "grad_norm": 2.7984184007321597,
      "language_loss": 0.84179306,
      "learning_rate": 3.6042808418507084e-06,
      "loss": 0.86482358,
      "num_input_tokens_seen": 40213620,
      "step": 1893,
      "time_per_iteration": 3.5499744415283203
    },
    {
      "auxiliary_loss_clip": 0.01271383,
      "auxiliary_loss_mlp": 0.01031346,
      "balance_loss_clip": 1.06540823,
      "balance_loss_mlp": 1.02219701,
      "epoch": 0.22774003487043828,
      "flos": 18806777827200.0,
      "grad_norm": 2.3763668305747547,
      "language_loss": 0.77170593,
      "learning_rate": 3.6038155685935976e-06,
      "loss": 0.79473323,
      "num_input_tokens_seen": 40230190,
      "step": 1894,
      "time_per_iteration": 2.6338815689086914
    },
    {
      "auxiliary_loss_clip": 0.01268701,
      "auxiliary_loss_mlp": 0.01034074,
      "balance_loss_clip": 1.06223714,
      "balance_loss_mlp": 1.02463269,
      "epoch": 0.22786027776107737,
      "flos": 23002544476800.0,
      "grad_norm": 2.6789459480266427,
      "language_loss": 0.70868111,
      "learning_rate": 3.6033500520389404e-06,
      "loss": 0.73170882,
      "num_input_tokens_seen": 40246860,
      "step": 1895,
      "time_per_iteration": 2.806927442550659
    },
    {
      "auxiliary_loss_clip": 0.01326426,
      "auxiliary_loss_mlp": 0.01008635,
      "balance_loss_clip": 1.02681661,
      "balance_loss_mlp": 1.00710928,
      "epoch": 0.22798052065171648,
      "flos": 66706872600960.0,
      "grad_norm": 0.7982207930301609,
      "language_loss": 0.64735079,
      "learning_rate": 3.6028842922573553e-06,
      "loss": 0.67070138,
      "num_input_tokens_seen": 40311005,
      "step": 1896,
      "time_per_iteration": 3.383547782897949
    },
    {
      "auxiliary_loss_clip": 0.01282561,
      "auxiliary_loss_mlp": 0.02525571,
      "balance_loss_clip": 1.02764165,
      "balance_loss_mlp": 0.9996807,
      "epoch": 0.22810076354235556,
      "flos": 62080896758400.0,
      "grad_norm": 0.8534720171281481,
      "language_loss": 0.62867379,
      "learning_rate": 3.602418289319497e-06,
      "loss": 0.6667552,
      "num_input_tokens_seen": 40369560,
      "step": 1897,
      "time_per_iteration": 3.2421741485595703
    },
    {
      "auxiliary_loss_clip": 0.01417034,
      "auxiliary_loss_mlp": 0.010335,
      "balance_loss_clip": 1.05528259,
      "balance_loss_mlp": 1.02472591,
      "epoch": 0.22822100643299464,
      "flos": 23876358635520.0,
      "grad_norm": 2.272271658331969,
      "language_loss": 0.7371304,
      "learning_rate": 3.601952043296059e-06,
      "loss": 0.76163572,
      "num_input_tokens_seen": 40389555,
      "step": 1898,
      "time_per_iteration": 2.806100606918335
    },
    {
      "auxiliary_loss_clip": 0.01243993,
      "auxiliary_loss_mlp": 0.01033927,
      "balance_loss_clip": 1.05951512,
      "balance_loss_mlp": 1.02483153,
      "epoch": 0.22834124932363373,
      "flos": 20991331180800.0,
      "grad_norm": 2.018886016256577,
      "language_loss": 0.80679357,
      "learning_rate": 3.6014855542577696e-06,
      "loss": 0.8295728,
      "num_input_tokens_seen": 40406765,
      "step": 1899,
      "time_per_iteration": 2.653918743133545
    },
    {
      "auxiliary_loss_clip": 0.01320573,
      "auxiliary_loss_mlp": 0.01039071,
      "balance_loss_clip": 1.06154346,
      "balance_loss_mlp": 1.02979076,
      "epoch": 0.22846149221427284,
      "flos": 24901572620160.0,
      "grad_norm": 4.293655842748259,
      "language_loss": 0.84517992,
      "learning_rate": 3.6010188222753943e-06,
      "loss": 0.86877632,
      "num_input_tokens_seen": 40427535,
      "step": 1900,
      "time_per_iteration": 2.763356924057007
    },
    {
      "auxiliary_loss_clip": 0.01222633,
      "auxiliary_loss_mlp": 0.01004954,
      "balance_loss_clip": 1.03365898,
      "balance_loss_mlp": 1.00349987,
      "epoch": 0.22858173510491192,
      "flos": 56132294319360.0,
      "grad_norm": 0.8972490597652837,
      "language_loss": 0.64139819,
      "learning_rate": 3.6005518474197372e-06,
      "loss": 0.66367412,
      "num_input_tokens_seen": 40479580,
      "step": 1901,
      "time_per_iteration": 3.1565401554107666
    },
    {
      "auxiliary_loss_clip": 0.01269615,
      "auxiliary_loss_mlp": 0.01038613,
      "balance_loss_clip": 1.06427217,
      "balance_loss_mlp": 1.02885604,
      "epoch": 0.228701977995551,
      "flos": 24170826332160.0,
      "grad_norm": 2.7686144028701083,
      "language_loss": 0.78464198,
      "learning_rate": 3.6000846297616373e-06,
      "loss": 0.80772424,
      "num_input_tokens_seen": 40497880,
      "step": 1902,
      "time_per_iteration": 2.704374074935913
    },
    {
      "auxiliary_loss_clip": 0.01224562,
      "auxiliary_loss_mlp": 0.01039351,
      "balance_loss_clip": 1.06699908,
      "balance_loss_mlp": 1.02970707,
      "epoch": 0.22882222088619011,
      "flos": 21387892308480.0,
      "grad_norm": 2.367438596881115,
      "language_loss": 0.72540689,
      "learning_rate": 3.5996171693719717e-06,
      "loss": 0.74804604,
      "num_input_tokens_seen": 40513975,
      "step": 1903,
      "time_per_iteration": 2.6481473445892334
    },
    {
      "auxiliary_loss_clip": 0.01167727,
      "auxiliary_loss_mlp": 0.01006199,
      "balance_loss_clip": 1.02929807,
      "balance_loss_mlp": 1.00469673,
      "epoch": 0.2289424637768292,
      "flos": 64589615377920.0,
      "grad_norm": 0.8432532237632567,
      "language_loss": 0.64802945,
      "learning_rate": 3.5991494663216528e-06,
      "loss": 0.66976869,
      "num_input_tokens_seen": 40576960,
      "step": 1904,
      "time_per_iteration": 3.2695555686950684
    },
    {
      "auxiliary_loss_clip": 0.01222487,
      "auxiliary_loss_mlp": 0.01040668,
      "balance_loss_clip": 1.06617641,
      "balance_loss_mlp": 1.03197217,
      "epoch": 0.22906270666746828,
      "flos": 22163419877760.0,
      "grad_norm": 1.8875704472887151,
      "language_loss": 0.87689722,
      "learning_rate": 3.5986815206816314e-06,
      "loss": 0.89952874,
      "num_input_tokens_seen": 40595780,
      "step": 1905,
      "time_per_iteration": 2.6459643840789795
    },
    {
      "auxiliary_loss_clip": 0.01218542,
      "auxiliary_loss_mlp": 0.01029711,
      "balance_loss_clip": 1.06281257,
      "balance_loss_mlp": 1.02094293,
      "epoch": 0.2291829495581074,
      "flos": 25772334122880.0,
      "grad_norm": 1.7442029136648958,
      "language_loss": 0.74900228,
      "learning_rate": 3.598213332522895e-06,
      "loss": 0.77148473,
      "num_input_tokens_seen": 40615810,
      "step": 1906,
      "time_per_iteration": 2.6363868713378906
    },
    {
      "auxiliary_loss_clip": 0.01267545,
      "auxiliary_loss_mlp": 0.01033822,
      "balance_loss_clip": 1.06110835,
      "balance_loss_mlp": 1.02516115,
      "epoch": 0.22930319244874647,
      "flos": 31172760126720.0,
      "grad_norm": 1.905596044706425,
      "language_loss": 0.77201402,
      "learning_rate": 3.597744901916466e-06,
      "loss": 0.79502767,
      "num_input_tokens_seen": 40637095,
      "step": 1907,
      "time_per_iteration": 2.7319064140319824
    },
    {
      "auxiliary_loss_clip": 0.01221767,
      "auxiliary_loss_mlp": 0.01038801,
      "balance_loss_clip": 1.06146657,
      "balance_loss_mlp": 1.02899063,
      "epoch": 0.22942343533938556,
      "flos": 23254098399360.0,
      "grad_norm": 3.6923526987657236,
      "language_loss": 0.77040458,
      "learning_rate": 3.5972762289334058e-06,
      "loss": 0.79301023,
      "num_input_tokens_seen": 40656725,
      "step": 1908,
      "time_per_iteration": 2.6155734062194824
    },
    {
      "auxiliary_loss_clip": 0.01457842,
      "auxiliary_loss_mlp": 0.0103585,
      "balance_loss_clip": 1.05332804,
      "balance_loss_mlp": 1.02666473,
      "epoch": 0.22954367823002464,
      "flos": 14610903436800.0,
      "grad_norm": 2.8626261397450246,
      "language_loss": 0.85359251,
      "learning_rate": 3.5968073136448116e-06,
      "loss": 0.87852943,
      "num_input_tokens_seen": 40674745,
      "step": 1909,
      "time_per_iteration": 2.819629430770874
    },
    {
      "auxiliary_loss_clip": 0.0127405,
      "auxiliary_loss_mlp": 0.01035006,
      "balance_loss_clip": 1.06183195,
      "balance_loss_mlp": 1.02608967,
      "epoch": 0.22966392112066375,
      "flos": 16763604405120.0,
      "grad_norm": 1.6503690366170085,
      "language_loss": 0.91530478,
      "learning_rate": 3.596338156121818e-06,
      "loss": 0.93839526,
      "num_input_tokens_seen": 40693630,
      "step": 1910,
      "time_per_iteration": 3.6550562381744385
    },
    {
      "auxiliary_loss_clip": 0.01214533,
      "auxiliary_loss_mlp": 0.01011812,
      "balance_loss_clip": 1.02730894,
      "balance_loss_mlp": 1.01028621,
      "epoch": 0.22978416401130283,
      "flos": 67474247783040.0,
      "grad_norm": 0.7400251792914777,
      "language_loss": 0.59343737,
      "learning_rate": 3.595868756435595e-06,
      "loss": 0.61570084,
      "num_input_tokens_seen": 40761310,
      "step": 1911,
      "time_per_iteration": 3.3865480422973633
    },
    {
      "auxiliary_loss_clip": 0.01369881,
      "auxiliary_loss_mlp": 0.01038607,
      "balance_loss_clip": 1.06033862,
      "balance_loss_mlp": 1.0293982,
      "epoch": 0.22990440690194192,
      "flos": 19865137086720.0,
      "grad_norm": 2.454089329907637,
      "language_loss": 0.80313575,
      "learning_rate": 3.5953991146573504e-06,
      "loss": 0.82722062,
      "num_input_tokens_seen": 40779955,
      "step": 1912,
      "time_per_iteration": 2.7250075340270996
    },
    {
      "auxiliary_loss_clip": 0.01270642,
      "auxiliary_loss_mlp": 0.01034813,
      "balance_loss_clip": 1.05858624,
      "balance_loss_mlp": 1.02463865,
      "epoch": 0.23002464979258103,
      "flos": 13289246507520.0,
      "grad_norm": 2.630055236627394,
      "language_loss": 0.8331399,
      "learning_rate": 3.5949292308583294e-06,
      "loss": 0.8561945,
      "num_input_tokens_seen": 40793200,
      "step": 1913,
      "time_per_iteration": 3.4754066467285156
    },
    {
      "auxiliary_loss_clip": 0.01222972,
      "auxiliary_loss_mlp": 0.01035487,
      "balance_loss_clip": 1.06720245,
      "balance_loss_mlp": 1.02627814,
      "epoch": 0.2301448926832201,
      "flos": 22163779013760.0,
      "grad_norm": 2.758846829476453,
      "language_loss": 0.80932254,
      "learning_rate": 3.594459105109811e-06,
      "loss": 0.83190715,
      "num_input_tokens_seen": 40812380,
      "step": 1914,
      "time_per_iteration": 2.609114646911621
    },
    {
      "auxiliary_loss_clip": 0.01270105,
      "auxiliary_loss_mlp": 0.01034203,
      "balance_loss_clip": 1.06234014,
      "balance_loss_mlp": 1.02557206,
      "epoch": 0.2302651355738592,
      "flos": 20704477167360.0,
      "grad_norm": 2.4629271227468457,
      "language_loss": 0.81603265,
      "learning_rate": 3.593988737483115e-06,
      "loss": 0.83907574,
      "num_input_tokens_seen": 40832320,
      "step": 1915,
      "time_per_iteration": 3.6285440921783447
    },
    {
      "auxiliary_loss_clip": 0.01323002,
      "auxiliary_loss_mlp": 0.01031478,
      "balance_loss_clip": 1.06445265,
      "balance_loss_mlp": 1.02241254,
      "epoch": 0.23038537846449827,
      "flos": 18588943797120.0,
      "grad_norm": 2.0594492691988058,
      "language_loss": 0.78599274,
      "learning_rate": 3.5935181280495947e-06,
      "loss": 0.80953753,
      "num_input_tokens_seen": 40850900,
      "step": 1916,
      "time_per_iteration": 2.6871848106384277
    },
    {
      "auxiliary_loss_clip": 0.01208686,
      "auxiliary_loss_mlp": 0.00997473,
      "balance_loss_clip": 1.02456427,
      "balance_loss_mlp": 0.99604207,
      "epoch": 0.23050562135513739,
      "flos": 64224260190720.0,
      "grad_norm": 0.805585579546953,
      "language_loss": 0.54264432,
      "learning_rate": 3.5930472768806412e-06,
      "loss": 0.56470591,
      "num_input_tokens_seen": 40909570,
      "step": 1917,
      "time_per_iteration": 3.2540714740753174
    },
    {
      "auxiliary_loss_clip": 0.01220396,
      "auxiliary_loss_mlp": 0.01032388,
      "balance_loss_clip": 1.06514513,
      "balance_loss_mlp": 1.02279162,
      "epoch": 0.23062586424577647,
      "flos": 17313396952320.0,
      "grad_norm": 2.071968279026329,
      "language_loss": 0.7712605,
      "learning_rate": 3.5925761840476826e-06,
      "loss": 0.79378837,
      "num_input_tokens_seen": 40928180,
      "step": 1918,
      "time_per_iteration": 2.632484197616577
    },
    {
      "auxiliary_loss_clip": 0.01315792,
      "auxiliary_loss_mlp": 0.01031829,
      "balance_loss_clip": 1.06249249,
      "balance_loss_mlp": 1.02349651,
      "epoch": 0.23074610713641555,
      "flos": 27855979194240.0,
      "grad_norm": 2.0121463543404734,
      "language_loss": 0.81477618,
      "learning_rate": 3.592104849622183e-06,
      "loss": 0.83825231,
      "num_input_tokens_seen": 40950435,
      "step": 1919,
      "time_per_iteration": 3.6134703159332275
    },
    {
      "auxiliary_loss_clip": 0.01412288,
      "auxiliary_loss_mlp": 0.01027936,
      "balance_loss_clip": 1.05698705,
      "balance_loss_mlp": 1.01848853,
      "epoch": 0.23086635002705466,
      "flos": 28841798937600.0,
      "grad_norm": 1.4669158912866949,
      "language_loss": 0.73385346,
      "learning_rate": 3.591633273675644e-06,
      "loss": 0.75825572,
      "num_input_tokens_seen": 40972670,
      "step": 1920,
      "time_per_iteration": 2.78517746925354
    },
    {
      "auxiliary_loss_clip": 0.01265082,
      "auxiliary_loss_mlp": 0.01011973,
      "balance_loss_clip": 1.03677034,
      "balance_loss_mlp": 1.01042295,
      "epoch": 0.23098659291769374,
      "flos": 62923681566720.0,
      "grad_norm": 0.9079384273471515,
      "language_loss": 0.58161056,
      "learning_rate": 3.591161456279602e-06,
      "loss": 0.60438108,
      "num_input_tokens_seen": 41018215,
      "step": 1921,
      "time_per_iteration": 3.0940945148468018
    },
    {
      "auxiliary_loss_clip": 0.01322985,
      "auxiliary_loss_mlp": 0.01032734,
      "balance_loss_clip": 1.0601691,
      "balance_loss_mlp": 1.02340031,
      "epoch": 0.23110683580833283,
      "flos": 23476816679040.0,
      "grad_norm": 1.649944732958213,
      "language_loss": 0.80628037,
      "learning_rate": 3.590689397505633e-06,
      "loss": 0.82983756,
      "num_input_tokens_seen": 41039125,
      "step": 1922,
      "time_per_iteration": 2.7187061309814453
    },
    {
      "auxiliary_loss_clip": 0.01217488,
      "auxiliary_loss_mlp": 0.01026193,
      "balance_loss_clip": 1.06358457,
      "balance_loss_mlp": 1.01824772,
      "epoch": 0.2312270786989719,
      "flos": 27271066124160.0,
      "grad_norm": 1.7213519601113307,
      "language_loss": 0.86919832,
      "learning_rate": 3.590217097425347e-06,
      "loss": 0.89163512,
      "num_input_tokens_seen": 41059025,
      "step": 1923,
      "time_per_iteration": 2.7208385467529297
    },
    {
      "auxiliary_loss_clip": 0.01220258,
      "auxiliary_loss_mlp": 0.01034243,
      "balance_loss_clip": 1.06361926,
      "balance_loss_mlp": 1.02447414,
      "epoch": 0.23134732158961102,
      "flos": 13261344618240.0,
      "grad_norm": 3.071569785922368,
      "language_loss": 0.71327645,
      "learning_rate": 3.589744556110391e-06,
      "loss": 0.73582149,
      "num_input_tokens_seen": 41077015,
      "step": 1924,
      "time_per_iteration": 2.5946052074432373
    },
    {
      "auxiliary_loss_clip": 0.01317001,
      "auxiliary_loss_mlp": 0.01036347,
      "balance_loss_clip": 1.05834901,
      "balance_loss_mlp": 1.02670336,
      "epoch": 0.2314675644802501,
      "flos": 36977648250240.0,
      "grad_norm": 2.1133939564369784,
      "language_loss": 0.84680337,
      "learning_rate": 3.58927177363245e-06,
      "loss": 0.87033689,
      "num_input_tokens_seen": 41099840,
      "step": 1925,
      "time_per_iteration": 2.8584423065185547
    },
    {
      "auxiliary_loss_clip": 0.01284719,
      "auxiliary_loss_mlp": 0.01032632,
      "balance_loss_clip": 1.05574751,
      "balance_loss_mlp": 1.0230118,
      "epoch": 0.2315878073708892,
      "flos": 23842207779840.0,
      "grad_norm": 3.158271724916513,
      "language_loss": 0.7298606,
      "learning_rate": 3.5887987500632447e-06,
      "loss": 0.75303411,
      "num_input_tokens_seen": 41117845,
      "step": 1926,
      "time_per_iteration": 2.7693490982055664
    },
    {
      "auxiliary_loss_clip": 0.01372117,
      "auxiliary_loss_mlp": 0.01035433,
      "balance_loss_clip": 1.05811548,
      "balance_loss_mlp": 1.02774405,
      "epoch": 0.2317080502615283,
      "flos": 23039424766080.0,
      "grad_norm": 1.7248048440630026,
      "language_loss": 0.84626186,
      "learning_rate": 3.5883254854745325e-06,
      "loss": 0.87033737,
      "num_input_tokens_seen": 41136235,
      "step": 1927,
      "time_per_iteration": 2.7901201248168945
    },
    {
      "auxiliary_loss_clip": 0.01272883,
      "auxiliary_loss_mlp": 0.01035085,
      "balance_loss_clip": 1.06170475,
      "balance_loss_mlp": 1.02512527,
      "epoch": 0.23182829315216738,
      "flos": 11254656435840.0,
      "grad_norm": 2.30014954904484,
      "language_loss": 0.75218666,
      "learning_rate": 3.587851979938107e-06,
      "loss": 0.77526629,
      "num_input_tokens_seen": 41153125,
      "step": 1928,
      "time_per_iteration": 2.6089015007019043
    },
    {
      "auxiliary_loss_clip": 0.01267038,
      "auxiliary_loss_mlp": 0.01040678,
      "balance_loss_clip": 1.06240034,
      "balance_loss_mlp": 1.03110528,
      "epoch": 0.23194853604280646,
      "flos": 19828939155840.0,
      "grad_norm": 1.9445959771673278,
      "language_loss": 0.7733947,
      "learning_rate": 3.5873782335257985e-06,
      "loss": 0.79647183,
      "num_input_tokens_seen": 41171290,
      "step": 1929,
      "time_per_iteration": 2.6734976768493652
    },
    {
      "auxiliary_loss_clip": 0.01367525,
      "auxiliary_loss_mlp": 0.01031845,
      "balance_loss_clip": 1.06249404,
      "balance_loss_mlp": 1.02263606,
      "epoch": 0.23206877893344555,
      "flos": 15305020830720.0,
      "grad_norm": 2.0128483730448776,
      "language_loss": 0.78547311,
      "learning_rate": 3.5869042463094744e-06,
      "loss": 0.80946678,
      "num_input_tokens_seen": 41189005,
      "step": 1930,
      "time_per_iteration": 2.699697256088257
    },
    {
      "auxiliary_loss_clip": 0.01405283,
      "auxiliary_loss_mlp": 0.01036645,
      "balance_loss_clip": 1.05299592,
      "balance_loss_mlp": 1.0274601,
      "epoch": 0.23218902182408466,
      "flos": 22711488572160.0,
      "grad_norm": 2.00933247131545,
      "language_loss": 0.77351147,
      "learning_rate": 3.586430018361038e-06,
      "loss": 0.79793078,
      "num_input_tokens_seen": 41208775,
      "step": 1931,
      "time_per_iteration": 2.8040659427642822
    },
    {
      "auxiliary_loss_clip": 0.01311829,
      "auxiliary_loss_mlp": 0.01034633,
      "balance_loss_clip": 1.05833638,
      "balance_loss_mlp": 1.02535844,
      "epoch": 0.23230926471472374,
      "flos": 22710734386560.0,
      "grad_norm": 2.6798154731059967,
      "language_loss": 0.76530367,
      "learning_rate": 3.5859555497524283e-06,
      "loss": 0.78876829,
      "num_input_tokens_seen": 41226010,
      "step": 1932,
      "time_per_iteration": 2.7362613677978516
    },
    {
      "auxiliary_loss_clip": 0.01268825,
      "auxiliary_loss_mlp": 0.0103274,
      "balance_loss_clip": 1.06345952,
      "balance_loss_mlp": 1.02341795,
      "epoch": 0.23242950760536282,
      "flos": 20375499479040.0,
      "grad_norm": 2.134678306216583,
      "language_loss": 0.9222852,
      "learning_rate": 3.5854808405556237e-06,
      "loss": 0.94530094,
      "num_input_tokens_seen": 41245245,
      "step": 1933,
      "time_per_iteration": 2.64353346824646
    },
    {
      "auxiliary_loss_clip": 0.01367803,
      "auxiliary_loss_mlp": 0.0103399,
      "balance_loss_clip": 1.05652928,
      "balance_loss_mlp": 1.02586603,
      "epoch": 0.23254975049600193,
      "flos": 16908324301440.0,
      "grad_norm": 4.110975466271458,
      "language_loss": 0.75843859,
      "learning_rate": 3.5850058908426355e-06,
      "loss": 0.78245658,
      "num_input_tokens_seen": 41263795,
      "step": 1934,
      "time_per_iteration": 2.7357468605041504
    },
    {
      "auxiliary_loss_clip": 0.01319328,
      "auxiliary_loss_mlp": 0.01038723,
      "balance_loss_clip": 1.0573138,
      "balance_loss_mlp": 1.02886438,
      "epoch": 0.23266999338664102,
      "flos": 23294821443840.0,
      "grad_norm": 3.106339726433307,
      "language_loss": 0.85599416,
      "learning_rate": 3.584530700685514e-06,
      "loss": 0.87957466,
      "num_input_tokens_seen": 41284055,
      "step": 1935,
      "time_per_iteration": 3.5847437381744385
    },
    {
      "auxiliary_loss_clip": 0.01313342,
      "auxiliary_loss_mlp": 0.0103007,
      "balance_loss_clip": 1.06197453,
      "balance_loss_mlp": 1.02154064,
      "epoch": 0.2327902362772801,
      "flos": 19569987031680.0,
      "grad_norm": 2.773967743169451,
      "language_loss": 0.88719773,
      "learning_rate": 3.5840552701563448e-06,
      "loss": 0.91063178,
      "num_input_tokens_seen": 41300255,
      "step": 1936,
      "time_per_iteration": 2.640392780303955
    },
    {
      "auxiliary_loss_clip": 0.0121809,
      "auxiliary_loss_mlp": 0.01030158,
      "balance_loss_clip": 1.06332827,
      "balance_loss_mlp": 1.02087116,
      "epoch": 0.2329104791679192,
      "flos": 16727514215040.0,
      "grad_norm": 2.1616852845805195,
      "language_loss": 0.82291329,
      "learning_rate": 3.5835795993272513e-06,
      "loss": 0.8453958,
      "num_input_tokens_seen": 41318540,
      "step": 1937,
      "time_per_iteration": 2.5955958366394043
    },
    {
      "auxiliary_loss_clip": 0.01560411,
      "auxiliary_loss_mlp": 0.01038165,
      "balance_loss_clip": 1.04929137,
      "balance_loss_mlp": 1.02926636,
      "epoch": 0.2330307220585583,
      "flos": 22163743100160.0,
      "grad_norm": 1.8796297699986553,
      "language_loss": 0.71565819,
      "learning_rate": 3.583103688270391e-06,
      "loss": 0.74164391,
      "num_input_tokens_seen": 41338320,
      "step": 1938,
      "time_per_iteration": 4.017335653305054
    },
    {
      "auxiliary_loss_clip": 0.01310907,
      "auxiliary_loss_mlp": 0.01035524,
      "balance_loss_clip": 1.05752754,
      "balance_loss_mlp": 1.02649379,
      "epoch": 0.23315096494919738,
      "flos": 19317319787520.0,
      "grad_norm": 2.2487917354460043,
      "language_loss": 0.89787412,
      "learning_rate": 3.58262753705796e-06,
      "loss": 0.9213385,
      "num_input_tokens_seen": 41353210,
      "step": 1939,
      "time_per_iteration": 3.1744954586029053
    },
    {
      "auxiliary_loss_clip": 0.01203553,
      "auxiliary_loss_mlp": 0.0100709,
      "balance_loss_clip": 1.02303195,
      "balance_loss_mlp": 1.00573087,
      "epoch": 0.23327120783983646,
      "flos": 53031048946560.0,
      "grad_norm": 0.7681578756123835,
      "language_loss": 0.55518496,
      "learning_rate": 3.5821511457621902e-06,
      "loss": 0.57729137,
      "num_input_tokens_seen": 41410510,
      "step": 1940,
      "time_per_iteration": 3.305053949356079
    },
    {
      "auxiliary_loss_clip": 0.01315125,
      "auxiliary_loss_mlp": 0.0103521,
      "balance_loss_clip": 1.0605166,
      "balance_loss_mlp": 1.02479064,
      "epoch": 0.23339145073047557,
      "flos": 17126984344320.0,
      "grad_norm": 10.77120844767438,
      "language_loss": 0.81499714,
      "learning_rate": 3.5816745144553497e-06,
      "loss": 0.8385005,
      "num_input_tokens_seen": 41425830,
      "step": 1941,
      "time_per_iteration": 4.122735500335693
    },
    {
      "auxiliary_loss_clip": 0.01406752,
      "auxiliary_loss_mlp": 0.01035558,
      "balance_loss_clip": 1.05725837,
      "balance_loss_mlp": 1.02622938,
      "epoch": 0.23351169362111465,
      "flos": 13078918419840.0,
      "grad_norm": 2.022049390516949,
      "language_loss": 0.75497305,
      "learning_rate": 3.5811976432097424e-06,
      "loss": 0.77939612,
      "num_input_tokens_seen": 41443500,
      "step": 1942,
      "time_per_iteration": 2.7634754180908203
    },
    {
      "auxiliary_loss_clip": 0.01266659,
      "auxiliary_loss_mlp": 0.02574003,
      "balance_loss_clip": 1.06383562,
      "balance_loss_mlp": 1.00023746,
      "epoch": 0.23363193651175373,
      "flos": 15851257931520.0,
      "grad_norm": 2.2065232985677894,
      "language_loss": 0.85059702,
      "learning_rate": 3.58072053209771e-06,
      "loss": 0.88900363,
      "num_input_tokens_seen": 41460055,
      "step": 1943,
      "time_per_iteration": 2.597238063812256
    },
    {
      "auxiliary_loss_clip": 0.01314458,
      "auxiliary_loss_mlp": 0.01032224,
      "balance_loss_clip": 1.05587244,
      "balance_loss_mlp": 1.02298522,
      "epoch": 0.23375217940239285,
      "flos": 21025769345280.0,
      "grad_norm": 2.781758331341295,
      "language_loss": 0.7923435,
      "learning_rate": 3.5802431811916296e-06,
      "loss": 0.81581032,
      "num_input_tokens_seen": 41476665,
      "step": 1944,
      "time_per_iteration": 2.716317892074585
    },
    {
      "auxiliary_loss_clip": 0.01311858,
      "auxiliary_loss_mlp": 0.01037754,
      "balance_loss_clip": 1.05909705,
      "balance_loss_mlp": 1.02919507,
      "epoch": 0.23387242229303193,
      "flos": 20594698225920.0,
      "grad_norm": 1.7573575396385546,
      "language_loss": 0.80691469,
      "learning_rate": 3.579765590563916e-06,
      "loss": 0.83041084,
      "num_input_tokens_seen": 41496065,
      "step": 1945,
      "time_per_iteration": 2.6737658977508545
    },
    {
      "auxiliary_loss_clip": 0.01261556,
      "auxiliary_loss_mlp": 0.01044341,
      "balance_loss_clip": 1.05992639,
      "balance_loss_mlp": 1.03558517,
      "epoch": 0.233992665183671,
      "flos": 24279491952000.0,
      "grad_norm": 2.5308152510148147,
      "language_loss": 0.82059437,
      "learning_rate": 3.579287760287017e-06,
      "loss": 0.84365338,
      "num_input_tokens_seen": 41516815,
      "step": 1946,
      "time_per_iteration": 3.6124868392944336
    },
    {
      "auxiliary_loss_clip": 0.01265451,
      "auxiliary_loss_mlp": 0.01042863,
      "balance_loss_clip": 1.06311667,
      "balance_loss_mlp": 1.03421485,
      "epoch": 0.2341129080743101,
      "flos": 30154621121280.0,
      "grad_norm": 2.800109301967222,
      "language_loss": 0.72646242,
      "learning_rate": 3.578809690433421e-06,
      "loss": 0.74954557,
      "num_input_tokens_seen": 41538525,
      "step": 1947,
      "time_per_iteration": 2.7733728885650635
    },
    {
      "auxiliary_loss_clip": 0.01226059,
      "auxiliary_loss_mlp": 0.01030777,
      "balance_loss_clip": 1.06853926,
      "balance_loss_mlp": 1.021824,
      "epoch": 0.2342331509649492,
      "flos": 22784135829120.0,
      "grad_norm": 6.269405518569942,
      "language_loss": 0.81350422,
      "learning_rate": 3.578331381075651e-06,
      "loss": 0.83607256,
      "num_input_tokens_seen": 41559025,
      "step": 1948,
      "time_per_iteration": 2.6663758754730225
    },
    {
      "auxiliary_loss_clip": 0.01268693,
      "auxiliary_loss_mlp": 0.0103513,
      "balance_loss_clip": 1.06054354,
      "balance_loss_mlp": 1.02617121,
      "epoch": 0.2343533938555883,
      "flos": 23623152687360.0,
      "grad_norm": 2.1343837253037643,
      "language_loss": 0.70236021,
      "learning_rate": 3.5778528322862646e-06,
      "loss": 0.72539842,
      "num_input_tokens_seen": 41577845,
      "step": 1949,
      "time_per_iteration": 2.6780872344970703
    },
    {
      "auxiliary_loss_clip": 0.01192808,
      "auxiliary_loss_mlp": 0.01036014,
      "balance_loss_clip": 1.06351829,
      "balance_loss_mlp": 1.02643538,
      "epoch": 0.23447363674622737,
      "flos": 24570332375040.0,
      "grad_norm": 1.5636887199867735,
      "language_loss": 0.86689496,
      "learning_rate": 3.5773740441378585e-06,
      "loss": 0.88918316,
      "num_input_tokens_seen": 41598600,
      "step": 1950,
      "time_per_iteration": 2.7372612953186035
    },
    {
      "auxiliary_loss_clip": 0.01267018,
      "auxiliary_loss_mlp": 0.01037934,
      "balance_loss_clip": 1.06264508,
      "balance_loss_mlp": 1.0289638,
      "epoch": 0.23459387963686648,
      "flos": 53140322119680.0,
      "grad_norm": 1.8369772268409867,
      "language_loss": 0.73931539,
      "learning_rate": 3.5768950167030633e-06,
      "loss": 0.76236498,
      "num_input_tokens_seen": 41623300,
      "step": 1951,
      "time_per_iteration": 2.918297052383423
    },
    {
      "auxiliary_loss_clip": 0.01308948,
      "auxiliary_loss_mlp": 0.0103757,
      "balance_loss_clip": 1.05538547,
      "balance_loss_mlp": 1.02831984,
      "epoch": 0.23471412252750556,
      "flos": 23951412103680.0,
      "grad_norm": 2.1587274892082893,
      "language_loss": 0.78558046,
      "learning_rate": 3.576415750054548e-06,
      "loss": 0.80904567,
      "num_input_tokens_seen": 41643420,
      "step": 1952,
      "time_per_iteration": 2.755760431289673
    },
    {
      "auxiliary_loss_clip": 0.01310601,
      "auxiliary_loss_mlp": 0.01038935,
      "balance_loss_clip": 1.05669212,
      "balance_loss_mlp": 1.02987516,
      "epoch": 0.23483436541814465,
      "flos": 15706573948800.0,
      "grad_norm": 2.372522104521301,
      "language_loss": 0.85843158,
      "learning_rate": 3.5759362442650172e-06,
      "loss": 0.88192689,
      "num_input_tokens_seen": 41660170,
      "step": 1953,
      "time_per_iteration": 2.663092851638794
    },
    {
      "auxiliary_loss_clip": 0.01274665,
      "auxiliary_loss_mlp": 0.01036839,
      "balance_loss_clip": 1.06847405,
      "balance_loss_mlp": 1.02710593,
      "epoch": 0.23495460830878373,
      "flos": 24936262179840.0,
      "grad_norm": 3.307450071177933,
      "language_loss": 0.85340488,
      "learning_rate": 3.5754564994072113e-06,
      "loss": 0.87651998,
      "num_input_tokens_seen": 41679010,
      "step": 1954,
      "time_per_iteration": 2.74505877494812
    },
    {
      "auxiliary_loss_clip": 0.0131288,
      "auxiliary_loss_mlp": 0.01028638,
      "balance_loss_clip": 1.05630064,
      "balance_loss_mlp": 1.01975727,
      "epoch": 0.23507485119942284,
      "flos": 30482665056000.0,
      "grad_norm": 11.684594704474568,
      "language_loss": 0.60231233,
      "learning_rate": 3.5749765155539067e-06,
      "loss": 0.62572753,
      "num_input_tokens_seen": 41699495,
      "step": 1955,
      "time_per_iteration": 2.7644052505493164
    },
    {
      "auxiliary_loss_clip": 0.01359186,
      "auxiliary_loss_mlp": 0.01034059,
      "balance_loss_clip": 1.05582666,
      "balance_loss_mlp": 1.02411127,
      "epoch": 0.23519509409006192,
      "flos": 18329129746560.0,
      "grad_norm": 2.2800192627872304,
      "language_loss": 0.92112887,
      "learning_rate": 3.574496292777917e-06,
      "loss": 0.94506133,
      "num_input_tokens_seen": 41717705,
      "step": 1956,
      "time_per_iteration": 2.754769802093506
    },
    {
      "auxiliary_loss_clip": 0.01323638,
      "auxiliary_loss_mlp": 0.0103402,
      "balance_loss_clip": 1.06239367,
      "balance_loss_mlp": 1.0249002,
      "epoch": 0.235315336980701,
      "flos": 29643217234560.0,
      "grad_norm": 2.1288370011008677,
      "language_loss": 0.71735018,
      "learning_rate": 3.574015831152092e-06,
      "loss": 0.74092674,
      "num_input_tokens_seen": 41738120,
      "step": 1957,
      "time_per_iteration": 2.745117425918579
    },
    {
      "auxiliary_loss_clip": 0.01307505,
      "auxiliary_loss_mlp": 0.01029849,
      "balance_loss_clip": 1.05888104,
      "balance_loss_mlp": 1.02142727,
      "epoch": 0.23543557987134012,
      "flos": 18551704371840.0,
      "grad_norm": 5.404891455028431,
      "language_loss": 0.83219945,
      "learning_rate": 3.573535130749316e-06,
      "loss": 0.85557294,
      "num_input_tokens_seen": 41756070,
      "step": 1958,
      "time_per_iteration": 2.6845388412475586
    },
    {
      "auxiliary_loss_clip": 0.0131169,
      "auxiliary_loss_mlp": 0.01030623,
      "balance_loss_clip": 1.06130385,
      "balance_loss_mlp": 1.02159917,
      "epoch": 0.2355558227619792,
      "flos": 24679033908480.0,
      "grad_norm": 1.71806240589035,
      "language_loss": 0.73680335,
      "learning_rate": 3.5730541916425127e-06,
      "loss": 0.76022649,
      "num_input_tokens_seen": 41777550,
      "step": 1959,
      "time_per_iteration": 2.762404203414917
    },
    {
      "auxiliary_loss_clip": 0.01362397,
      "auxiliary_loss_mlp": 0.01030627,
      "balance_loss_clip": 1.05571818,
      "balance_loss_mlp": 1.02152562,
      "epoch": 0.23567606565261828,
      "flos": 21944795748480.0,
      "grad_norm": 2.231724518467874,
      "language_loss": 0.86245835,
      "learning_rate": 3.572573013904639e-06,
      "loss": 0.8863886,
      "num_input_tokens_seen": 41797460,
      "step": 1960,
      "time_per_iteration": 2.7309954166412354
    },
    {
      "auxiliary_loss_clip": 0.0121586,
      "auxiliary_loss_mlp": 0.01037409,
      "balance_loss_clip": 1.06296539,
      "balance_loss_mlp": 1.02857542,
      "epoch": 0.2357963085432574,
      "flos": 13589352639360.0,
      "grad_norm": 1.8784314343320545,
      "language_loss": 0.92044175,
      "learning_rate": 3.572091597608689e-06,
      "loss": 0.94297445,
      "num_input_tokens_seen": 41815585,
      "step": 1961,
      "time_per_iteration": 2.5806539058685303
    },
    {
      "auxiliary_loss_clip": 0.01319417,
      "auxiliary_loss_mlp": 0.01034203,
      "balance_loss_clip": 1.05955577,
      "balance_loss_mlp": 1.02547681,
      "epoch": 0.23591655143389648,
      "flos": 22088689632000.0,
      "grad_norm": 2.3151343868669114,
      "language_loss": 0.73922431,
      "learning_rate": 3.571609942827694e-06,
      "loss": 0.76276052,
      "num_input_tokens_seen": 41834700,
      "step": 1962,
      "time_per_iteration": 3.689704179763794
    },
    {
      "auxiliary_loss_clip": 0.01313408,
      "auxiliary_loss_mlp": 0.01033653,
      "balance_loss_clip": 1.05775845,
      "balance_loss_mlp": 1.02488494,
      "epoch": 0.23603679432453556,
      "flos": 17017349057280.0,
      "grad_norm": 2.315185929686582,
      "language_loss": 0.88805771,
      "learning_rate": 3.57112804963472e-06,
      "loss": 0.91152829,
      "num_input_tokens_seen": 41852915,
      "step": 1963,
      "time_per_iteration": 2.7121055126190186
    },
    {
      "auxiliary_loss_clip": 0.01358915,
      "auxiliary_loss_mlp": 0.01030618,
      "balance_loss_clip": 1.0601573,
      "balance_loss_mlp": 1.02232075,
      "epoch": 0.23615703721517464,
      "flos": 19171307001600.0,
      "grad_norm": 1.97287172454595,
      "language_loss": 0.76365477,
      "learning_rate": 3.57064591810287e-06,
      "loss": 0.78755009,
      "num_input_tokens_seen": 41870415,
      "step": 1964,
      "time_per_iteration": 3.5276012420654297
    },
    {
      "auxiliary_loss_clip": 0.01216366,
      "auxiliary_loss_mlp": 0.02573999,
      "balance_loss_clip": 1.06324303,
      "balance_loss_mlp": 1.00018704,
      "epoch": 0.23627728010581375,
      "flos": 19098803399040.0,
      "grad_norm": 2.2420860372933076,
      "language_loss": 0.80667353,
      "learning_rate": 3.570163548305284e-06,
      "loss": 0.84457719,
      "num_input_tokens_seen": 41889345,
      "step": 1965,
      "time_per_iteration": 2.691075563430786
    },
    {
      "auxiliary_loss_clip": 0.01312744,
      "auxiliary_loss_mlp": 0.01032523,
      "balance_loss_clip": 1.0587399,
      "balance_loss_mlp": 1.02330852,
      "epoch": 0.23639752299645284,
      "flos": 14282213057280.0,
      "grad_norm": 2.1560587593258442,
      "language_loss": 0.69773656,
      "learning_rate": 3.569680940315135e-06,
      "loss": 0.72118926,
      "num_input_tokens_seen": 41905745,
      "step": 1966,
      "time_per_iteration": 2.62320613861084
    },
    {
      "auxiliary_loss_clip": 0.01366355,
      "auxiliary_loss_mlp": 0.01035774,
      "balance_loss_clip": 1.05538809,
      "balance_loss_mlp": 1.02645183,
      "epoch": 0.23651776588709192,
      "flos": 22893411980160.0,
      "grad_norm": 1.8147274837027487,
      "language_loss": 0.81827688,
      "learning_rate": 3.5691980942056356e-06,
      "loss": 0.84229815,
      "num_input_tokens_seen": 41925115,
      "step": 1967,
      "time_per_iteration": 2.7448320388793945
    },
    {
      "auxiliary_loss_clip": 0.01267062,
      "auxiliary_loss_mlp": 0.0103266,
      "balance_loss_clip": 1.0581491,
      "balance_loss_mlp": 1.02358246,
      "epoch": 0.23663800877773103,
      "flos": 18624531196800.0,
      "grad_norm": 1.9691872527670764,
      "language_loss": 0.79817414,
      "learning_rate": 3.5687150100500332e-06,
      "loss": 0.8211714,
      "num_input_tokens_seen": 41944815,
      "step": 1968,
      "time_per_iteration": 3.673537015914917
    },
    {
      "auxiliary_loss_clip": 0.01270196,
      "auxiliary_loss_mlp": 0.01034175,
      "balance_loss_clip": 1.0625664,
      "balance_loss_mlp": 1.0252645,
      "epoch": 0.2367582516683701,
      "flos": 25555828896000.0,
      "grad_norm": 1.8866810138466232,
      "language_loss": 0.74434906,
      "learning_rate": 3.568231687921611e-06,
      "loss": 0.76739275,
      "num_input_tokens_seen": 41964990,
      "step": 1969,
      "time_per_iteration": 2.8294010162353516
    },
    {
      "auxiliary_loss_clip": 0.01212969,
      "auxiliary_loss_mlp": 0.01034756,
      "balance_loss_clip": 1.06124902,
      "balance_loss_mlp": 1.02666163,
      "epoch": 0.2368784945590092,
      "flos": 23295072839040.0,
      "grad_norm": 1.611411390724804,
      "language_loss": 0.8067534,
      "learning_rate": 3.5677481278936883e-06,
      "loss": 0.82923067,
      "num_input_tokens_seen": 41984570,
      "step": 1970,
      "time_per_iteration": 2.633720874786377
    },
    {
      "auxiliary_loss_clip": 0.01214663,
      "auxiliary_loss_mlp": 0.01001824,
      "balance_loss_clip": 1.02942157,
      "balance_loss_mlp": 1.00039911,
      "epoch": 0.23699873744964828,
      "flos": 69859291875840.0,
      "grad_norm": 0.8380039256783458,
      "language_loss": 0.57857513,
      "learning_rate": 3.5672643300396214e-06,
      "loss": 0.60073996,
      "num_input_tokens_seen": 42053715,
      "step": 1971,
      "time_per_iteration": 3.3568155765533447
    },
    {
      "auxiliary_loss_clip": 0.01361967,
      "auxiliary_loss_mlp": 0.01029079,
      "balance_loss_clip": 1.05771732,
      "balance_loss_mlp": 1.02097893,
      "epoch": 0.2371189803402874,
      "flos": 21835052720640.0,
      "grad_norm": 2.486289940663306,
      "language_loss": 0.67199159,
      "learning_rate": 3.566780294432802e-06,
      "loss": 0.69590205,
      "num_input_tokens_seen": 42070890,
      "step": 1972,
      "time_per_iteration": 3.578482151031494
    },
    {
      "auxiliary_loss_clip": 0.01212149,
      "auxiliary_loss_mlp": 0.01037378,
      "balance_loss_clip": 1.05975878,
      "balance_loss_mlp": 1.02877676,
      "epoch": 0.23723922323092647,
      "flos": 21908490076800.0,
      "grad_norm": 2.7585687578577693,
      "language_loss": 0.74966359,
      "learning_rate": 3.566296021146657e-06,
      "loss": 0.77215886,
      "num_input_tokens_seen": 42090270,
      "step": 1973,
      "time_per_iteration": 2.655172348022461
    },
    {
      "auxiliary_loss_clip": 0.01217586,
      "auxiliary_loss_mlp": 0.01038596,
      "balance_loss_clip": 1.06187582,
      "balance_loss_mlp": 1.02922583,
      "epoch": 0.23735946612156555,
      "flos": 32708803380480.0,
      "grad_norm": 2.0995161708564365,
      "language_loss": 0.73226601,
      "learning_rate": 3.565811510254652e-06,
      "loss": 0.7548278,
      "num_input_tokens_seen": 42111150,
      "step": 1974,
      "time_per_iteration": 2.792651414871216
    },
    {
      "auxiliary_loss_clip": 0.0116899,
      "auxiliary_loss_mlp": 0.01003555,
      "balance_loss_clip": 1.03577042,
      "balance_loss_mlp": 1.00207698,
      "epoch": 0.23747970901220466,
      "flos": 70546944821760.0,
      "grad_norm": 0.8417150792464798,
      "language_loss": 0.58221865,
      "learning_rate": 3.5653267618302845e-06,
      "loss": 0.60394406,
      "num_input_tokens_seen": 42178730,
      "step": 1975,
      "time_per_iteration": 3.2486581802368164
    },
    {
      "auxiliary_loss_clip": 0.01214581,
      "auxiliary_loss_mlp": 0.01028898,
      "balance_loss_clip": 1.06205821,
      "balance_loss_mlp": 1.02068472,
      "epoch": 0.23759995190284375,
      "flos": 20849807594880.0,
      "grad_norm": 2.020923440478174,
      "language_loss": 0.85912204,
      "learning_rate": 3.564841775947093e-06,
      "loss": 0.88155681,
      "num_input_tokens_seen": 42199620,
      "step": 1976,
      "time_per_iteration": 2.642845630645752
    },
    {
      "auxiliary_loss_clip": 0.01363986,
      "auxiliary_loss_mlp": 0.01038245,
      "balance_loss_clip": 1.05520058,
      "balance_loss_mlp": 1.02973986,
      "epoch": 0.23772019479348283,
      "flos": 32921645420160.0,
      "grad_norm": 2.106009236525713,
      "language_loss": 0.76187587,
      "learning_rate": 3.5643565526786475e-06,
      "loss": 0.78589821,
      "num_input_tokens_seen": 42219560,
      "step": 1977,
      "time_per_iteration": 2.781310796737671
    },
    {
      "auxiliary_loss_clip": 0.01218005,
      "auxiliary_loss_mlp": 0.01031576,
      "balance_loss_clip": 1.06349182,
      "balance_loss_mlp": 1.02338004,
      "epoch": 0.2378404376841219,
      "flos": 32342765834880.0,
      "grad_norm": 1.6646778667112287,
      "language_loss": 0.77335358,
      "learning_rate": 3.5638710920985574e-06,
      "loss": 0.79584944,
      "num_input_tokens_seen": 42241020,
      "step": 1978,
      "time_per_iteration": 2.7182579040527344
    },
    {
      "auxiliary_loss_clip": 0.01268689,
      "auxiliary_loss_mlp": 0.02575929,
      "balance_loss_clip": 1.05738401,
      "balance_loss_mlp": 1.00015271,
      "epoch": 0.23796068057476102,
      "flos": 22997624313600.0,
      "grad_norm": 2.2701671042219673,
      "language_loss": 0.82114089,
      "learning_rate": 3.5633853942804655e-06,
      "loss": 0.85958707,
      "num_input_tokens_seen": 42259345,
      "step": 1979,
      "time_per_iteration": 2.6937849521636963
    },
    {
      "auxiliary_loss_clip": 0.01358531,
      "auxiliary_loss_mlp": 0.01043075,
      "balance_loss_clip": 1.05047262,
      "balance_loss_mlp": 1.03332412,
      "epoch": 0.2380809234654001,
      "flos": 13480938414720.0,
      "grad_norm": 2.5441354146983355,
      "language_loss": 0.77011275,
      "learning_rate": 3.5628994592980527e-06,
      "loss": 0.79412889,
      "num_input_tokens_seen": 42277250,
      "step": 1980,
      "time_per_iteration": 2.6846225261688232
    },
    {
      "auxiliary_loss_clip": 0.01214338,
      "auxiliary_loss_mlp": 0.01029291,
      "balance_loss_clip": 1.06030369,
      "balance_loss_mlp": 1.02029061,
      "epoch": 0.2382011663560392,
      "flos": 16871803148160.0,
      "grad_norm": 1.881965822172395,
      "language_loss": 0.70491612,
      "learning_rate": 3.562413287225034e-06,
      "loss": 0.72735244,
      "num_input_tokens_seen": 42295360,
      "step": 1981,
      "time_per_iteration": 2.671572208404541
    },
    {
      "auxiliary_loss_clip": 0.01260361,
      "auxiliary_loss_mlp": 0.0103509,
      "balance_loss_clip": 1.05882716,
      "balance_loss_mlp": 1.02587533,
      "epoch": 0.2383214092466783,
      "flos": 18441135331200.0,
      "grad_norm": 2.507571590224718,
      "language_loss": 0.89286542,
      "learning_rate": 3.5619268781351623e-06,
      "loss": 0.91581994,
      "num_input_tokens_seen": 42313430,
      "step": 1982,
      "time_per_iteration": 2.736248731613159
    },
    {
      "auxiliary_loss_clip": 0.01306625,
      "auxiliary_loss_mlp": 0.01036559,
      "balance_loss_clip": 1.0591445,
      "balance_loss_mlp": 1.02762437,
      "epoch": 0.23844165213731738,
      "flos": 19755717281280.0,
      "grad_norm": 1.8050434164249045,
      "language_loss": 0.76920247,
      "learning_rate": 3.5614402321022256e-06,
      "loss": 0.79263437,
      "num_input_tokens_seen": 42331260,
      "step": 1983,
      "time_per_iteration": 2.777867317199707
    },
    {
      "auxiliary_loss_clip": 0.01405862,
      "auxiliary_loss_mlp": 0.01042337,
      "balance_loss_clip": 1.05546391,
      "balance_loss_mlp": 1.03346765,
      "epoch": 0.23856189502795647,
      "flos": 23367360960000.0,
      "grad_norm": 1.9057890217624283,
      "language_loss": 0.87069786,
      "learning_rate": 3.5609533492000463e-06,
      "loss": 0.89517975,
      "num_input_tokens_seen": 42350150,
      "step": 1984,
      "time_per_iteration": 2.8368828296661377
    },
    {
      "auxiliary_loss_clip": 0.01309934,
      "auxiliary_loss_mlp": 0.01027565,
      "balance_loss_clip": 1.06023574,
      "balance_loss_mlp": 1.01854038,
      "epoch": 0.23868213791859555,
      "flos": 23475056912640.0,
      "grad_norm": 2.2607608470244025,
      "language_loss": 0.78625858,
      "learning_rate": 3.560466229502485e-06,
      "loss": 0.80963349,
      "num_input_tokens_seen": 42369495,
      "step": 1985,
      "time_per_iteration": 2.8222124576568604
    },
    {
      "auxiliary_loss_clip": 0.01309288,
      "auxiliary_loss_mlp": 0.02572892,
      "balance_loss_clip": 1.05953538,
      "balance_loss_mlp": 1.00016475,
      "epoch": 0.23880238080923466,
      "flos": 16617340224000.0,
      "grad_norm": 2.097917112169926,
      "language_loss": 0.90552491,
      "learning_rate": 3.5599788730834384e-06,
      "loss": 0.94434667,
      "num_input_tokens_seen": 42387455,
      "step": 1986,
      "time_per_iteration": 2.7352869510650635
    },
    {
      "auxiliary_loss_clip": 0.01268022,
      "auxiliary_loss_mlp": 0.01028866,
      "balance_loss_clip": 1.05988789,
      "balance_loss_mlp": 1.01972818,
      "epoch": 0.23892262369987374,
      "flos": 17348409734400.0,
      "grad_norm": 2.11975285821781,
      "language_loss": 0.78729534,
      "learning_rate": 3.559491280016836e-06,
      "loss": 0.81026423,
      "num_input_tokens_seen": 42405400,
      "step": 1987,
      "time_per_iteration": 2.6619064807891846
    },
    {
      "auxiliary_loss_clip": 0.01309793,
      "auxiliary_loss_mlp": 0.01032637,
      "balance_loss_clip": 1.05692649,
      "balance_loss_mlp": 1.02419674,
      "epoch": 0.23904286659051283,
      "flos": 22309899540480.0,
      "grad_norm": 1.650851691020631,
      "language_loss": 0.70798135,
      "learning_rate": 3.5590034503766465e-06,
      "loss": 0.73140562,
      "num_input_tokens_seen": 42425065,
      "step": 1988,
      "time_per_iteration": 3.600346565246582
    },
    {
      "auxiliary_loss_clip": 0.01216045,
      "auxiliary_loss_mlp": 0.01032173,
      "balance_loss_clip": 1.06330311,
      "balance_loss_mlp": 1.02353644,
      "epoch": 0.23916310948115194,
      "flos": 21178246579200.0,
      "grad_norm": 2.0397553111468887,
      "language_loss": 0.81040114,
      "learning_rate": 3.558515384236874e-06,
      "loss": 0.83288336,
      "num_input_tokens_seen": 42442495,
      "step": 1989,
      "time_per_iteration": 2.5338199138641357
    },
    {
      "auxiliary_loss_clip": 0.01353023,
      "auxiliary_loss_mlp": 0.02574616,
      "balance_loss_clip": 1.05504668,
      "balance_loss_mlp": 1.00017416,
      "epoch": 0.23928335237179102,
      "flos": 14137349506560.0,
      "grad_norm": 1.8592027016786858,
      "language_loss": 0.84102112,
      "learning_rate": 3.558027081671556e-06,
      "loss": 0.88029748,
      "num_input_tokens_seen": 42459480,
      "step": 1990,
      "time_per_iteration": 3.3376665115356445
    },
    {
      "auxiliary_loss_clip": 0.01266204,
      "auxiliary_loss_mlp": 0.01030024,
      "balance_loss_clip": 1.05793333,
      "balance_loss_mlp": 1.02114844,
      "epoch": 0.2394035952624301,
      "flos": 23769596436480.0,
      "grad_norm": 1.6927789625105827,
      "language_loss": 0.6882571,
      "learning_rate": 3.557538542754769e-06,
      "loss": 0.71121937,
      "num_input_tokens_seen": 42479175,
      "step": 1991,
      "time_per_iteration": 2.5560810565948486
    },
    {
      "auxiliary_loss_clip": 0.0121992,
      "auxiliary_loss_mlp": 0.01035888,
      "balance_loss_clip": 1.06575692,
      "balance_loss_mlp": 1.02659559,
      "epoch": 0.2395238381530692,
      "flos": 24206198250240.0,
      "grad_norm": 2.379044128226998,
      "language_loss": 0.66700256,
      "learning_rate": 3.557049767560623e-06,
      "loss": 0.68956071,
      "num_input_tokens_seen": 42498090,
      "step": 1992,
      "time_per_iteration": 2.6965601444244385
    },
    {
      "auxiliary_loss_clip": 0.01416011,
      "auxiliary_loss_mlp": 0.01033342,
      "balance_loss_clip": 1.06122541,
      "balance_loss_mlp": 1.02488422,
      "epoch": 0.2396440810437083,
      "flos": 25295763450240.0,
      "grad_norm": 2.7165730784434112,
      "language_loss": 0.85916483,
      "learning_rate": 3.5565607561632655e-06,
      "loss": 0.88365829,
      "num_input_tokens_seen": 42516930,
      "step": 1993,
      "time_per_iteration": 3.685983419418335
    },
    {
      "auxiliary_loss_clip": 0.01311788,
      "auxiliary_loss_mlp": 0.01031145,
      "balance_loss_clip": 1.06021059,
      "balance_loss_mlp": 1.02232385,
      "epoch": 0.23976432393434738,
      "flos": 28543093436160.0,
      "grad_norm": 2.7150633278523983,
      "language_loss": 0.79951727,
      "learning_rate": 3.5560715086368787e-06,
      "loss": 0.82294667,
      "num_input_tokens_seen": 42534800,
      "step": 1994,
      "time_per_iteration": 2.750213623046875
    },
    {
      "auxiliary_loss_clip": 0.01309537,
      "auxiliary_loss_mlp": 0.0103627,
      "balance_loss_clip": 1.0612725,
      "balance_loss_mlp": 1.0273416,
      "epoch": 0.23988456682498646,
      "flos": 19494358945920.0,
      "grad_norm": 2.116510283387766,
      "language_loss": 0.82538503,
      "learning_rate": 3.5555820250556816e-06,
      "loss": 0.8488431,
      "num_input_tokens_seen": 42552000,
      "step": 1995,
      "time_per_iteration": 2.6314034461975098
    },
    {
      "auxiliary_loss_clip": 0.01241419,
      "auxiliary_loss_mlp": 0.0103151,
      "balance_loss_clip": 1.06253767,
      "balance_loss_mlp": 1.02220619,
      "epoch": 0.24000480971562557,
      "flos": 20266331068800.0,
      "grad_norm": 2.377665055768994,
      "language_loss": 0.69521999,
      "learning_rate": 3.5550923054939278e-06,
      "loss": 0.71794927,
      "num_input_tokens_seen": 42571455,
      "step": 1996,
      "time_per_iteration": 2.6452934741973877
    },
    {
      "auxiliary_loss_clip": 0.01338127,
      "auxiliary_loss_mlp": 0.01039711,
      "balance_loss_clip": 1.05845046,
      "balance_loss_mlp": 1.03147936,
      "epoch": 0.24012505260626466,
      "flos": 25443176866560.0,
      "grad_norm": 2.1074794213469943,
      "language_loss": 0.74738097,
      "learning_rate": 3.5546023500259083e-06,
      "loss": 0.77115935,
      "num_input_tokens_seen": 42592550,
      "step": 1997,
      "time_per_iteration": 2.725383758544922
    },
    {
      "auxiliary_loss_clip": 0.01418492,
      "auxiliary_loss_mlp": 0.01035021,
      "balance_loss_clip": 1.05386472,
      "balance_loss_mlp": 1.02637875,
      "epoch": 0.24024529549690374,
      "flos": 15553342529280.0,
      "grad_norm": 2.18626864242386,
      "language_loss": 0.8076126,
      "learning_rate": 3.5541121587259477e-06,
      "loss": 0.83214772,
      "num_input_tokens_seen": 42610385,
      "step": 1998,
      "time_per_iteration": 3.613677740097046
    },
    {
      "auxiliary_loss_clip": 0.01158875,
      "auxiliary_loss_mlp": 0.009993,
      "balance_loss_clip": 1.03155887,
      "balance_loss_mlp": 0.99806005,
      "epoch": 0.24036553838754285,
      "flos": 57122351867520.0,
      "grad_norm": 0.8471247679829159,
      "language_loss": 0.57873303,
      "learning_rate": 3.553621731668408e-06,
      "loss": 0.60031474,
      "num_input_tokens_seen": 42673595,
      "step": 1999,
      "time_per_iteration": 3.187566041946411
    },
    {
      "auxiliary_loss_clip": 0.01263527,
      "auxiliary_loss_mlp": 0.01030522,
      "balance_loss_clip": 1.05807412,
      "balance_loss_mlp": 1.02145588,
      "epoch": 0.24048578127818193,
      "flos": 24969946158720.0,
      "grad_norm": 1.9242364563531915,
      "language_loss": 0.83397371,
      "learning_rate": 3.553131068927688e-06,
      "loss": 0.85691416,
      "num_input_tokens_seen": 42692000,
      "step": 2000,
      "time_per_iteration": 2.661674737930298
    },
    {
      "auxiliary_loss_clip": 0.01362067,
      "auxiliary_loss_mlp": 0.01031467,
      "balance_loss_clip": 1.05913949,
      "balance_loss_mlp": 1.02263355,
      "epoch": 0.24060602416882101,
      "flos": 23330947547520.0,
      "grad_norm": 1.8435314135481022,
      "language_loss": 0.80364478,
      "learning_rate": 3.552640170578219e-06,
      "loss": 0.82758009,
      "num_input_tokens_seen": 42712250,
      "step": 2001,
      "time_per_iteration": 2.7319180965423584
    },
    {
      "auxiliary_loss_clip": 0.01312829,
      "auxiliary_loss_mlp": 0.01031744,
      "balance_loss_clip": 1.0576216,
      "balance_loss_mlp": 1.02279711,
      "epoch": 0.2407262670594601,
      "flos": 14173260128640.0,
      "grad_norm": 2.247766143998083,
      "language_loss": 0.78108573,
      "learning_rate": 3.5521490366944703e-06,
      "loss": 0.80453146,
      "num_input_tokens_seen": 42729900,
      "step": 2002,
      "time_per_iteration": 2.619098663330078
    },
    {
      "auxiliary_loss_clip": 0.01357781,
      "auxiliary_loss_mlp": 0.01038717,
      "balance_loss_clip": 1.05494809,
      "balance_loss_mlp": 1.030092,
      "epoch": 0.2408465099500992,
      "flos": 13663113217920.0,
      "grad_norm": 2.6306154895084375,
      "language_loss": 0.80094069,
      "learning_rate": 3.5516576673509474e-06,
      "loss": 0.82490563,
      "num_input_tokens_seen": 42747900,
      "step": 2003,
      "time_per_iteration": 2.7244203090667725
    },
    {
      "auxiliary_loss_clip": 0.01218792,
      "auxiliary_loss_mlp": 0.01035916,
      "balance_loss_clip": 1.06334496,
      "balance_loss_mlp": 1.02716589,
      "epoch": 0.2409667528407383,
      "flos": 31248029076480.0,
      "grad_norm": 1.9714584207925971,
      "language_loss": 0.86039865,
      "learning_rate": 3.5511660626221896e-06,
      "loss": 0.88294578,
      "num_input_tokens_seen": 42768540,
      "step": 2004,
      "time_per_iteration": 2.702199935913086
    },
    {
      "auxiliary_loss_clip": 0.01310774,
      "auxiliary_loss_mlp": 0.02576519,
      "balance_loss_clip": 1.06084776,
      "balance_loss_mlp": 1.00027585,
      "epoch": 0.24108699573137737,
      "flos": 22199941031040.0,
      "grad_norm": 2.362305996985324,
      "language_loss": 0.89356732,
      "learning_rate": 3.5506742225827744e-06,
      "loss": 0.93244016,
      "num_input_tokens_seen": 42785395,
      "step": 2005,
      "time_per_iteration": 2.7343859672546387
    },
    {
      "auxiliary_loss_clip": 0.01360394,
      "auxiliary_loss_mlp": 0.01035261,
      "balance_loss_clip": 1.05509853,
      "balance_loss_mlp": 1.02651143,
      "epoch": 0.24120723862201648,
      "flos": 26103035664000.0,
      "grad_norm": 2.2577006207401342,
      "language_loss": 0.90309417,
      "learning_rate": 3.5501821473073116e-06,
      "loss": 0.92705065,
      "num_input_tokens_seen": 42801980,
      "step": 2006,
      "time_per_iteration": 2.799283266067505
    },
    {
      "auxiliary_loss_clip": 0.0136762,
      "auxiliary_loss_mlp": 0.0103922,
      "balance_loss_clip": 1.06252253,
      "balance_loss_mlp": 1.02981424,
      "epoch": 0.24132748151265557,
      "flos": 18624926246400.0,
      "grad_norm": 2.355780744988493,
      "language_loss": 0.86886299,
      "learning_rate": 3.54968983687045e-06,
      "loss": 0.89293134,
      "num_input_tokens_seen": 42818850,
      "step": 2007,
      "time_per_iteration": 2.805720090866089
    },
    {
      "auxiliary_loss_clip": 0.01327076,
      "auxiliary_loss_mlp": 0.01034896,
      "balance_loss_clip": 1.06501758,
      "balance_loss_mlp": 1.02528739,
      "epoch": 0.24144772440329465,
      "flos": 15267673664640.0,
      "grad_norm": 4.704649427077454,
      "language_loss": 0.8913883,
      "learning_rate": 3.549197291346872e-06,
      "loss": 0.91500795,
      "num_input_tokens_seen": 42835375,
      "step": 2008,
      "time_per_iteration": 2.692507266998291
    },
    {
      "auxiliary_loss_clip": 0.01267802,
      "auxiliary_loss_mlp": 0.01033382,
      "balance_loss_clip": 1.05963373,
      "balance_loss_mlp": 1.0236429,
      "epoch": 0.24156796729393373,
      "flos": 24024274842240.0,
      "grad_norm": 2.553275814649865,
      "language_loss": 0.79225433,
      "learning_rate": 3.548704510811297e-06,
      "loss": 0.81526619,
      "num_input_tokens_seen": 42854570,
      "step": 2009,
      "time_per_iteration": 2.692596197128296
    },
    {
      "auxiliary_loss_clip": 0.01337367,
      "auxiliary_loss_mlp": 0.01034317,
      "balance_loss_clip": 1.0578258,
      "balance_loss_mlp": 1.02455926,
      "epoch": 0.24168821018457284,
      "flos": 26286790665600.0,
      "grad_norm": 3.780559895009379,
      "language_loss": 0.74971807,
      "learning_rate": 3.5482114953384787e-06,
      "loss": 0.77343494,
      "num_input_tokens_seen": 42873800,
      "step": 2010,
      "time_per_iteration": 2.83145809173584
    },
    {
      "auxiliary_loss_clip": 0.01267707,
      "auxiliary_loss_mlp": 0.01028913,
      "balance_loss_clip": 1.05987489,
      "balance_loss_mlp": 1.01998973,
      "epoch": 0.24180845307521193,
      "flos": 18223193560320.0,
      "grad_norm": 2.8863378163734335,
      "language_loss": 0.84732389,
      "learning_rate": 3.5477182450032077e-06,
      "loss": 0.87029004,
      "num_input_tokens_seen": 42892400,
      "step": 2011,
      "time_per_iteration": 2.62113356590271
    },
    {
      "auxiliary_loss_clip": 0.01264079,
      "auxiliary_loss_mlp": 0.01035719,
      "balance_loss_clip": 1.06093526,
      "balance_loss_mlp": 1.02550304,
      "epoch": 0.241928695965851,
      "flos": 20449260057600.0,
      "grad_norm": 2.4706478739861994,
      "language_loss": 0.83265519,
      "learning_rate": 3.5472247598803097e-06,
      "loss": 0.85565317,
      "num_input_tokens_seen": 42911745,
      "step": 2012,
      "time_per_iteration": 2.6657283306121826
    },
    {
      "auxiliary_loss_clip": 0.01216903,
      "auxiliary_loss_mlp": 0.0102583,
      "balance_loss_clip": 1.06109309,
      "balance_loss_mlp": 1.01607907,
      "epoch": 0.24204893885649012,
      "flos": 25556475340800.0,
      "grad_norm": 4.04297839284616,
      "language_loss": 0.85286772,
      "learning_rate": 3.546731040044645e-06,
      "loss": 0.87529504,
      "num_input_tokens_seen": 42926915,
      "step": 2013,
      "time_per_iteration": 2.614208459854126
    },
    {
      "auxiliary_loss_clip": 0.0121723,
      "auxiliary_loss_mlp": 0.01040813,
      "balance_loss_clip": 1.06363845,
      "balance_loss_mlp": 1.0311811,
      "epoch": 0.2421691817471292,
      "flos": 30660207004800.0,
      "grad_norm": 2.2108174196615433,
      "language_loss": 0.75252712,
      "learning_rate": 3.546237085571112e-06,
      "loss": 0.7751075,
      "num_input_tokens_seen": 42945350,
      "step": 2014,
      "time_per_iteration": 3.5687689781188965
    },
    {
      "auxiliary_loss_clip": 0.01268409,
      "auxiliary_loss_mlp": 0.01030749,
      "balance_loss_clip": 1.0627507,
      "balance_loss_mlp": 1.02164114,
      "epoch": 0.24228942463776829,
      "flos": 21945011230080.0,
      "grad_norm": 3.1361228160049675,
      "language_loss": 0.72493315,
      "learning_rate": 3.5457428965346425e-06,
      "loss": 0.74792469,
      "num_input_tokens_seen": 42964290,
      "step": 2015,
      "time_per_iteration": 2.6901602745056152
    },
    {
      "auxiliary_loss_clip": 0.01465297,
      "auxiliary_loss_mlp": 0.01032528,
      "balance_loss_clip": 1.05375671,
      "balance_loss_mlp": 1.02343798,
      "epoch": 0.2424096675284074,
      "flos": 33984493879680.0,
      "grad_norm": 1.6285959609266867,
      "language_loss": 0.74736071,
      "learning_rate": 3.545248473010205e-06,
      "loss": 0.77233893,
      "num_input_tokens_seen": 42987095,
      "step": 2016,
      "time_per_iteration": 3.8339364528656006
    },
    {
      "auxiliary_loss_clip": 0.01219904,
      "auxiliary_loss_mlp": 0.02582677,
      "balance_loss_clip": 1.06295061,
      "balance_loss_mlp": 1.00026536,
      "epoch": 0.24252991041904648,
      "flos": 21653416621440.0,
      "grad_norm": 1.8164999601366036,
      "language_loss": 0.880404,
      "learning_rate": 3.544753815072802e-06,
      "loss": 0.91842979,
      "num_input_tokens_seen": 43005750,
      "step": 2017,
      "time_per_iteration": 2.8230113983154297
    },
    {
      "auxiliary_loss_clip": 0.01514801,
      "auxiliary_loss_mlp": 0.01035001,
      "balance_loss_clip": 1.05293524,
      "balance_loss_mlp": 1.02533889,
      "epoch": 0.24265015330968556,
      "flos": 21870065502720.0,
      "grad_norm": 1.9165965199080641,
      "language_loss": 0.88270915,
      "learning_rate": 3.544258922797474e-06,
      "loss": 0.90820718,
      "num_input_tokens_seen": 43023870,
      "step": 2018,
      "time_per_iteration": 2.9394783973693848
    },
    {
      "auxiliary_loss_clip": 0.01214447,
      "auxiliary_loss_mlp": 0.01033969,
      "balance_loss_clip": 1.06110215,
      "balance_loss_mlp": 1.02478421,
      "epoch": 0.24277039620032465,
      "flos": 25628260671360.0,
      "grad_norm": 1.6914831417435723,
      "language_loss": 0.78350824,
      "learning_rate": 3.543763796259295e-06,
      "loss": 0.80599236,
      "num_input_tokens_seen": 43043825,
      "step": 2019,
      "time_per_iteration": 4.181771755218506
    },
    {
      "auxiliary_loss_clip": 0.01261113,
      "auxiliary_loss_mlp": 0.01037179,
      "balance_loss_clip": 1.05909586,
      "balance_loss_mlp": 1.02760696,
      "epoch": 0.24289063909096376,
      "flos": 26286575184000.0,
      "grad_norm": 1.8065392786198953,
      "language_loss": 0.90800297,
      "learning_rate": 3.5432684355333754e-06,
      "loss": 0.93098593,
      "num_input_tokens_seen": 43062480,
      "step": 2020,
      "time_per_iteration": 2.7367825508117676
    },
    {
      "auxiliary_loss_clip": 0.01265048,
      "auxiliary_loss_mlp": 0.01031704,
      "balance_loss_clip": 1.05884027,
      "balance_loss_mlp": 1.02251899,
      "epoch": 0.24301088198160284,
      "flos": 25075056332160.0,
      "grad_norm": 3.0508284778579164,
      "language_loss": 0.76728487,
      "learning_rate": 3.5427728406948613e-06,
      "loss": 0.79025239,
      "num_input_tokens_seen": 43081595,
      "step": 2021,
      "time_per_iteration": 2.6627392768859863
    },
    {
      "auxiliary_loss_clip": 0.01215785,
      "auxiliary_loss_mlp": 0.01001111,
      "balance_loss_clip": 1.03045678,
      "balance_loss_mlp": 0.99981135,
      "epoch": 0.24313112487224192,
      "flos": 69900948673920.0,
      "grad_norm": 0.7497471874378326,
      "language_loss": 0.57918411,
      "learning_rate": 3.542277011818934e-06,
      "loss": 0.60135305,
      "num_input_tokens_seen": 43145430,
      "step": 2022,
      "time_per_iteration": 3.4131600856781006
    },
    {
      "auxiliary_loss_clip": 0.01235061,
      "auxiliary_loss_mlp": 0.01034721,
      "balance_loss_clip": 1.0626725,
      "balance_loss_mlp": 1.02616763,
      "epoch": 0.24325136776288103,
      "flos": 40662334235520.0,
      "grad_norm": 2.014982228662073,
      "language_loss": 0.74118161,
      "learning_rate": 3.5417809489808104e-06,
      "loss": 0.76387942,
      "num_input_tokens_seen": 43167040,
      "step": 2023,
      "time_per_iteration": 2.872576951980591
    },
    {
      "auxiliary_loss_clip": 0.0126822,
      "auxiliary_loss_mlp": 0.01037005,
      "balance_loss_clip": 1.06218314,
      "balance_loss_mlp": 1.02793896,
      "epoch": 0.24337161065352012,
      "flos": 25046400257280.0,
      "grad_norm": 2.002194433395579,
      "language_loss": 0.72552025,
      "learning_rate": 3.5412846522557422e-06,
      "loss": 0.74857253,
      "num_input_tokens_seen": 43187930,
      "step": 2024,
      "time_per_iteration": 3.6466939449310303
    },
    {
      "auxiliary_loss_clip": 0.01218411,
      "auxiliary_loss_mlp": 0.01037313,
      "balance_loss_clip": 1.06427062,
      "balance_loss_mlp": 1.02787161,
      "epoch": 0.2434918535441592,
      "flos": 18661160090880.0,
      "grad_norm": 2.095112210630518,
      "language_loss": 0.73884177,
      "learning_rate": 3.540788121719018e-06,
      "loss": 0.76139903,
      "num_input_tokens_seen": 43206350,
      "step": 2025,
      "time_per_iteration": 2.598159074783325
    },
    {
      "auxiliary_loss_clip": 0.0136401,
      "auxiliary_loss_mlp": 0.01038537,
      "balance_loss_clip": 1.06395447,
      "balance_loss_mlp": 1.02960205,
      "epoch": 0.24361209643479828,
      "flos": 23915142345600.0,
      "grad_norm": 2.131651190888136,
      "language_loss": 0.82259047,
      "learning_rate": 3.5402913574459604e-06,
      "loss": 0.84661597,
      "num_input_tokens_seen": 43226255,
      "step": 2026,
      "time_per_iteration": 2.7055678367614746
    },
    {
      "auxiliary_loss_clip": 0.01377833,
      "auxiliary_loss_mlp": 0.01029067,
      "balance_loss_clip": 1.05007672,
      "balance_loss_mlp": 1.02025771,
      "epoch": 0.2437323393254374,
      "flos": 28657505232000.0,
      "grad_norm": 5.4704559427499175,
      "language_loss": 0.86235315,
      "learning_rate": 3.5397943595119297e-06,
      "loss": 0.88642216,
      "num_input_tokens_seen": 43247675,
      "step": 2027,
      "time_per_iteration": 2.838533878326416
    },
    {
      "auxiliary_loss_clip": 0.01312439,
      "auxiliary_loss_mlp": 0.01033824,
      "balance_loss_clip": 1.06176436,
      "balance_loss_mlp": 1.02503812,
      "epoch": 0.24385258221607647,
      "flos": 23550325862400.0,
      "grad_norm": 2.7597959833512826,
      "language_loss": 0.7724818,
      "learning_rate": 3.5392971279923177e-06,
      "loss": 0.79594433,
      "num_input_tokens_seen": 43265895,
      "step": 2028,
      "time_per_iteration": 2.691261053085327
    },
    {
      "auxiliary_loss_clip": 0.01361591,
      "auxiliary_loss_mlp": 0.01038411,
      "balance_loss_clip": 1.05517542,
      "balance_loss_mlp": 1.02910674,
      "epoch": 0.24397282510671556,
      "flos": 25336091445120.0,
      "grad_norm": 2.3025071263391492,
      "language_loss": 0.82986832,
      "learning_rate": 3.5387996629625557e-06,
      "loss": 0.85386837,
      "num_input_tokens_seen": 43283485,
      "step": 2029,
      "time_per_iteration": 2.8049063682556152
    },
    {
      "auxiliary_loss_clip": 0.01111175,
      "auxiliary_loss_mlp": 0.0100408,
      "balance_loss_clip": 1.02996111,
      "balance_loss_mlp": 1.00282192,
      "epoch": 0.24409306799735467,
      "flos": 65187421430400.0,
      "grad_norm": 0.805736003588865,
      "language_loss": 0.54988122,
      "learning_rate": 3.5383019644981083e-06,
      "loss": 0.57103372,
      "num_input_tokens_seen": 43347180,
      "step": 2030,
      "time_per_iteration": 3.201660394668579
    },
    {
      "auxiliary_loss_clip": 0.01309384,
      "auxiliary_loss_mlp": 0.010323,
      "balance_loss_clip": 1.05774641,
      "balance_loss_mlp": 1.02319813,
      "epoch": 0.24421331088799375,
      "flos": 19537093152000.0,
      "grad_norm": 2.577626830585538,
      "language_loss": 0.73181158,
      "learning_rate": 3.5378040326744763e-06,
      "loss": 0.7552284,
      "num_input_tokens_seen": 43366665,
      "step": 2031,
      "time_per_iteration": 2.637068748474121
    },
    {
      "auxiliary_loss_clip": 0.01358184,
      "auxiliary_loss_mlp": 0.01030341,
      "balance_loss_clip": 1.05798578,
      "balance_loss_mlp": 1.02159071,
      "epoch": 0.24433355377863283,
      "flos": 21068575378560.0,
      "grad_norm": 2.127481236335183,
      "language_loss": 0.86031979,
      "learning_rate": 3.5373058675671946e-06,
      "loss": 0.88420498,
      "num_input_tokens_seen": 43384670,
      "step": 2032,
      "time_per_iteration": 2.7284598350524902
    },
    {
      "auxiliary_loss_clip": 0.01408005,
      "auxiliary_loss_mlp": 0.0103152,
      "balance_loss_clip": 1.05329823,
      "balance_loss_mlp": 1.02277637,
      "epoch": 0.24445379666927192,
      "flos": 22637189289600.0,
      "grad_norm": 2.0772509916994375,
      "language_loss": 0.72653651,
      "learning_rate": 3.536807469251836e-06,
      "loss": 0.75093174,
      "num_input_tokens_seen": 43403825,
      "step": 2033,
      "time_per_iteration": 2.7652342319488525
    },
    {
      "auxiliary_loss_clip": 0.01288093,
      "auxiliary_loss_mlp": 0.01029565,
      "balance_loss_clip": 1.05492878,
      "balance_loss_mlp": 1.02048171,
      "epoch": 0.24457403955991103,
      "flos": 21251612108160.0,
      "grad_norm": 1.6904256824664252,
      "language_loss": 0.83272541,
      "learning_rate": 3.5363088378040055e-06,
      "loss": 0.85590196,
      "num_input_tokens_seen": 43422715,
      "step": 2034,
      "time_per_iteration": 2.7444818019866943
    },
    {
      "auxiliary_loss_clip": 0.01110614,
      "auxiliary_loss_mlp": 0.02524748,
      "balance_loss_clip": 1.02908707,
      "balance_loss_mlp": 0.99987918,
      "epoch": 0.2446942824505501,
      "flos": 66997820764800.0,
      "grad_norm": 0.7525979418515761,
      "language_loss": 0.64283001,
      "learning_rate": 3.5358099732993463e-06,
      "loss": 0.6791836,
      "num_input_tokens_seen": 43481825,
      "step": 2035,
      "time_per_iteration": 3.1334311962127686
    },
    {
      "auxiliary_loss_clip": 0.01321297,
      "auxiliary_loss_mlp": 0.0103061,
      "balance_loss_clip": 1.06030846,
      "balance_loss_mlp": 1.02157438,
      "epoch": 0.2448145253411892,
      "flos": 20411122792320.0,
      "grad_norm": 2.0804820566197324,
      "language_loss": 0.896474,
      "learning_rate": 3.535310875813535e-06,
      "loss": 0.91999304,
      "num_input_tokens_seen": 43500220,
      "step": 2036,
      "time_per_iteration": 2.6430015563964844
    },
    {
      "auxiliary_loss_clip": 0.01259242,
      "auxiliary_loss_mlp": 0.01034172,
      "balance_loss_clip": 1.05697691,
      "balance_loss_mlp": 1.02481365,
      "epoch": 0.2449347682318283,
      "flos": 28804739080320.0,
      "grad_norm": 2.345070387204629,
      "language_loss": 0.81579685,
      "learning_rate": 3.5348115454222843e-06,
      "loss": 0.83873093,
      "num_input_tokens_seen": 43522805,
      "step": 2037,
      "time_per_iteration": 2.8425345420837402
    },
    {
      "auxiliary_loss_clip": 0.01311352,
      "auxiliary_loss_mlp": 0.01024902,
      "balance_loss_clip": 1.0552671,
      "balance_loss_mlp": 1.01633704,
      "epoch": 0.2450550111224674,
      "flos": 22528990546560.0,
      "grad_norm": 2.1641272436421732,
      "language_loss": 0.86343288,
      "learning_rate": 3.5343119822013425e-06,
      "loss": 0.88679546,
      "num_input_tokens_seen": 43541915,
      "step": 2038,
      "time_per_iteration": 2.695773124694824
    },
    {
      "auxiliary_loss_clip": 0.01270193,
      "auxiliary_loss_mlp": 0.01039053,
      "balance_loss_clip": 1.06030273,
      "balance_loss_mlp": 1.02937317,
      "epoch": 0.24517525401310647,
      "flos": 21759137326080.0,
      "grad_norm": 1.9116681504461093,
      "language_loss": 0.77666271,
      "learning_rate": 3.533812186226493e-06,
      "loss": 0.7997551,
      "num_input_tokens_seen": 43562625,
      "step": 2039,
      "time_per_iteration": 2.7093608379364014
    },
    {
      "auxiliary_loss_clip": 0.01210774,
      "auxiliary_loss_mlp": 0.01031852,
      "balance_loss_clip": 1.05957568,
      "balance_loss_mlp": 1.02365041,
      "epoch": 0.24529549690374555,
      "flos": 25043311687680.0,
      "grad_norm": 1.8954267613516815,
      "language_loss": 0.76008409,
      "learning_rate": 3.5333121575735545e-06,
      "loss": 0.78251034,
      "num_input_tokens_seen": 43582265,
      "step": 2040,
      "time_per_iteration": 3.978466272354126
    },
    {
      "auxiliary_loss_clip": 0.01316516,
      "auxiliary_loss_mlp": 0.01037777,
      "balance_loss_clip": 1.06169999,
      "balance_loss_mlp": 1.0292536,
      "epoch": 0.24541573979438466,
      "flos": 32123638915200.0,
      "grad_norm": 2.1234129530789194,
      "language_loss": 0.75567067,
      "learning_rate": 3.532811896318381e-06,
      "loss": 0.77921355,
      "num_input_tokens_seen": 43604335,
      "step": 2041,
      "time_per_iteration": 3.6529383659362793
    },
    {
      "auxiliary_loss_clip": 0.01367139,
      "auxiliary_loss_mlp": 0.01037759,
      "balance_loss_clip": 1.05795801,
      "balance_loss_mlp": 1.02789426,
      "epoch": 0.24553598268502375,
      "flos": 31357556622720.0,
      "grad_norm": 3.9333314974963347,
      "language_loss": 0.81498367,
      "learning_rate": 3.5323114025368615e-06,
      "loss": 0.83903265,
      "num_input_tokens_seen": 43619400,
      "step": 2042,
      "time_per_iteration": 2.8374757766723633
    },
    {
      "auxiliary_loss_clip": 0.01260902,
      "auxiliary_loss_mlp": 0.01034766,
      "balance_loss_clip": 1.05554879,
      "balance_loss_mlp": 1.02573001,
      "epoch": 0.24565622557566283,
      "flos": 14027462824320.0,
      "grad_norm": 2.389929338981685,
      "language_loss": 0.81738389,
      "learning_rate": 3.53181067630492e-06,
      "loss": 0.84034055,
      "num_input_tokens_seen": 43636870,
      "step": 2043,
      "time_per_iteration": 2.6508355140686035
    },
    {
      "auxiliary_loss_clip": 0.01307375,
      "auxiliary_loss_mlp": 0.01036737,
      "balance_loss_clip": 1.05759656,
      "balance_loss_mlp": 1.02748013,
      "epoch": 0.24577646846630194,
      "flos": 16581465515520.0,
      "grad_norm": 2.600404985603515,
      "language_loss": 0.75845599,
      "learning_rate": 3.5313097176985175e-06,
      "loss": 0.78189707,
      "num_input_tokens_seen": 43655180,
      "step": 2044,
      "time_per_iteration": 2.6776084899902344
    },
    {
      "auxiliary_loss_clip": 0.01265244,
      "auxiliary_loss_mlp": 0.01030754,
      "balance_loss_clip": 1.06186438,
      "balance_loss_mlp": 1.02203989,
      "epoch": 0.24589671135694102,
      "flos": 18807424272000.0,
      "grad_norm": 1.747399534571349,
      "language_loss": 0.81243813,
      "learning_rate": 3.5308085267936482e-06,
      "loss": 0.83539808,
      "num_input_tokens_seen": 43672895,
      "step": 2045,
      "time_per_iteration": 3.5886030197143555
    },
    {
      "auxiliary_loss_clip": 0.01379747,
      "auxiliary_loss_mlp": 0.025748,
      "balance_loss_clip": 1.05474663,
      "balance_loss_mlp": 1.0002563,
      "epoch": 0.2460169542475801,
      "flos": 19938538529280.0,
      "grad_norm": 2.387163053043289,
      "language_loss": 0.8999666,
      "learning_rate": 3.530307103666342e-06,
      "loss": 0.93951201,
      "num_input_tokens_seen": 43691975,
      "step": 2046,
      "time_per_iteration": 2.801833391189575
    },
    {
      "auxiliary_loss_clip": 0.01289242,
      "auxiliary_loss_mlp": 0.01037837,
      "balance_loss_clip": 1.05882597,
      "balance_loss_mlp": 1.02866995,
      "epoch": 0.24613719713821922,
      "flos": 24171221381760.0,
      "grad_norm": 1.9452759067769267,
      "language_loss": 0.80524856,
      "learning_rate": 3.5298054483926658e-06,
      "loss": 0.82851934,
      "num_input_tokens_seen": 43712670,
      "step": 2047,
      "time_per_iteration": 2.756282329559326
    },
    {
      "auxiliary_loss_clip": 0.01272458,
      "auxiliary_loss_mlp": 0.01037605,
      "balance_loss_clip": 1.06160808,
      "balance_loss_mlp": 1.02802682,
      "epoch": 0.2462574400288583,
      "flos": 30221055325440.0,
      "grad_norm": 2.2521245691047396,
      "language_loss": 0.82735711,
      "learning_rate": 3.5293035610487187e-06,
      "loss": 0.85045779,
      "num_input_tokens_seen": 43732035,
      "step": 2048,
      "time_per_iteration": 2.7086665630340576
    },
    {
      "auxiliary_loss_clip": 0.0121328,
      "auxiliary_loss_mlp": 0.01017324,
      "balance_loss_clip": 1.02761531,
      "balance_loss_mlp": 1.01607251,
      "epoch": 0.24637768291949738,
      "flos": 68943030819840.0,
      "grad_norm": 0.725926517229073,
      "language_loss": 0.61961913,
      "learning_rate": 3.5288014417106374e-06,
      "loss": 0.64192522,
      "num_input_tokens_seen": 43798055,
      "step": 2049,
      "time_per_iteration": 3.3027215003967285
    },
    {
      "auxiliary_loss_clip": 0.01362284,
      "auxiliary_loss_mlp": 0.01030611,
      "balance_loss_clip": 1.05705214,
      "balance_loss_mlp": 1.02165222,
      "epoch": 0.24649792581013646,
      "flos": 34383999922560.0,
      "grad_norm": 1.9391540580481164,
      "language_loss": 0.76025289,
      "learning_rate": 3.528299090454593e-06,
      "loss": 0.78418183,
      "num_input_tokens_seen": 43818590,
      "step": 2050,
      "time_per_iteration": 3.712082624435425
    },
    {
      "auxiliary_loss_clip": 0.01265602,
      "auxiliary_loss_mlp": 0.01036677,
      "balance_loss_clip": 1.05806434,
      "balance_loss_mlp": 1.02678216,
      "epoch": 0.24661816870077558,
      "flos": 19680448331520.0,
      "grad_norm": 2.433731197839955,
      "language_loss": 0.82634252,
      "learning_rate": 3.527796507356792e-06,
      "loss": 0.84936529,
      "num_input_tokens_seen": 43832480,
      "step": 2051,
      "time_per_iteration": 2.8650405406951904
    },
    {
      "auxiliary_loss_clip": 0.01187206,
      "auxiliary_loss_mlp": 0.01033494,
      "balance_loss_clip": 1.06058455,
      "balance_loss_mlp": 1.02498877,
      "epoch": 0.24673841159141466,
      "flos": 20002279213440.0,
      "grad_norm": 3.5796500636831614,
      "language_loss": 0.89902198,
      "learning_rate": 3.527293692493475e-06,
      "loss": 0.921229,
      "num_input_tokens_seen": 43848345,
      "step": 2052,
      "time_per_iteration": 2.6718039512634277
    },
    {
      "auxiliary_loss_clip": 0.01268805,
      "auxiliary_loss_mlp": 0.01034749,
      "balance_loss_clip": 1.05908835,
      "balance_loss_mlp": 1.02586162,
      "epoch": 0.24685865448205374,
      "flos": 21646593037440.0,
      "grad_norm": 3.0697322782814473,
      "language_loss": 0.73400229,
      "learning_rate": 3.52679064594092e-06,
      "loss": 0.75703788,
      "num_input_tokens_seen": 43865685,
      "step": 2053,
      "time_per_iteration": 2.6493000984191895
    },
    {
      "auxiliary_loss_clip": 0.01405501,
      "auxiliary_loss_mlp": 0.01034306,
      "balance_loss_clip": 1.04668212,
      "balance_loss_mlp": 1.02524567,
      "epoch": 0.24697889737269285,
      "flos": 17960470508160.0,
      "grad_norm": 2.656216529934147,
      "language_loss": 0.74858856,
      "learning_rate": 3.5262873677754375e-06,
      "loss": 0.77298653,
      "num_input_tokens_seen": 43883690,
      "step": 2054,
      "time_per_iteration": 2.833987236022949
    },
    {
      "auxiliary_loss_clip": 0.01215841,
      "auxiliary_loss_mlp": 0.01030403,
      "balance_loss_clip": 1.06273711,
      "balance_loss_mlp": 1.02198696,
      "epoch": 0.24709914026333193,
      "flos": 27344611221120.0,
      "grad_norm": 2.3586044359649803,
      "language_loss": 0.80327135,
      "learning_rate": 3.5257838580733745e-06,
      "loss": 0.82573378,
      "num_input_tokens_seen": 43903295,
      "step": 2055,
      "time_per_iteration": 2.614351511001587
    },
    {
      "auxiliary_loss_clip": 0.01271289,
      "auxiliary_loss_mlp": 0.01029563,
      "balance_loss_clip": 1.06232548,
      "balance_loss_mlp": 1.02093208,
      "epoch": 0.24721938315397102,
      "flos": 19275519335040.0,
      "grad_norm": 1.8678067558067382,
      "language_loss": 0.872558,
      "learning_rate": 3.5252801169111138e-06,
      "loss": 0.89556658,
      "num_input_tokens_seen": 43920960,
      "step": 2056,
      "time_per_iteration": 2.6933176517486572
    },
    {
      "auxiliary_loss_clip": 0.01311357,
      "auxiliary_loss_mlp": 0.01032143,
      "balance_loss_clip": 1.06051016,
      "balance_loss_mlp": 1.02333951,
      "epoch": 0.2473396260446101,
      "flos": 23185796688000.0,
      "grad_norm": 1.881456010199446,
      "language_loss": 0.80060267,
      "learning_rate": 3.524776144365072e-06,
      "loss": 0.82403773,
      "num_input_tokens_seen": 43939415,
      "step": 2057,
      "time_per_iteration": 2.70794939994812
    },
    {
      "auxiliary_loss_clip": 0.01310598,
      "auxiliary_loss_mlp": 0.01039729,
      "balance_loss_clip": 1.06386697,
      "balance_loss_mlp": 1.03139591,
      "epoch": 0.2474598689352492,
      "flos": 21142443697920.0,
      "grad_norm": 1.5915427943763425,
      "language_loss": 0.79380101,
      "learning_rate": 3.5242719405117016e-06,
      "loss": 0.81730425,
      "num_input_tokens_seen": 43959220,
      "step": 2058,
      "time_per_iteration": 2.7502689361572266
    },
    {
      "auxiliary_loss_clip": 0.01325637,
      "auxiliary_loss_mlp": 0.02577951,
      "balance_loss_clip": 1.06519699,
      "balance_loss_mlp": 1.00025189,
      "epoch": 0.2475801118258883,
      "flos": 21648352803840.0,
      "grad_norm": 13.530613666607833,
      "language_loss": 0.751293,
      "learning_rate": 3.5237675054274893e-06,
      "loss": 0.79032892,
      "num_input_tokens_seen": 43978420,
      "step": 2059,
      "time_per_iteration": 2.7842822074890137
    },
    {
      "auxiliary_loss_clip": 0.01265437,
      "auxiliary_loss_mlp": 0.01035208,
      "balance_loss_clip": 1.0591104,
      "balance_loss_mlp": 1.02621365,
      "epoch": 0.24770035471652738,
      "flos": 22674500542080.0,
      "grad_norm": 1.9006991718800461,
      "language_loss": 0.80257404,
      "learning_rate": 3.5232628391889584e-06,
      "loss": 0.82558048,
      "num_input_tokens_seen": 43996710,
      "step": 2060,
      "time_per_iteration": 2.6563146114349365
    },
    {
      "auxiliary_loss_clip": 0.01406328,
      "auxiliary_loss_mlp": 0.01036875,
      "balance_loss_clip": 1.05485487,
      "balance_loss_mlp": 1.02862573,
      "epoch": 0.2478205976071665,
      "flos": 22163814927360.0,
      "grad_norm": 3.398369017724221,
      "language_loss": 0.64552951,
      "learning_rate": 3.522757941872666e-06,
      "loss": 0.66996145,
      "num_input_tokens_seen": 44014865,
      "step": 2061,
      "time_per_iteration": 2.80312180519104
    },
    {
      "auxiliary_loss_clip": 0.01216729,
      "auxiliary_loss_mlp": 0.02576031,
      "balance_loss_clip": 1.06388235,
      "balance_loss_mlp": 1.00015664,
      "epoch": 0.24794084049780557,
      "flos": 24973106555520.0,
      "grad_norm": 1.8342288610258561,
      "language_loss": 0.82701504,
      "learning_rate": 3.5222528135552042e-06,
      "loss": 0.86494255,
      "num_input_tokens_seen": 44036325,
      "step": 2062,
      "time_per_iteration": 2.6286051273345947
    },
    {
      "auxiliary_loss_clip": 0.01263599,
      "auxiliary_loss_mlp": 0.0103872,
      "balance_loss_clip": 1.06345177,
      "balance_loss_mlp": 1.02899241,
      "epoch": 0.24806108338844465,
      "flos": 18296379521280.0,
      "grad_norm": 1.884168704698469,
      "language_loss": 0.80819267,
      "learning_rate": 3.521747454313201e-06,
      "loss": 0.8312158,
      "num_input_tokens_seen": 44055005,
      "step": 2063,
      "time_per_iteration": 2.966130018234253
    },
    {
      "auxiliary_loss_clip": 0.01356995,
      "auxiliary_loss_mlp": 0.0104313,
      "balance_loss_clip": 1.05336332,
      "balance_loss_mlp": 1.03365314,
      "epoch": 0.24818132627908374,
      "flos": 19282163351040.0,
      "grad_norm": 2.2550812402142735,
      "language_loss": 0.66750771,
      "learning_rate": 3.521241864223319e-06,
      "loss": 0.69150895,
      "num_input_tokens_seen": 44073965,
      "step": 2064,
      "time_per_iteration": 2.7609410285949707
    },
    {
      "auxiliary_loss_clip": 0.01229684,
      "auxiliary_loss_mlp": 0.01005501,
      "balance_loss_clip": 1.03521538,
      "balance_loss_mlp": 1.00415981,
      "epoch": 0.24830156916972285,
      "flos": 70285837881600.0,
      "grad_norm": 0.7905224995353267,
      "language_loss": 0.61910224,
      "learning_rate": 3.5207360433622552e-06,
      "loss": 0.6414541,
      "num_input_tokens_seen": 44135965,
      "step": 2065,
      "time_per_iteration": 3.2126245498657227
    },
    {
      "auxiliary_loss_clip": 0.01307836,
      "auxiliary_loss_mlp": 0.01038162,
      "balance_loss_clip": 1.05751967,
      "balance_loss_mlp": 1.02943039,
      "epoch": 0.24842181206036193,
      "flos": 40409128287360.0,
      "grad_norm": 1.8369356770645453,
      "language_loss": 0.74975824,
      "learning_rate": 3.5202299918067437e-06,
      "loss": 0.77321815,
      "num_input_tokens_seen": 44159560,
      "step": 2066,
      "time_per_iteration": 3.8206698894500732
    },
    {
      "auxiliary_loss_clip": 0.01266233,
      "auxiliary_loss_mlp": 0.01036264,
      "balance_loss_clip": 1.06242728,
      "balance_loss_mlp": 1.02756763,
      "epoch": 0.248542054951001,
      "flos": 20082432412800.0,
      "grad_norm": 2.271287958378783,
      "language_loss": 0.69314259,
      "learning_rate": 3.519723709633551e-06,
      "loss": 0.71616757,
      "num_input_tokens_seen": 44178320,
      "step": 2067,
      "time_per_iteration": 3.4979257583618164
    },
    {
      "auxiliary_loss_clip": 0.01309233,
      "auxiliary_loss_mlp": 0.01036134,
      "balance_loss_clip": 1.05864716,
      "balance_loss_mlp": 1.02747929,
      "epoch": 0.24866229784164012,
      "flos": 23513948363520.0,
      "grad_norm": 3.596672403844209,
      "language_loss": 0.83592469,
      "learning_rate": 3.519217196919479e-06,
      "loss": 0.8593784,
      "num_input_tokens_seen": 44197305,
      "step": 2068,
      "time_per_iteration": 2.772091865539551
    },
    {
      "auxiliary_loss_clip": 0.01320331,
      "auxiliary_loss_mlp": 0.01039726,
      "balance_loss_clip": 1.06511903,
      "balance_loss_mlp": 1.03062487,
      "epoch": 0.2487825407322792,
      "flos": 19865101173120.0,
      "grad_norm": 2.040256574424427,
      "language_loss": 0.7362448,
      "learning_rate": 3.518710453741367e-06,
      "loss": 0.75984538,
      "num_input_tokens_seen": 44216505,
      "step": 2069,
      "time_per_iteration": 2.7082159519195557
    },
    {
      "auxiliary_loss_clip": 0.01306977,
      "auxiliary_loss_mlp": 0.02577202,
      "balance_loss_clip": 1.05878472,
      "balance_loss_mlp": 1.00013542,
      "epoch": 0.2489027836229183,
      "flos": 22017622573440.0,
      "grad_norm": 2.232891729698551,
      "language_loss": 0.68204629,
      "learning_rate": 3.518203480176086e-06,
      "loss": 0.72088808,
      "num_input_tokens_seen": 44235435,
      "step": 2070,
      "time_per_iteration": 2.714195489883423
    },
    {
      "auxiliary_loss_clip": 0.01515056,
      "auxiliary_loss_mlp": 0.01038182,
      "balance_loss_clip": 1.04883313,
      "balance_loss_mlp": 1.02938437,
      "epoch": 0.2490230265135574,
      "flos": 23294354567040.0,
      "grad_norm": 1.7380605899636838,
      "language_loss": 0.80938196,
      "learning_rate": 3.517696276300545e-06,
      "loss": 0.83491433,
      "num_input_tokens_seen": 44256975,
      "step": 2071,
      "time_per_iteration": 3.9173824787139893
    },
    {
      "auxiliary_loss_clip": 0.01267178,
      "auxiliary_loss_mlp": 0.01037671,
      "balance_loss_clip": 1.06422997,
      "balance_loss_mlp": 1.02857506,
      "epoch": 0.24914326940419648,
      "flos": 19826784339840.0,
      "grad_norm": 2.7710384864394872,
      "language_loss": 0.69061017,
      "learning_rate": 3.517188842191685e-06,
      "loss": 0.71365869,
      "num_input_tokens_seen": 44275125,
      "step": 2072,
      "time_per_iteration": 2.7400810718536377
    },
    {
      "auxiliary_loss_clip": 0.01263235,
      "auxiliary_loss_mlp": 0.01038811,
      "balance_loss_clip": 1.05906463,
      "balance_loss_mlp": 1.02979326,
      "epoch": 0.24926351229483557,
      "flos": 20229271211520.0,
      "grad_norm": 1.6441911484839018,
      "language_loss": 0.74405682,
      "learning_rate": 3.5166811779264837e-06,
      "loss": 0.76707733,
      "num_input_tokens_seen": 44295445,
      "step": 2073,
      "time_per_iteration": 2.7787890434265137
    },
    {
      "auxiliary_loss_clip": 0.01214691,
      "auxiliary_loss_mlp": 0.01037386,
      "balance_loss_clip": 1.06024885,
      "balance_loss_mlp": 1.0280931,
      "epoch": 0.24938375518547465,
      "flos": 23294570048640.0,
      "grad_norm": 2.931356372241926,
      "language_loss": 0.78199726,
      "learning_rate": 3.5161732835819545e-06,
      "loss": 0.80451804,
      "num_input_tokens_seen": 44314755,
      "step": 2074,
      "time_per_iteration": 2.659229278564453
    },
    {
      "auxiliary_loss_clip": 0.01217002,
      "auxiliary_loss_mlp": 0.01033564,
      "balance_loss_clip": 1.06382632,
      "balance_loss_mlp": 1.02474248,
      "epoch": 0.24950399807611376,
      "flos": 17311673099520.0,
      "grad_norm": 2.5786485815105347,
      "language_loss": 0.83692467,
      "learning_rate": 3.515665159235143e-06,
      "loss": 0.85943037,
      "num_input_tokens_seen": 44333640,
      "step": 2075,
      "time_per_iteration": 2.5719804763793945
    },
    {
      "auxiliary_loss_clip": 0.01313583,
      "auxiliary_loss_mlp": 0.01036494,
      "balance_loss_clip": 1.05682135,
      "balance_loss_mlp": 1.02792275,
      "epoch": 0.24962424096675284,
      "flos": 19024863252480.0,
      "grad_norm": 1.6004943147790314,
      "language_loss": 0.75116974,
      "learning_rate": 3.5151568049631318e-06,
      "loss": 0.77467048,
      "num_input_tokens_seen": 44352355,
      "step": 2076,
      "time_per_iteration": 3.5829272270202637
    },
    {
      "auxiliary_loss_clip": 0.01216331,
      "auxiliary_loss_mlp": 0.0103621,
      "balance_loss_clip": 1.0626719,
      "balance_loss_mlp": 1.02746558,
      "epoch": 0.24974448385739192,
      "flos": 33398790710400.0,
      "grad_norm": 1.8209681672205338,
      "language_loss": 0.80366945,
      "learning_rate": 3.5146482208430385e-06,
      "loss": 0.82619488,
      "num_input_tokens_seen": 44374185,
      "step": 2077,
      "time_per_iteration": 2.6953911781311035
    },
    {
      "auxiliary_loss_clip": 0.01451232,
      "auxiliary_loss_mlp": 0.01042046,
      "balance_loss_clip": 1.0471431,
      "balance_loss_mlp": 1.03296268,
      "epoch": 0.24986472674803104,
      "flos": 30007279532160.0,
      "grad_norm": 1.9934084007231414,
      "language_loss": 0.6764158,
      "learning_rate": 3.514139406952014e-06,
      "loss": 0.70134854,
      "num_input_tokens_seen": 44396210,
      "step": 2078,
      "time_per_iteration": 2.9102559089660645
    },
    {
      "auxiliary_loss_clip": 0.01262839,
      "auxiliary_loss_mlp": 0.0103704,
      "balance_loss_clip": 1.06114376,
      "balance_loss_mlp": 1.02856374,
      "epoch": 0.24998496963867012,
      "flos": 26613074833920.0,
      "grad_norm": 1.819495316238732,
      "language_loss": 0.83178645,
      "learning_rate": 3.5136303633672454e-06,
      "loss": 0.85478526,
      "num_input_tokens_seen": 44416340,
      "step": 2079,
      "time_per_iteration": 2.685487985610962
    },
    {
      "auxiliary_loss_clip": 0.01373109,
      "auxiliary_loss_mlp": 0.02577556,
      "balance_loss_clip": 1.05872834,
      "balance_loss_mlp": 1.00014973,
      "epoch": 0.25010521252930923,
      "flos": 23553989049600.0,
      "grad_norm": 1.984031145120208,
      "language_loss": 0.7449106,
      "learning_rate": 3.5131210901659544e-06,
      "loss": 0.78441727,
      "num_input_tokens_seen": 44438095,
      "step": 2080,
      "time_per_iteration": 2.7983696460723877
    },
    {
      "auxiliary_loss_clip": 0.0135926,
      "auxiliary_loss_mlp": 0.01030811,
      "balance_loss_clip": 1.05435407,
      "balance_loss_mlp": 1.02195418,
      "epoch": 0.2502254554199483,
      "flos": 23441193365760.0,
      "grad_norm": 2.5604702567001403,
      "language_loss": 0.82191527,
      "learning_rate": 3.5126115874253967e-06,
      "loss": 0.84581602,
      "num_input_tokens_seen": 44457650,
      "step": 2081,
      "time_per_iteration": 2.787261486053467
    },
    {
      "auxiliary_loss_clip": 0.01362858,
      "auxiliary_loss_mlp": 0.01034645,
      "balance_loss_clip": 1.06048417,
      "balance_loss_mlp": 1.02560043,
      "epoch": 0.2503456983105874,
      "flos": 28761681651840.0,
      "grad_norm": 2.275655687113159,
      "language_loss": 0.81241137,
      "learning_rate": 3.5121018552228644e-06,
      "loss": 0.83638638,
      "num_input_tokens_seen": 44476155,
      "step": 2082,
      "time_per_iteration": 2.72613525390625
    },
    {
      "auxiliary_loss_clip": 0.01358166,
      "auxiliary_loss_mlp": 0.01035561,
      "balance_loss_clip": 1.05580187,
      "balance_loss_mlp": 1.02657247,
      "epoch": 0.2504659412012265,
      "flos": 18770256673920.0,
      "grad_norm": 2.2809020667008215,
      "language_loss": 0.76461506,
      "learning_rate": 3.5115918936356827e-06,
      "loss": 0.7885524,
      "num_input_tokens_seen": 44492910,
      "step": 2083,
      "time_per_iteration": 2.704159736633301
    },
    {
      "auxiliary_loss_clip": 0.01347205,
      "auxiliary_loss_mlp": 0.01043672,
      "balance_loss_clip": 1.05248833,
      "balance_loss_mlp": 1.03479671,
      "epoch": 0.25058618409186556,
      "flos": 16873383346560.0,
      "grad_norm": 3.3356737692011627,
      "language_loss": 0.78992558,
      "learning_rate": 3.5110817027412123e-06,
      "loss": 0.81383431,
      "num_input_tokens_seen": 44512000,
      "step": 2084,
      "time_per_iteration": 2.7328057289123535
    },
    {
      "auxiliary_loss_clip": 0.01355304,
      "auxiliary_loss_mlp": 0.0103647,
      "balance_loss_clip": 1.05081332,
      "balance_loss_mlp": 1.02788138,
      "epoch": 0.25070642698250467,
      "flos": 24425540651520.0,
      "grad_norm": 7.720349247777,
      "language_loss": 0.69066954,
      "learning_rate": 3.5105712826168493e-06,
      "loss": 0.71458733,
      "num_input_tokens_seen": 44531650,
      "step": 2085,
      "time_per_iteration": 2.7854232788085938
    },
    {
      "auxiliary_loss_clip": 0.01262027,
      "auxiliary_loss_mlp": 0.02573487,
      "balance_loss_clip": 1.05865371,
      "balance_loss_mlp": 1.00013638,
      "epoch": 0.2508266698731437,
      "flos": 20260944028800.0,
      "grad_norm": 2.0461252280800974,
      "language_loss": 0.70707971,
      "learning_rate": 3.5100606333400235e-06,
      "loss": 0.74543488,
      "num_input_tokens_seen": 44548785,
      "step": 2086,
      "time_per_iteration": 2.8493502140045166
    },
    {
      "auxiliary_loss_clip": 0.01324959,
      "auxiliary_loss_mlp": 0.01040409,
      "balance_loss_clip": 1.05917573,
      "balance_loss_mlp": 1.03066349,
      "epoch": 0.25094691276378284,
      "flos": 19245318975360.0,
      "grad_norm": 2.381979659880762,
      "language_loss": 0.77453971,
      "learning_rate": 3.5095497549882006e-06,
      "loss": 0.79819334,
      "num_input_tokens_seen": 44567230,
      "step": 2087,
      "time_per_iteration": 2.706061840057373
    },
    {
      "auxiliary_loss_clip": 0.01262505,
      "auxiliary_loss_mlp": 0.01032039,
      "balance_loss_clip": 1.06166458,
      "balance_loss_mlp": 1.02312803,
      "epoch": 0.25106715565442195,
      "flos": 26943237671040.0,
      "grad_norm": 2.5951480949247347,
      "language_loss": 0.72836936,
      "learning_rate": 3.50903864763888e-06,
      "loss": 0.75131482,
      "num_input_tokens_seen": 44588020,
      "step": 2088,
      "time_per_iteration": 2.6468725204467773
    },
    {
      "auxiliary_loss_clip": 0.01267711,
      "auxiliary_loss_mlp": 0.01033946,
      "balance_loss_clip": 1.06054032,
      "balance_loss_mlp": 1.02467763,
      "epoch": 0.251187398545061,
      "flos": 48359570572800.0,
      "grad_norm": 3.341761868234344,
      "language_loss": 0.75963843,
      "learning_rate": 3.5085273113695965e-06,
      "loss": 0.782655,
      "num_input_tokens_seen": 44612590,
      "step": 2089,
      "time_per_iteration": 3.0599327087402344
    },
    {
      "auxiliary_loss_clip": 0.01215827,
      "auxiliary_loss_mlp": 0.0103577,
      "balance_loss_clip": 1.06173551,
      "balance_loss_mlp": 1.02600074,
      "epoch": 0.2513076414357001,
      "flos": 27016100409600.0,
      "grad_norm": 5.370533109864265,
      "language_loss": 0.78668475,
      "learning_rate": 3.508015746257919e-06,
      "loss": 0.80920076,
      "num_input_tokens_seen": 44631630,
      "step": 2090,
      "time_per_iteration": 2.658648729324341
    },
    {
      "auxiliary_loss_clip": 0.01367281,
      "auxiliary_loss_mlp": 0.01034568,
      "balance_loss_clip": 1.05711246,
      "balance_loss_mlp": 1.02547836,
      "epoch": 0.2514278843263392,
      "flos": 19463619882240.0,
      "grad_norm": 2.365317586953289,
      "language_loss": 0.83471346,
      "learning_rate": 3.5075039523814518e-06,
      "loss": 0.85873193,
      "num_input_tokens_seen": 44650820,
      "step": 2091,
      "time_per_iteration": 3.559981107711792
    },
    {
      "auxiliary_loss_clip": 0.01269705,
      "auxiliary_loss_mlp": 0.01036562,
      "balance_loss_clip": 1.05855513,
      "balance_loss_mlp": 1.02684093,
      "epoch": 0.2515481272169783,
      "flos": 16866092885760.0,
      "grad_norm": 2.5464428688938465,
      "language_loss": 0.81914479,
      "learning_rate": 3.506991929817834e-06,
      "loss": 0.84220743,
      "num_input_tokens_seen": 44667540,
      "step": 2092,
      "time_per_iteration": 2.6777875423431396
    },
    {
      "auxiliary_loss_clip": 0.01214903,
      "auxiliary_loss_mlp": 0.0103677,
      "balance_loss_clip": 1.06559289,
      "balance_loss_mlp": 1.02812123,
      "epoch": 0.2516683701076174,
      "flos": 23732464752000.0,
      "grad_norm": 1.7995155141542611,
      "language_loss": 0.82720071,
      "learning_rate": 3.506479678644738e-06,
      "loss": 0.84971744,
      "num_input_tokens_seen": 44687935,
      "step": 2093,
      "time_per_iteration": 3.496401071548462
    },
    {
      "auxiliary_loss_clip": 0.01404798,
      "auxiliary_loss_mlp": 0.01036301,
      "balance_loss_clip": 1.05598974,
      "balance_loss_mlp": 1.02740157,
      "epoch": 0.2517886129982565,
      "flos": 27635954434560.0,
      "grad_norm": 10.322329295364929,
      "language_loss": 0.74012482,
      "learning_rate": 3.505967198939873e-06,
      "loss": 0.76453578,
      "num_input_tokens_seen": 44704975,
      "step": 2094,
      "time_per_iteration": 2.777081251144409
    },
    {
      "auxiliary_loss_clip": 0.01309739,
      "auxiliary_loss_mlp": 0.01027965,
      "balance_loss_clip": 1.05637324,
      "balance_loss_mlp": 1.0193696,
      "epoch": 0.25190885588889556,
      "flos": 38104596529920.0,
      "grad_norm": 3.519736407577444,
      "language_loss": 0.7864567,
      "learning_rate": 3.5054544907809813e-06,
      "loss": 0.80983377,
      "num_input_tokens_seen": 44725475,
      "step": 2095,
      "time_per_iteration": 2.840287208557129
    },
    {
      "auxiliary_loss_clip": 0.0131405,
      "auxiliary_loss_mlp": 0.02578828,
      "balance_loss_clip": 1.06157148,
      "balance_loss_mlp": 1.00014591,
      "epoch": 0.25202909877953467,
      "flos": 22269894768000.0,
      "grad_norm": 2.0835205638558203,
      "language_loss": 0.80542094,
      "learning_rate": 3.50494155424584e-06,
      "loss": 0.84434968,
      "num_input_tokens_seen": 44744380,
      "step": 2096,
      "time_per_iteration": 2.691432237625122
    },
    {
      "auxiliary_loss_clip": 0.01269231,
      "auxiliary_loss_mlp": 0.01037925,
      "balance_loss_clip": 1.06118941,
      "balance_loss_mlp": 1.02913332,
      "epoch": 0.2521493416701738,
      "flos": 21761759018880.0,
      "grad_norm": 4.383030156319143,
      "language_loss": 0.83278072,
      "learning_rate": 3.504428389412262e-06,
      "loss": 0.85585225,
      "num_input_tokens_seen": 44765190,
      "step": 2097,
      "time_per_iteration": 3.5834972858428955
    },
    {
      "auxiliary_loss_clip": 0.01264733,
      "auxiliary_loss_mlp": 0.01034409,
      "balance_loss_clip": 1.05970764,
      "balance_loss_mlp": 1.02519965,
      "epoch": 0.25226958456081283,
      "flos": 27746738956800.0,
      "grad_norm": 1.9332186087415986,
      "language_loss": 0.73045325,
      "learning_rate": 3.5039149963580927e-06,
      "loss": 0.75344467,
      "num_input_tokens_seen": 44785210,
      "step": 2098,
      "time_per_iteration": 2.6844053268432617
    },
    {
      "auxiliary_loss_clip": 0.01308548,
      "auxiliary_loss_mlp": 0.01027493,
      "balance_loss_clip": 1.06223452,
      "balance_loss_mlp": 1.01862955,
      "epoch": 0.25238982745145194,
      "flos": 30732171903360.0,
      "grad_norm": 2.210299983510216,
      "language_loss": 0.70183492,
      "learning_rate": 3.503401375161215e-06,
      "loss": 0.72519535,
      "num_input_tokens_seen": 44804955,
      "step": 2099,
      "time_per_iteration": 2.7357375621795654
    },
    {
      "auxiliary_loss_clip": 0.01212898,
      "auxiliary_loss_mlp": 0.01032462,
      "balance_loss_clip": 1.06154084,
      "balance_loss_mlp": 1.02339637,
      "epoch": 0.252510070342091,
      "flos": 20266331068800.0,
      "grad_norm": 1.698581230253737,
      "language_loss": 0.83839226,
      "learning_rate": 3.502887525899544e-06,
      "loss": 0.8608458,
      "num_input_tokens_seen": 44823935,
      "step": 2100,
      "time_per_iteration": 2.6516335010528564
    },
    {
      "auxiliary_loss_clip": 0.01315169,
      "auxiliary_loss_mlp": 0.0103238,
      "balance_loss_clip": 1.05985165,
      "balance_loss_mlp": 1.02324879,
      "epoch": 0.2526303132327301,
      "flos": 22747399194240.0,
      "grad_norm": 1.9194913849806892,
      "language_loss": 0.8302691,
      "learning_rate": 3.50237344865103e-06,
      "loss": 0.85374463,
      "num_input_tokens_seen": 44844935,
      "step": 2101,
      "time_per_iteration": 2.732715606689453
    },
    {
      "auxiliary_loss_clip": 0.01214603,
      "auxiliary_loss_mlp": 0.01030996,
      "balance_loss_clip": 1.06175947,
      "balance_loss_mlp": 1.021608,
      "epoch": 0.2527505561233692,
      "flos": 30263466309120.0,
      "grad_norm": 3.222781087631139,
      "language_loss": 0.7661972,
      "learning_rate": 3.501859143493658e-06,
      "loss": 0.78865314,
      "num_input_tokens_seen": 44865565,
      "step": 2102,
      "time_per_iteration": 2.7172234058380127
    },
    {
      "auxiliary_loss_clip": 0.01111341,
      "auxiliary_loss_mlp": 0.01022349,
      "balance_loss_clip": 1.03062022,
      "balance_loss_mlp": 1.02121055,
      "epoch": 0.2528707990140083,
      "flos": 58492917164160.0,
      "grad_norm": 0.9308656554730081,
      "language_loss": 0.60547173,
      "learning_rate": 3.5013446105054488e-06,
      "loss": 0.62680864,
      "num_input_tokens_seen": 44918485,
      "step": 2103,
      "time_per_iteration": 3.869779586791992
    },
    {
      "auxiliary_loss_clip": 0.01349736,
      "auxiliary_loss_mlp": 0.01028391,
      "balance_loss_clip": 1.05434442,
      "balance_loss_mlp": 1.019593,
      "epoch": 0.2529910419046474,
      "flos": 24645134448000.0,
      "grad_norm": 2.4838000500464403,
      "language_loss": 0.7491219,
      "learning_rate": 3.5008298497644555e-06,
      "loss": 0.77290326,
      "num_input_tokens_seen": 44937530,
      "step": 2104,
      "time_per_iteration": 2.7732651233673096
    },
    {
      "auxiliary_loss_clip": 0.01282295,
      "auxiliary_loss_mlp": 0.0102852,
      "balance_loss_clip": 1.05800617,
      "balance_loss_mlp": 1.01984775,
      "epoch": 0.2531112847952865,
      "flos": 23842135952640.0,
      "grad_norm": 1.5680943777453873,
      "language_loss": 0.88067371,
      "learning_rate": 3.500314861348767e-06,
      "loss": 0.90378183,
      "num_input_tokens_seen": 44958165,
      "step": 2105,
      "time_per_iteration": 2.741002321243286
    },
    {
      "auxiliary_loss_clip": 0.01357787,
      "auxiliary_loss_mlp": 0.01031964,
      "balance_loss_clip": 1.05819452,
      "balance_loss_mlp": 1.02327919,
      "epoch": 0.25323152768592555,
      "flos": 16143822207360.0,
      "grad_norm": 3.0069140865927637,
      "language_loss": 0.77396315,
      "learning_rate": 3.499799645336507e-06,
      "loss": 0.79786068,
      "num_input_tokens_seen": 44975060,
      "step": 2106,
      "time_per_iteration": 2.7395431995391846
    },
    {
      "auxiliary_loss_clip": 0.01267866,
      "auxiliary_loss_mlp": 0.01032729,
      "balance_loss_clip": 1.06406355,
      "balance_loss_mlp": 1.02443802,
      "epoch": 0.25335177057656466,
      "flos": 28405161210240.0,
      "grad_norm": 1.806552502484042,
      "language_loss": 0.8708961,
      "learning_rate": 3.4992842018058336e-06,
      "loss": 0.89390206,
      "num_input_tokens_seen": 44997960,
      "step": 2107,
      "time_per_iteration": 2.699760675430298
    },
    {
      "auxiliary_loss_clip": 0.01366258,
      "auxiliary_loss_mlp": 0.0103431,
      "balance_loss_clip": 1.05774283,
      "balance_loss_mlp": 1.02613854,
      "epoch": 0.25347201346720377,
      "flos": 18799666934400.0,
      "grad_norm": 2.347927930181925,
      "language_loss": 0.88648367,
      "learning_rate": 3.4987685308349384e-06,
      "loss": 0.91048938,
      "num_input_tokens_seen": 45015690,
      "step": 2108,
      "time_per_iteration": 2.73976469039917
    },
    {
      "auxiliary_loss_clip": 0.01364231,
      "auxiliary_loss_mlp": 0.01030372,
      "balance_loss_clip": 1.0533793,
      "balance_loss_mlp": 1.02163339,
      "epoch": 0.2535922563578428,
      "flos": 15815490963840.0,
      "grad_norm": 1.9446106240353969,
      "language_loss": 0.61257857,
      "learning_rate": 3.4982526325020497e-06,
      "loss": 0.63652456,
      "num_input_tokens_seen": 45032660,
      "step": 2109,
      "time_per_iteration": 2.6917262077331543
    },
    {
      "auxiliary_loss_clip": 0.01231487,
      "auxiliary_loss_mlp": 0.01036687,
      "balance_loss_clip": 1.05866194,
      "balance_loss_mlp": 1.02841973,
      "epoch": 0.25371249924848194,
      "flos": 16318922031360.0,
      "grad_norm": 3.1136731344595012,
      "language_loss": 0.82037055,
      "learning_rate": 3.4977365068854273e-06,
      "loss": 0.84305227,
      "num_input_tokens_seen": 45048280,
      "step": 2110,
      "time_per_iteration": 2.644906759262085
    },
    {
      "auxiliary_loss_clip": 0.01303609,
      "auxiliary_loss_mlp": 0.01030153,
      "balance_loss_clip": 1.05538356,
      "balance_loss_mlp": 1.02087283,
      "epoch": 0.25383274213912105,
      "flos": 21761615364480.0,
      "grad_norm": 1.7421840485645965,
      "language_loss": 0.73786056,
      "learning_rate": 3.4972201540633676e-06,
      "loss": 0.76119816,
      "num_input_tokens_seen": 45067635,
      "step": 2111,
      "time_per_iteration": 2.7079527378082275
    },
    {
      "auxiliary_loss_clip": 0.01307567,
      "auxiliary_loss_mlp": 0.01032727,
      "balance_loss_clip": 1.05738401,
      "balance_loss_mlp": 1.02396488,
      "epoch": 0.2539529850297601,
      "flos": 21396870708480.0,
      "grad_norm": 2.001815661947714,
      "language_loss": 0.85154188,
      "learning_rate": 3.4967035741142008e-06,
      "loss": 0.87494487,
      "num_input_tokens_seen": 45086455,
      "step": 2112,
      "time_per_iteration": 2.7456154823303223
    },
    {
      "auxiliary_loss_clip": 0.01312543,
      "auxiliary_loss_mlp": 0.01031012,
      "balance_loss_clip": 1.06704307,
      "balance_loss_mlp": 1.02301931,
      "epoch": 0.2540732279203992,
      "flos": 25228467319680.0,
      "grad_norm": 1.9320800262121287,
      "language_loss": 0.81844884,
      "learning_rate": 3.4961867671162917e-06,
      "loss": 0.84188437,
      "num_input_tokens_seen": 45106385,
      "step": 2113,
      "time_per_iteration": 2.8515405654907227
    },
    {
      "auxiliary_loss_clip": 0.01216944,
      "auxiliary_loss_mlp": 0.01026929,
      "balance_loss_clip": 1.06266534,
      "balance_loss_mlp": 1.01783895,
      "epoch": 0.2541934708110383,
      "flos": 19427386037760.0,
      "grad_norm": 3.222156962600459,
      "language_loss": 0.77175558,
      "learning_rate": 3.4956697331480402e-06,
      "loss": 0.79419422,
      "num_input_tokens_seen": 45124955,
      "step": 2114,
      "time_per_iteration": 2.629101276397705
    },
    {
      "auxiliary_loss_clip": 0.01280014,
      "auxiliary_loss_mlp": 0.01033114,
      "balance_loss_clip": 1.05560708,
      "balance_loss_mlp": 1.02471566,
      "epoch": 0.2543137137016774,
      "flos": 23949436855680.0,
      "grad_norm": 2.4477968651293134,
      "language_loss": 0.8010484,
      "learning_rate": 3.495152472287879e-06,
      "loss": 0.82417965,
      "num_input_tokens_seen": 45145665,
      "step": 2115,
      "time_per_iteration": 2.8409862518310547
    },
    {
      "auxiliary_loss_clip": 0.01362869,
      "auxiliary_loss_mlp": 0.01031495,
      "balance_loss_clip": 1.05951631,
      "balance_loss_mlp": 1.02268529,
      "epoch": 0.2544339565923165,
      "flos": 25593283802880.0,
      "grad_norm": 2.1068506950526804,
      "language_loss": 0.73762524,
      "learning_rate": 3.4946349846142766e-06,
      "loss": 0.76156878,
      "num_input_tokens_seen": 45164805,
      "step": 2116,
      "time_per_iteration": 2.777482509613037
    },
    {
      "auxiliary_loss_clip": 0.01215275,
      "auxiliary_loss_mlp": 0.01030958,
      "balance_loss_clip": 1.06265545,
      "balance_loss_mlp": 1.02221417,
      "epoch": 0.25455419948295555,
      "flos": 21689470897920.0,
      "grad_norm": 2.0367403871445653,
      "language_loss": 0.76138771,
      "learning_rate": 3.4941172702057353e-06,
      "loss": 0.78385001,
      "num_input_tokens_seen": 45184865,
      "step": 2117,
      "time_per_iteration": 2.6294572353363037
    },
    {
      "auxiliary_loss_clip": 0.01311714,
      "auxiliary_loss_mlp": 0.01035357,
      "balance_loss_clip": 1.06073093,
      "balance_loss_mlp": 1.02690554,
      "epoch": 0.25467444237359466,
      "flos": 26250341339520.0,
      "grad_norm": 2.0375575724219854,
      "language_loss": 0.80608559,
      "learning_rate": 3.4935993291407924e-06,
      "loss": 0.82955635,
      "num_input_tokens_seen": 45203690,
      "step": 2118,
      "time_per_iteration": 3.7872841358184814
    },
    {
      "auxiliary_loss_clip": 0.01313282,
      "auxiliary_loss_mlp": 0.01037773,
      "balance_loss_clip": 1.0598073,
      "balance_loss_mlp": 1.02858758,
      "epoch": 0.25479468526423377,
      "flos": 26979686997120.0,
      "grad_norm": 2.693397627287194,
      "language_loss": 0.71834236,
      "learning_rate": 3.4930811614980183e-06,
      "loss": 0.74185288,
      "num_input_tokens_seen": 45225385,
      "step": 2119,
      "time_per_iteration": 3.549647331237793
    },
    {
      "auxiliary_loss_clip": 0.01261112,
      "auxiliary_loss_mlp": 0.01033255,
      "balance_loss_clip": 1.05947089,
      "balance_loss_mlp": 1.02412987,
      "epoch": 0.2549149281548728,
      "flos": 23475811098240.0,
      "grad_norm": 1.6329885786635523,
      "language_loss": 0.79339677,
      "learning_rate": 3.4925627673560198e-06,
      "loss": 0.81634045,
      "num_input_tokens_seen": 45246045,
      "step": 2120,
      "time_per_iteration": 2.7023160457611084
    },
    {
      "auxiliary_loss_clip": 0.01368019,
      "auxiliary_loss_mlp": 0.01038996,
      "balance_loss_clip": 1.06162167,
      "balance_loss_mlp": 1.03008473,
      "epoch": 0.25503517104551193,
      "flos": 25812302981760.0,
      "grad_norm": 1.748343717781985,
      "language_loss": 0.88977683,
      "learning_rate": 3.4920441467934357e-06,
      "loss": 0.91384697,
      "num_input_tokens_seen": 45266560,
      "step": 2121,
      "time_per_iteration": 2.7547953128814697
    },
    {
      "auxiliary_loss_clip": 0.0135153,
      "auxiliary_loss_mlp": 0.01025982,
      "balance_loss_clip": 1.05479753,
      "balance_loss_mlp": 1.0174768,
      "epoch": 0.25515541393615104,
      "flos": 26645106787200.0,
      "grad_norm": 3.414135646689568,
      "language_loss": 0.82969439,
      "learning_rate": 3.491525299888941e-06,
      "loss": 0.85346961,
      "num_input_tokens_seen": 45285405,
      "step": 2122,
      "time_per_iteration": 2.8044161796569824
    },
    {
      "auxiliary_loss_clip": 0.01204114,
      "auxiliary_loss_mlp": 0.02521632,
      "balance_loss_clip": 1.02987528,
      "balance_loss_mlp": 1.0000689,
      "epoch": 0.2552756568267901,
      "flos": 65955945847680.0,
      "grad_norm": 0.8780792939835298,
      "language_loss": 0.62616408,
      "learning_rate": 3.491006226721244e-06,
      "loss": 0.66342151,
      "num_input_tokens_seen": 45349615,
      "step": 2123,
      "time_per_iteration": 4.208857297897339
    },
    {
      "auxiliary_loss_clip": 0.01230002,
      "auxiliary_loss_mlp": 0.02571317,
      "balance_loss_clip": 1.06028676,
      "balance_loss_mlp": 1.00005519,
      "epoch": 0.2553958997174292,
      "flos": 17931096161280.0,
      "grad_norm": 2.0934304193269595,
      "language_loss": 0.77699196,
      "learning_rate": 3.4904869273690882e-06,
      "loss": 0.81500518,
      "num_input_tokens_seen": 45367505,
      "step": 2124,
      "time_per_iteration": 2.639619827270508
    },
    {
      "auxiliary_loss_clip": 0.01264669,
      "auxiliary_loss_mlp": 0.01033935,
      "balance_loss_clip": 1.0594703,
      "balance_loss_mlp": 1.0256319,
      "epoch": 0.2555161426080683,
      "flos": 23367791923200.0,
      "grad_norm": 1.953977431628158,
      "language_loss": 0.89135075,
      "learning_rate": 3.489967401911251e-06,
      "loss": 0.9143368,
      "num_input_tokens_seen": 45386805,
      "step": 2125,
      "time_per_iteration": 2.7364842891693115
    },
    {
      "auxiliary_loss_clip": 0.01217278,
      "auxiliary_loss_mlp": 0.01033901,
      "balance_loss_clip": 1.06302881,
      "balance_loss_mlp": 1.02516341,
      "epoch": 0.2556363854987074,
      "flos": 40625130723840.0,
      "grad_norm": 1.653358442155824,
      "language_loss": 0.69801664,
      "learning_rate": 3.4894476504265428e-06,
      "loss": 0.72052842,
      "num_input_tokens_seen": 45411045,
      "step": 2126,
      "time_per_iteration": 2.909564971923828
    },
    {
      "auxiliary_loss_clip": 0.0115686,
      "auxiliary_loss_mlp": 0.0100029,
      "balance_loss_clip": 1.02848577,
      "balance_loss_mlp": 0.99898452,
      "epoch": 0.2557566283893465,
      "flos": 68019443389440.0,
      "grad_norm": 0.7641333655239898,
      "language_loss": 0.54396951,
      "learning_rate": 3.4889276729938104e-06,
      "loss": 0.56554103,
      "num_input_tokens_seen": 45469575,
      "step": 2127,
      "time_per_iteration": 3.2201597690582275
    },
    {
      "auxiliary_loss_clip": 0.01308352,
      "auxiliary_loss_mlp": 0.01037915,
      "balance_loss_clip": 1.05929077,
      "balance_loss_mlp": 1.02974296,
      "epoch": 0.2558768712799856,
      "flos": 22635645004800.0,
      "grad_norm": 1.8839439402009803,
      "language_loss": 0.80878019,
      "learning_rate": 3.488407469691934e-06,
      "loss": 0.83224297,
      "num_input_tokens_seen": 45490270,
      "step": 2128,
      "time_per_iteration": 3.6668334007263184
    },
    {
      "auxiliary_loss_clip": 0.01310863,
      "auxiliary_loss_mlp": 0.01037452,
      "balance_loss_clip": 1.05634379,
      "balance_loss_mlp": 1.02807057,
      "epoch": 0.25599711417062465,
      "flos": 26396354125440.0,
      "grad_norm": 2.1059810600546074,
      "language_loss": 0.80860889,
      "learning_rate": 3.487887040599828e-06,
      "loss": 0.83209205,
      "num_input_tokens_seen": 45510070,
      "step": 2129,
      "time_per_iteration": 2.7118170261383057
    },
    {
      "auxiliary_loss_clip": 0.01216608,
      "auxiliary_loss_mlp": 0.01032992,
      "balance_loss_clip": 1.06557226,
      "balance_loss_mlp": 1.02480292,
      "epoch": 0.25611735706126376,
      "flos": 22852042490880.0,
      "grad_norm": 2.550759557262873,
      "language_loss": 0.76030374,
      "learning_rate": 3.4873663857964407e-06,
      "loss": 0.78279978,
      "num_input_tokens_seen": 45527285,
      "step": 2130,
      "time_per_iteration": 2.6365292072296143
    },
    {
      "auxiliary_loss_clip": 0.01416604,
      "auxiliary_loss_mlp": 0.01029932,
      "balance_loss_clip": 1.05992651,
      "balance_loss_mlp": 1.02133739,
      "epoch": 0.2562375999519028,
      "flos": 23367863750400.0,
      "grad_norm": 1.9092480929834914,
      "language_loss": 0.6660322,
      "learning_rate": 3.4868455053607556e-06,
      "loss": 0.69049752,
      "num_input_tokens_seen": 45546900,
      "step": 2131,
      "time_per_iteration": 2.8564846515655518
    },
    {
      "auxiliary_loss_clip": 0.0126667,
      "auxiliary_loss_mlp": 0.01032539,
      "balance_loss_clip": 1.05996919,
      "balance_loss_mlp": 1.02335978,
      "epoch": 0.2563578428425419,
      "flos": 22856962654080.0,
      "grad_norm": 5.453924127160086,
      "language_loss": 0.71404397,
      "learning_rate": 3.486324399371789e-06,
      "loss": 0.73703611,
      "num_input_tokens_seen": 45566200,
      "step": 2132,
      "time_per_iteration": 2.7104945182800293
    },
    {
      "auxiliary_loss_clip": 0.01354298,
      "auxiliary_loss_mlp": 0.01033022,
      "balance_loss_clip": 1.05785,
      "balance_loss_mlp": 1.02517796,
      "epoch": 0.25647808573318104,
      "flos": 21653883498240.0,
      "grad_norm": 1.9516594667820069,
      "language_loss": 0.78773093,
      "learning_rate": 3.485803067908593e-06,
      "loss": 0.81160414,
      "num_input_tokens_seen": 45585710,
      "step": 2133,
      "time_per_iteration": 2.849024534225464
    },
    {
      "auxiliary_loss_clip": 0.0144156,
      "auxiliary_loss_mlp": 0.01036042,
      "balance_loss_clip": 1.04723954,
      "balance_loss_mlp": 1.0268513,
      "epoch": 0.2565983286238201,
      "flos": 33730569659520.0,
      "grad_norm": 2.278550821955213,
      "language_loss": 0.7987932,
      "learning_rate": 3.485281511050253e-06,
      "loss": 0.82356924,
      "num_input_tokens_seen": 45607845,
      "step": 2134,
      "time_per_iteration": 2.8973453044891357
    },
    {
      "auxiliary_loss_clip": 0.01178337,
      "auxiliary_loss_mlp": 0.0102757,
      "balance_loss_clip": 1.05980527,
      "balance_loss_mlp": 1.01973486,
      "epoch": 0.2567185715144592,
      "flos": 16216002587520.0,
      "grad_norm": 3.351097008584336,
      "language_loss": 0.89820993,
      "learning_rate": 3.484759728875889e-06,
      "loss": 0.92026901,
      "num_input_tokens_seen": 45623210,
      "step": 2135,
      "time_per_iteration": 2.6023073196411133
    },
    {
      "auxiliary_loss_clip": 0.0140278,
      "auxiliary_loss_mlp": 0.01032962,
      "balance_loss_clip": 1.05456519,
      "balance_loss_mlp": 1.02492762,
      "epoch": 0.2568388144050983,
      "flos": 17458475984640.0,
      "grad_norm": 2.013686792094541,
      "language_loss": 0.81158507,
      "learning_rate": 3.4842377214646543e-06,
      "loss": 0.83594251,
      "num_input_tokens_seen": 45641505,
      "step": 2136,
      "time_per_iteration": 2.718172788619995
    },
    {
      "auxiliary_loss_clip": 0.01212453,
      "auxiliary_loss_mlp": 0.01033329,
      "balance_loss_clip": 1.06246853,
      "balance_loss_mlp": 1.02512753,
      "epoch": 0.25695905729573737,
      "flos": 20887442069760.0,
      "grad_norm": 2.212225082069032,
      "language_loss": 0.66970968,
      "learning_rate": 3.483715488895737e-06,
      "loss": 0.69216752,
      "num_input_tokens_seen": 45661835,
      "step": 2137,
      "time_per_iteration": 2.633713960647583
    },
    {
      "auxiliary_loss_clip": 0.01406867,
      "auxiliary_loss_mlp": 0.01028723,
      "balance_loss_clip": 1.05231833,
      "balance_loss_mlp": 1.02032495,
      "epoch": 0.2570793001863765,
      "flos": 24717278914560.0,
      "grad_norm": 1.8359237349706017,
      "language_loss": 0.78877747,
      "learning_rate": 3.48319303124836e-06,
      "loss": 0.81313348,
      "num_input_tokens_seen": 45682215,
      "step": 2138,
      "time_per_iteration": 2.7595648765563965
    },
    {
      "auxiliary_loss_clip": 0.01317327,
      "auxiliary_loss_mlp": 0.01034217,
      "balance_loss_clip": 1.06658173,
      "balance_loss_mlp": 1.02578926,
      "epoch": 0.2571995430770156,
      "flos": 26906896085760.0,
      "grad_norm": 2.057350873921361,
      "language_loss": 0.67038524,
      "learning_rate": 3.4826703486017798e-06,
      "loss": 0.6939007,
      "num_input_tokens_seen": 45701840,
      "step": 2139,
      "time_per_iteration": 2.6950290203094482
    },
    {
      "auxiliary_loss_clip": 0.0125704,
      "auxiliary_loss_mlp": 0.01028535,
      "balance_loss_clip": 1.06164169,
      "balance_loss_mlp": 1.02042866,
      "epoch": 0.25731978596765465,
      "flos": 19792561656960.0,
      "grad_norm": 1.6260119460030162,
      "language_loss": 0.76814294,
      "learning_rate": 3.4821474410352867e-06,
      "loss": 0.79099876,
      "num_input_tokens_seen": 45720500,
      "step": 2140,
      "time_per_iteration": 2.6531729698181152
    },
    {
      "auxiliary_loss_clip": 0.01197909,
      "auxiliary_loss_mlp": 0.01003125,
      "balance_loss_clip": 1.04515016,
      "balance_loss_mlp": 1.00176001,
      "epoch": 0.25744002885829376,
      "flos": 70564970471040.0,
      "grad_norm": 0.895463969643903,
      "language_loss": 0.62661147,
      "learning_rate": 3.481624308628205e-06,
      "loss": 0.6486218,
      "num_input_tokens_seen": 45781870,
      "step": 2141,
      "time_per_iteration": 3.3645691871643066
    },
    {
      "auxiliary_loss_clip": 0.01311375,
      "auxiliary_loss_mlp": 0.01030189,
      "balance_loss_clip": 1.05718839,
      "balance_loss_mlp": 1.0214088,
      "epoch": 0.25756027174893287,
      "flos": 18038181582720.0,
      "grad_norm": 2.6627487861583616,
      "language_loss": 1.00097632,
      "learning_rate": 3.481100951459893e-06,
      "loss": 1.02439189,
      "num_input_tokens_seen": 45794890,
      "step": 2142,
      "time_per_iteration": 2.7363739013671875
    },
    {
      "auxiliary_loss_clip": 0.01255906,
      "auxiliary_loss_mlp": 0.01028209,
      "balance_loss_clip": 1.05988514,
      "balance_loss_mlp": 1.01995385,
      "epoch": 0.2576805146395719,
      "flos": 22674069578880.0,
      "grad_norm": 1.614839190866429,
      "language_loss": 0.78698885,
      "learning_rate": 3.4805773696097453e-06,
      "loss": 0.80983007,
      "num_input_tokens_seen": 45815780,
      "step": 2143,
      "time_per_iteration": 3.542092800140381
    },
    {
      "auxiliary_loss_clip": 0.01311008,
      "auxiliary_loss_mlp": 0.01030036,
      "balance_loss_clip": 1.06189549,
      "balance_loss_mlp": 1.02152455,
      "epoch": 0.25780075753021103,
      "flos": 16472225278080.0,
      "grad_norm": 4.0727654795966455,
      "language_loss": 0.8760947,
      "learning_rate": 3.4800535631571874e-06,
      "loss": 0.89950514,
      "num_input_tokens_seen": 45831310,
      "step": 2144,
      "time_per_iteration": 2.670546293258667
    },
    {
      "auxiliary_loss_clip": 0.01312193,
      "auxiliary_loss_mlp": 0.01028991,
      "balance_loss_clip": 1.05746603,
      "balance_loss_mlp": 1.02037275,
      "epoch": 0.25792100042085014,
      "flos": 22820297846400.0,
      "grad_norm": 2.111043317527862,
      "language_loss": 0.76303661,
      "learning_rate": 3.4795295321816804e-06,
      "loss": 0.78644848,
      "num_input_tokens_seen": 45850135,
      "step": 2145,
      "time_per_iteration": 3.582796812057495
    },
    {
      "auxiliary_loss_clip": 0.01300538,
      "auxiliary_loss_mlp": 0.01026259,
      "balance_loss_clip": 1.0535419,
      "balance_loss_mlp": 1.01777732,
      "epoch": 0.2580412433114892,
      "flos": 18697286194560.0,
      "grad_norm": 2.218194824346025,
      "language_loss": 0.91104865,
      "learning_rate": 3.47900527676272e-06,
      "loss": 0.93431664,
      "num_input_tokens_seen": 45868470,
      "step": 2146,
      "time_per_iteration": 2.6335608959198
    },
    {
      "auxiliary_loss_clip": 0.01213455,
      "auxiliary_loss_mlp": 0.01032263,
      "balance_loss_clip": 1.06459355,
      "balance_loss_mlp": 1.02401996,
      "epoch": 0.2581614862021283,
      "flos": 14283146810880.0,
      "grad_norm": 2.3601264285681354,
      "language_loss": 0.8820107,
      "learning_rate": 3.478480796979835e-06,
      "loss": 0.90446788,
      "num_input_tokens_seen": 45886355,
      "step": 2147,
      "time_per_iteration": 2.586411952972412
    },
    {
      "auxiliary_loss_clip": 0.01305246,
      "auxiliary_loss_mlp": 0.01027692,
      "balance_loss_clip": 1.05886507,
      "balance_loss_mlp": 1.01950192,
      "epoch": 0.25828172909276736,
      "flos": 29498281856640.0,
      "grad_norm": 1.5552183395315247,
      "language_loss": 0.77801859,
      "learning_rate": 3.4779560929125894e-06,
      "loss": 0.80134797,
      "num_input_tokens_seen": 45907900,
      "step": 2148,
      "time_per_iteration": 2.750880241394043
    },
    {
      "auxiliary_loss_clip": 0.01260602,
      "auxiliary_loss_mlp": 0.0100086,
      "balance_loss_clip": 1.02807033,
      "balance_loss_mlp": 0.99957818,
      "epoch": 0.2584019719834065,
      "flos": 67114387376640.0,
      "grad_norm": 0.6644409971108939,
      "language_loss": 0.56896412,
      "learning_rate": 3.4774311646405783e-06,
      "loss": 0.59157872,
      "num_input_tokens_seen": 45977805,
      "step": 2149,
      "time_per_iteration": 4.318928480148315
    },
    {
      "auxiliary_loss_clip": 0.01350184,
      "auxiliary_loss_mlp": 0.01035975,
      "balance_loss_clip": 1.0511831,
      "balance_loss_mlp": 1.02812481,
      "epoch": 0.2585222148740456,
      "flos": 22893555634560.0,
      "grad_norm": 2.2999836092756785,
      "language_loss": 0.83569765,
      "learning_rate": 3.476906012243435e-06,
      "loss": 0.85955924,
      "num_input_tokens_seen": 45996715,
      "step": 2150,
      "time_per_iteration": 2.7770752906799316
    },
    {
      "auxiliary_loss_clip": 0.01256421,
      "auxiliary_loss_mlp": 0.01035196,
      "balance_loss_clip": 1.06003165,
      "balance_loss_mlp": 1.02651143,
      "epoch": 0.25864245776468464,
      "flos": 28909202808960.0,
      "grad_norm": 1.5561622398827049,
      "language_loss": 0.81080431,
      "learning_rate": 3.476380635800824e-06,
      "loss": 0.83372045,
      "num_input_tokens_seen": 46017915,
      "step": 2151,
      "time_per_iteration": 2.696713447570801
    },
    {
      "auxiliary_loss_clip": 0.01304782,
      "auxiliary_loss_mlp": 0.01036275,
      "balance_loss_clip": 1.05850255,
      "balance_loss_mlp": 1.02723861,
      "epoch": 0.25876270065532375,
      "flos": 14793185980800.0,
      "grad_norm": 5.912714476774329,
      "language_loss": 0.86369789,
      "learning_rate": 3.475855035392444e-06,
      "loss": 0.88710845,
      "num_input_tokens_seen": 46033235,
      "step": 2152,
      "time_per_iteration": 2.702138900756836
    },
    {
      "auxiliary_loss_clip": 0.01370675,
      "auxiliary_loss_mlp": 0.01033689,
      "balance_loss_clip": 1.05369925,
      "balance_loss_mlp": 1.02542794,
      "epoch": 0.25888294354596286,
      "flos": 60467821810560.0,
      "grad_norm": 2.0529203924295523,
      "language_loss": 0.71160114,
      "learning_rate": 3.475329211098029e-06,
      "loss": 0.73564482,
      "num_input_tokens_seen": 46056390,
      "step": 2153,
      "time_per_iteration": 3.126304864883423
    },
    {
      "auxiliary_loss_clip": 0.01411115,
      "auxiliary_loss_mlp": 0.01036465,
      "balance_loss_clip": 1.05805981,
      "balance_loss_mlp": 1.02750611,
      "epoch": 0.2590031864366019,
      "flos": 27851166771840.0,
      "grad_norm": 1.9289744925876198,
      "language_loss": 0.82287586,
      "learning_rate": 3.4748031629973453e-06,
      "loss": 0.84735161,
      "num_input_tokens_seen": 46077120,
      "step": 2154,
      "time_per_iteration": 3.882110118865967
    },
    {
      "auxiliary_loss_clip": 0.01306306,
      "auxiliary_loss_mlp": 0.00999177,
      "balance_loss_clip": 1.02674031,
      "balance_loss_mlp": 0.99797863,
      "epoch": 0.25912342932724103,
      "flos": 62422444206720.0,
      "grad_norm": 1.085221336813201,
      "language_loss": 0.56521481,
      "learning_rate": 3.4742768911701944e-06,
      "loss": 0.58826965,
      "num_input_tokens_seen": 46139815,
      "step": 2155,
      "time_per_iteration": 3.651494264602661
    },
    {
      "auxiliary_loss_clip": 0.0126657,
      "auxiliary_loss_mlp": 0.01033752,
      "balance_loss_clip": 1.06186259,
      "balance_loss_mlp": 1.02460885,
      "epoch": 0.25924367221788014,
      "flos": 12378839368320.0,
      "grad_norm": 3.928310236166233,
      "language_loss": 0.70375329,
      "learning_rate": 3.4737503956964113e-06,
      "loss": 0.72675651,
      "num_input_tokens_seen": 46152120,
      "step": 2156,
      "time_per_iteration": 2.722292900085449
    },
    {
      "auxiliary_loss_clip": 0.01308795,
      "auxiliary_loss_mlp": 0.01039177,
      "balance_loss_clip": 1.05649471,
      "balance_loss_mlp": 1.02971768,
      "epoch": 0.2593639151085192,
      "flos": 14575208296320.0,
      "grad_norm": 2.1934606914579065,
      "language_loss": 0.67327601,
      "learning_rate": 3.473223676655865e-06,
      "loss": 0.69675577,
      "num_input_tokens_seen": 46170120,
      "step": 2157,
      "time_per_iteration": 2.6231279373168945
    },
    {
      "auxiliary_loss_clip": 0.01308309,
      "auxiliary_loss_mlp": 0.01033456,
      "balance_loss_clip": 1.05446088,
      "balance_loss_mlp": 1.02422929,
      "epoch": 0.2594841579991583,
      "flos": 15230937029760.0,
      "grad_norm": 1.9500598531246045,
      "language_loss": 0.80168486,
      "learning_rate": 3.472696734128459e-06,
      "loss": 0.82510245,
      "num_input_tokens_seen": 46187985,
      "step": 2158,
      "time_per_iteration": 2.6766035556793213
    },
    {
      "auxiliary_loss_clip": 0.01260596,
      "auxiliary_loss_mlp": 0.01034863,
      "balance_loss_clip": 1.0606643,
      "balance_loss_mlp": 1.02574325,
      "epoch": 0.2596044008897974,
      "flos": 23623583650560.0,
      "grad_norm": 1.848576009106137,
      "language_loss": 0.75996858,
      "learning_rate": 3.4721695681941286e-06,
      "loss": 0.78292316,
      "num_input_tokens_seen": 46207025,
      "step": 2159,
      "time_per_iteration": 2.702679395675659
    },
    {
      "auxiliary_loss_clip": 0.01309242,
      "auxiliary_loss_mlp": 0.02575626,
      "balance_loss_clip": 1.05899811,
      "balance_loss_mlp": 1.00029778,
      "epoch": 0.25972464378043647,
      "flos": 13772281628160.0,
      "grad_norm": 2.8692325798842293,
      "language_loss": 0.8181957,
      "learning_rate": 3.471642178932845e-06,
      "loss": 0.8570444,
      "num_input_tokens_seen": 46225670,
      "step": 2160,
      "time_per_iteration": 2.769883871078491
    },
    {
      "auxiliary_loss_clip": 0.01309373,
      "auxiliary_loss_mlp": 0.01029832,
      "balance_loss_clip": 1.0546844,
      "balance_loss_mlp": 1.02102208,
      "epoch": 0.2598448866710756,
      "flos": 19573578391680.0,
      "grad_norm": 2.5708488598250714,
      "language_loss": 0.89839524,
      "learning_rate": 3.471114566424613e-06,
      "loss": 0.92178726,
      "num_input_tokens_seen": 46244130,
      "step": 2161,
      "time_per_iteration": 2.6563563346862793
    },
    {
      "auxiliary_loss_clip": 0.01307689,
      "auxiliary_loss_mlp": 0.01025996,
      "balance_loss_clip": 1.06100154,
      "balance_loss_mlp": 1.01767516,
      "epoch": 0.25996512956171464,
      "flos": 21653237053440.0,
      "grad_norm": 1.9000342737733447,
      "language_loss": 0.7587713,
      "learning_rate": 3.4705867307494715e-06,
      "loss": 0.78210819,
      "num_input_tokens_seen": 46263200,
      "step": 2162,
      "time_per_iteration": 2.7112088203430176
    },
    {
      "auxiliary_loss_clip": 0.01262551,
      "auxiliary_loss_mlp": 0.01031855,
      "balance_loss_clip": 1.05742741,
      "balance_loss_mlp": 1.02325368,
      "epoch": 0.26008537245235375,
      "flos": 18223480869120.0,
      "grad_norm": 2.2682545951961335,
      "language_loss": 0.84973633,
      "learning_rate": 3.470058671987492e-06,
      "loss": 0.87268043,
      "num_input_tokens_seen": 46281465,
      "step": 2163,
      "time_per_iteration": 2.6640963554382324
    },
    {
      "auxiliary_loss_clip": 0.01265128,
      "auxiliary_loss_mlp": 0.01034542,
      "balance_loss_clip": 1.058393,
      "balance_loss_mlp": 1.02576852,
      "epoch": 0.26020561534299286,
      "flos": 24645385843200.0,
      "grad_norm": 2.2252646183173064,
      "language_loss": 0.84596586,
      "learning_rate": 3.4695303902187805e-06,
      "loss": 0.86896259,
      "num_input_tokens_seen": 46301020,
      "step": 2164,
      "time_per_iteration": 2.6923372745513916
    },
    {
      "auxiliary_loss_clip": 0.01275174,
      "auxiliary_loss_mlp": 0.01033624,
      "balance_loss_clip": 1.05404139,
      "balance_loss_mlp": 1.0255239,
      "epoch": 0.2603258582336319,
      "flos": 25773662926080.0,
      "grad_norm": 2.10416482096449,
      "language_loss": 0.78618997,
      "learning_rate": 3.469001885523478e-06,
      "loss": 0.80927801,
      "num_input_tokens_seen": 46321740,
      "step": 2165,
      "time_per_iteration": 2.7587087154388428
    },
    {
      "auxiliary_loss_clip": 0.01211113,
      "auxiliary_loss_mlp": 0.01035955,
      "balance_loss_clip": 1.06160021,
      "balance_loss_mlp": 1.0275625,
      "epoch": 0.260446101124271,
      "flos": 28766314506240.0,
      "grad_norm": 1.9594474788474612,
      "language_loss": 0.81260365,
      "learning_rate": 3.4684731579817568e-06,
      "loss": 0.83507437,
      "num_input_tokens_seen": 46342730,
      "step": 2166,
      "time_per_iteration": 2.678591251373291
    },
    {
      "auxiliary_loss_clip": 0.0145181,
      "auxiliary_loss_mlp": 0.01031649,
      "balance_loss_clip": 1.05309439,
      "balance_loss_mlp": 1.02300656,
      "epoch": 0.26056634401491013,
      "flos": 25666757072640.0,
      "grad_norm": 1.6135391186986425,
      "language_loss": 0.76677185,
      "learning_rate": 3.4679442076738247e-06,
      "loss": 0.79160643,
      "num_input_tokens_seen": 46362445,
      "step": 2167,
      "time_per_iteration": 2.8043630123138428
    },
    {
      "auxiliary_loss_clip": 0.01212493,
      "auxiliary_loss_mlp": 0.01030858,
      "balance_loss_clip": 1.06074166,
      "balance_loss_mlp": 1.02132702,
      "epoch": 0.2606865869055492,
      "flos": 27052765217280.0,
      "grad_norm": 3.5028602317290782,
      "language_loss": 0.83535039,
      "learning_rate": 3.4674150346799245e-06,
      "loss": 0.85778391,
      "num_input_tokens_seen": 46382145,
      "step": 2168,
      "time_per_iteration": 2.655402183532715
    },
    {
      "auxiliary_loss_clip": 0.01310796,
      "auxiliary_loss_mlp": 0.01028799,
      "balance_loss_clip": 1.05978656,
      "balance_loss_mlp": 1.0195787,
      "epoch": 0.2608068297961883,
      "flos": 17712615686400.0,
      "grad_norm": 2.2155517150885915,
      "language_loss": 0.80461729,
      "learning_rate": 3.4668856390803295e-06,
      "loss": 0.82801324,
      "num_input_tokens_seen": 46400025,
      "step": 2169,
      "time_per_iteration": 3.6878297328948975
    },
    {
      "auxiliary_loss_clip": 0.01249905,
      "auxiliary_loss_mlp": 0.01028283,
      "balance_loss_clip": 1.05549598,
      "balance_loss_mlp": 1.01956844,
      "epoch": 0.2609270726868274,
      "flos": 18551632544640.0,
      "grad_norm": 2.0342034949662007,
      "language_loss": 0.89934254,
      "learning_rate": 3.4663560209553495e-06,
      "loss": 0.92212439,
      "num_input_tokens_seen": 46418090,
      "step": 2170,
      "time_per_iteration": 2.6436030864715576
    },
    {
      "auxiliary_loss_clip": 0.01306574,
      "auxiliary_loss_mlp": 0.01041673,
      "balance_loss_clip": 1.05713665,
      "balance_loss_mlp": 1.03242266,
      "epoch": 0.26104731557746647,
      "flos": 21835699165440.0,
      "grad_norm": 1.8700466334659973,
      "language_loss": 0.79644406,
      "learning_rate": 3.4658261803853267e-06,
      "loss": 0.81992656,
      "num_input_tokens_seen": 46436015,
      "step": 2171,
      "time_per_iteration": 3.6166367530822754
    },
    {
      "auxiliary_loss_clip": 0.01305467,
      "auxiliary_loss_mlp": 0.01032096,
      "balance_loss_clip": 1.0581708,
      "balance_loss_mlp": 1.02310765,
      "epoch": 0.2611675584681056,
      "flos": 21689650465920.0,
      "grad_norm": 4.396741786005645,
      "language_loss": 0.80692291,
      "learning_rate": 3.4652961174506383e-06,
      "loss": 0.8302986,
      "num_input_tokens_seen": 46455885,
      "step": 2172,
      "time_per_iteration": 2.6488633155822754
    },
    {
      "auxiliary_loss_clip": 0.01153297,
      "auxiliary_loss_mlp": 0.01000152,
      "balance_loss_clip": 1.02814722,
      "balance_loss_mlp": 0.99893612,
      "epoch": 0.2612878013587447,
      "flos": 71862101389440.0,
      "grad_norm": 0.9632052947704106,
      "language_loss": 0.58078551,
      "learning_rate": 3.464765832231694e-06,
      "loss": 0.60232008,
      "num_input_tokens_seen": 46510050,
      "step": 2173,
      "time_per_iteration": 3.1993191242218018
    },
    {
      "auxiliary_loss_clip": 0.01259552,
      "auxiliary_loss_mlp": 0.01030994,
      "balance_loss_clip": 1.05998707,
      "balance_loss_mlp": 1.02280402,
      "epoch": 0.26140804424938374,
      "flos": 20227511445120.0,
      "grad_norm": 1.896866464081546,
      "language_loss": 0.7088238,
      "learning_rate": 3.4642353248089373e-06,
      "loss": 0.73172927,
      "num_input_tokens_seen": 46528810,
      "step": 2174,
      "time_per_iteration": 2.6495962142944336
    },
    {
      "auxiliary_loss_clip": 0.01304748,
      "auxiliary_loss_mlp": 0.01036282,
      "balance_loss_clip": 1.05709219,
      "balance_loss_mlp": 1.0272398,
      "epoch": 0.26152828714002285,
      "flos": 25557085872000.0,
      "grad_norm": 1.904770715445126,
      "language_loss": 0.80498481,
      "learning_rate": 3.463704595262846e-06,
      "loss": 0.82839513,
      "num_input_tokens_seen": 46549690,
      "step": 2175,
      "time_per_iteration": 3.6258015632629395
    },
    {
      "auxiliary_loss_clip": 0.01357444,
      "auxiliary_loss_mlp": 0.01032524,
      "balance_loss_clip": 1.05771828,
      "balance_loss_mlp": 1.02434587,
      "epoch": 0.26164853003066196,
      "flos": 25446516831360.0,
      "grad_norm": 2.730367454747495,
      "language_loss": 0.70706838,
      "learning_rate": 3.463173643673931e-06,
      "loss": 0.73096806,
      "num_input_tokens_seen": 46572215,
      "step": 2176,
      "time_per_iteration": 2.8494210243225098
    },
    {
      "auxiliary_loss_clip": 0.01164179,
      "auxiliary_loss_mlp": 0.01001995,
      "balance_loss_clip": 1.02827978,
      "balance_loss_mlp": 1.00076151,
      "epoch": 0.261768772921301,
      "flos": 53944580568960.0,
      "grad_norm": 0.8990415152395262,
      "language_loss": 0.6352731,
      "learning_rate": 3.4626424701227387e-06,
      "loss": 0.65693486,
      "num_input_tokens_seen": 46627275,
      "step": 2177,
      "time_per_iteration": 3.1439626216888428
    },
    {
      "auxiliary_loss_clip": 0.01108413,
      "auxiliary_loss_mlp": 0.01002137,
      "balance_loss_clip": 1.029392,
      "balance_loss_mlp": 1.00090909,
      "epoch": 0.26188901581194013,
      "flos": 70687606481280.0,
      "grad_norm": 0.8255215866016168,
      "language_loss": 0.55822289,
      "learning_rate": 3.4621110746898452e-06,
      "loss": 0.57932842,
      "num_input_tokens_seen": 46695135,
      "step": 2178,
      "time_per_iteration": 3.3095543384552
    },
    {
      "auxiliary_loss_clip": 0.01263895,
      "auxiliary_loss_mlp": 0.01028785,
      "balance_loss_clip": 1.05957651,
      "balance_loss_mlp": 1.02010632,
      "epoch": 0.2620092587025792,
      "flos": 21069580959360.0,
      "grad_norm": 1.5898420584295405,
      "language_loss": 0.74800754,
      "learning_rate": 3.4615794574558654e-06,
      "loss": 0.77093434,
      "num_input_tokens_seen": 46714145,
      "step": 2179,
      "time_per_iteration": 2.6709303855895996
    },
    {
      "auxiliary_loss_clip": 0.01306139,
      "auxiliary_loss_mlp": 0.01029537,
      "balance_loss_clip": 1.05608845,
      "balance_loss_mlp": 1.02104378,
      "epoch": 0.2621295015932183,
      "flos": 18369601395840.0,
      "grad_norm": 4.342899274853047,
      "language_loss": 0.8427195,
      "learning_rate": 3.4610476185014436e-06,
      "loss": 0.86607623,
      "num_input_tokens_seen": 46731405,
      "step": 2180,
      "time_per_iteration": 3.6585946083068848
    },
    {
      "auxiliary_loss_clip": 0.01212546,
      "auxiliary_loss_mlp": 0.01028524,
      "balance_loss_clip": 1.06013799,
      "balance_loss_mlp": 1.01916647,
      "epoch": 0.2622497444838574,
      "flos": 23659997063040.0,
      "grad_norm": 2.3235009259842125,
      "language_loss": 0.79412639,
      "learning_rate": 3.4605155579072597e-06,
      "loss": 0.81653702,
      "num_input_tokens_seen": 46751260,
      "step": 2181,
      "time_per_iteration": 2.668334722518921
    },
    {
      "auxiliary_loss_clip": 0.01410866,
      "auxiliary_loss_mlp": 0.01028656,
      "balance_loss_clip": 1.05399191,
      "balance_loss_mlp": 1.01980472,
      "epoch": 0.26236998737449646,
      "flos": 22123810154880.0,
      "grad_norm": 1.8668416330993023,
      "language_loss": 0.71226531,
      "learning_rate": 3.459983275754027e-06,
      "loss": 0.73666054,
      "num_input_tokens_seen": 46770155,
      "step": 2182,
      "time_per_iteration": 2.822453498840332
    },
    {
      "auxiliary_loss_clip": 0.01209594,
      "auxiliary_loss_mlp": 0.01028298,
      "balance_loss_clip": 1.06019068,
      "balance_loss_mlp": 1.01945853,
      "epoch": 0.26249023026513557,
      "flos": 17895185539200.0,
      "grad_norm": 2.413608914997237,
      "language_loss": 0.79700464,
      "learning_rate": 3.4594507721224918e-06,
      "loss": 0.8193835,
      "num_input_tokens_seen": 46788805,
      "step": 2183,
      "time_per_iteration": 2.586663007736206
    },
    {
      "auxiliary_loss_clip": 0.01310299,
      "auxiliary_loss_mlp": 0.01032503,
      "balance_loss_clip": 1.05669332,
      "balance_loss_mlp": 1.02420604,
      "epoch": 0.2626104731557747,
      "flos": 18332936588160.0,
      "grad_norm": 1.6957093972117632,
      "language_loss": 0.82224262,
      "learning_rate": 3.4589180470934353e-06,
      "loss": 0.84567058,
      "num_input_tokens_seen": 46808670,
      "step": 2184,
      "time_per_iteration": 2.7564425468444824
    },
    {
      "auxiliary_loss_clip": 0.01178322,
      "auxiliary_loss_mlp": 0.01033737,
      "balance_loss_clip": 1.0558008,
      "balance_loss_mlp": 1.0244205,
      "epoch": 0.26273071604641374,
      "flos": 19317714837120.0,
      "grad_norm": 2.1181531397235376,
      "language_loss": 0.7675845,
      "learning_rate": 3.4583851007476713e-06,
      "loss": 0.78970516,
      "num_input_tokens_seen": 46827140,
      "step": 2185,
      "time_per_iteration": 2.6043379306793213
    },
    {
      "auxiliary_loss_clip": 0.01359876,
      "auxiliary_loss_mlp": 0.01032246,
      "balance_loss_clip": 1.05735862,
      "balance_loss_mlp": 1.02301311,
      "epoch": 0.26285095893705285,
      "flos": 18327477720960.0,
      "grad_norm": 2.191422171046224,
      "language_loss": 0.68511546,
      "learning_rate": 3.4578519331660464e-06,
      "loss": 0.70903671,
      "num_input_tokens_seen": 46844135,
      "step": 2186,
      "time_per_iteration": 2.7517526149749756
    },
    {
      "auxiliary_loss_clip": 0.01256012,
      "auxiliary_loss_mlp": 0.01030606,
      "balance_loss_clip": 1.06010413,
      "balance_loss_mlp": 1.02245498,
      "epoch": 0.26297120182769196,
      "flos": 20193827466240.0,
      "grad_norm": 2.020668483165944,
      "language_loss": 0.82321018,
      "learning_rate": 3.4573185444294426e-06,
      "loss": 0.84607637,
      "num_input_tokens_seen": 46862500,
      "step": 2187,
      "time_per_iteration": 2.6699087619781494
    },
    {
      "auxiliary_loss_clip": 0.01311753,
      "auxiliary_loss_mlp": 0.02576972,
      "balance_loss_clip": 1.05778396,
      "balance_loss_mlp": 1.00027156,
      "epoch": 0.263091444718331,
      "flos": 22418421505920.0,
      "grad_norm": 1.6966713552667727,
      "language_loss": 0.78527445,
      "learning_rate": 3.456784934618774e-06,
      "loss": 0.82416165,
      "num_input_tokens_seen": 46883665,
      "step": 2188,
      "time_per_iteration": 2.676276922225952
    },
    {
      "auxiliary_loss_clip": 0.01308881,
      "auxiliary_loss_mlp": 0.0103333,
      "balance_loss_clip": 1.05561686,
      "balance_loss_mlp": 1.02531922,
      "epoch": 0.2632116876089701,
      "flos": 19024827338880.0,
      "grad_norm": 2.5754527447692697,
      "language_loss": 0.80148149,
      "learning_rate": 3.4562511038149897e-06,
      "loss": 0.82490361,
      "num_input_tokens_seen": 46899160,
      "step": 2189,
      "time_per_iteration": 2.6550188064575195
    },
    {
      "auxiliary_loss_clip": 0.01312204,
      "auxiliary_loss_mlp": 0.01000158,
      "balance_loss_clip": 1.02466989,
      "balance_loss_mlp": 0.99899596,
      "epoch": 0.26333193049960923,
      "flos": 67308054531840.0,
      "grad_norm": 0.8585703582039322,
      "language_loss": 0.57713628,
      "learning_rate": 3.4557170520990705e-06,
      "loss": 0.6002599,
      "num_input_tokens_seen": 46959835,
      "step": 2190,
      "time_per_iteration": 3.32315993309021
    },
    {
      "auxiliary_loss_clip": 0.01253925,
      "auxiliary_loss_mlp": 0.01030144,
      "balance_loss_clip": 1.05826592,
      "balance_loss_mlp": 1.02206707,
      "epoch": 0.2634521733902483,
      "flos": 25048806468480.0,
      "grad_norm": 1.5871908819980611,
      "language_loss": 0.86337012,
      "learning_rate": 3.4551827795520324e-06,
      "loss": 0.8862108,
      "num_input_tokens_seen": 46982720,
      "step": 2191,
      "time_per_iteration": 2.7211551666259766
    },
    {
      "auxiliary_loss_clip": 0.01257008,
      "auxiliary_loss_mlp": 0.010293,
      "balance_loss_clip": 1.05553055,
      "balance_loss_mlp": 1.02084196,
      "epoch": 0.2635724162808874,
      "flos": 20594985534720.0,
      "grad_norm": 1.8309520368702337,
      "language_loss": 0.85086465,
      "learning_rate": 3.4546482862549226e-06,
      "loss": 0.87372768,
      "num_input_tokens_seen": 47003035,
      "step": 2192,
      "time_per_iteration": 2.6574840545654297
    },
    {
      "auxiliary_loss_clip": 0.01353524,
      "auxiliary_loss_mlp": 0.01032476,
      "balance_loss_clip": 1.05337071,
      "balance_loss_mlp": 1.02332044,
      "epoch": 0.2636926591715265,
      "flos": 19244636616960.0,
      "grad_norm": 2.6165937983710856,
      "language_loss": 0.78626543,
      "learning_rate": 3.4541135722888253e-06,
      "loss": 0.81012541,
      "num_input_tokens_seen": 47019625,
      "step": 2193,
      "time_per_iteration": 2.720285177230835
    },
    {
      "auxiliary_loss_clip": 0.01208786,
      "auxiliary_loss_mlp": 0.01032398,
      "balance_loss_clip": 1.05835497,
      "balance_loss_mlp": 1.02427971,
      "epoch": 0.26381290206216557,
      "flos": 28804882734720.0,
      "grad_norm": 1.7530555626194175,
      "language_loss": 0.80310857,
      "learning_rate": 3.453578637734854e-06,
      "loss": 0.8255204,
      "num_input_tokens_seen": 47040815,
      "step": 2194,
      "time_per_iteration": 2.6565017700195312
    },
    {
      "auxiliary_loss_clip": 0.0120903,
      "auxiliary_loss_mlp": 0.01034017,
      "balance_loss_clip": 1.06154275,
      "balance_loss_mlp": 1.02563095,
      "epoch": 0.2639331449528047,
      "flos": 25008909436800.0,
      "grad_norm": 7.8753897067186225,
      "language_loss": 0.78800148,
      "learning_rate": 3.4530434826741605e-06,
      "loss": 0.81043196,
      "num_input_tokens_seen": 47061755,
      "step": 2195,
      "time_per_iteration": 3.5613107681274414
    },
    {
      "auxiliary_loss_clip": 0.01298886,
      "auxiliary_loss_mlp": 0.01031723,
      "balance_loss_clip": 1.05559313,
      "balance_loss_mlp": 1.02409351,
      "epoch": 0.26405338784344373,
      "flos": 46535775465600.0,
      "grad_norm": 3.7609442697821556,
      "language_loss": 0.68898088,
      "learning_rate": 3.452508107187926e-06,
      "loss": 0.71228701,
      "num_input_tokens_seen": 47085130,
      "step": 2196,
      "time_per_iteration": 3.800170660018921
    },
    {
      "auxiliary_loss_clip": 0.01454196,
      "auxiliary_loss_mlp": 0.01027873,
      "balance_loss_clip": 1.04904962,
      "balance_loss_mlp": 1.01919472,
      "epoch": 0.26417363073408284,
      "flos": 21179467641600.0,
      "grad_norm": 1.9133199867494894,
      "language_loss": 0.77657568,
      "learning_rate": 3.451972511357366e-06,
      "loss": 0.80139637,
      "num_input_tokens_seen": 47104675,
      "step": 2197,
      "time_per_iteration": 2.819772481918335
    },
    {
      "auxiliary_loss_clip": 0.01255349,
      "auxiliary_loss_mlp": 0.01035673,
      "balance_loss_clip": 1.05920684,
      "balance_loss_mlp": 1.02766168,
      "epoch": 0.26429387362472195,
      "flos": 22674751937280.0,
      "grad_norm": 1.9125108629388943,
      "language_loss": 0.85526001,
      "learning_rate": 3.45143669526373e-06,
      "loss": 0.87817019,
      "num_input_tokens_seen": 47124435,
      "step": 2198,
      "time_per_iteration": 2.670285940170288
    },
    {
      "auxiliary_loss_clip": 0.01217042,
      "auxiliary_loss_mlp": 0.01003974,
      "balance_loss_clip": 1.02832842,
      "balance_loss_mlp": 1.00282407,
      "epoch": 0.264414116515361,
      "flos": 67180534272000.0,
      "grad_norm": 0.7945682572921017,
      "language_loss": 0.63162959,
      "learning_rate": 3.450900658988302e-06,
      "loss": 0.65383977,
      "num_input_tokens_seen": 47185985,
      "step": 2199,
      "time_per_iteration": 3.1840710639953613
    },
    {
      "auxiliary_loss_clip": 0.01303367,
      "auxiliary_loss_mlp": 0.01033878,
      "balance_loss_clip": 1.05733371,
      "balance_loss_mlp": 1.02520609,
      "epoch": 0.2645343594060001,
      "flos": 25664709997440.0,
      "grad_norm": 18.66571005151277,
      "language_loss": 0.77674276,
      "learning_rate": 3.450364402612397e-06,
      "loss": 0.80011523,
      "num_input_tokens_seen": 47203140,
      "step": 2200,
      "time_per_iteration": 2.7286102771759033
    },
    {
      "auxiliary_loss_clip": 0.0130618,
      "auxiliary_loss_mlp": 0.01036132,
      "balance_loss_clip": 1.05849516,
      "balance_loss_mlp": 1.02757335,
      "epoch": 0.26465460229663923,
      "flos": 22491822948480.0,
      "grad_norm": 2.262737306146407,
      "language_loss": 0.83795154,
      "learning_rate": 3.449827926217366e-06,
      "loss": 0.86137468,
      "num_input_tokens_seen": 47222575,
      "step": 2201,
      "time_per_iteration": 3.707426071166992
    },
    {
      "auxiliary_loss_clip": 0.0130825,
      "auxiliary_loss_mlp": 0.01028915,
      "balance_loss_clip": 1.05212545,
      "balance_loss_mlp": 1.02073169,
      "epoch": 0.2647748451872783,
      "flos": 29388036038400.0,
      "grad_norm": 1.9118968921945272,
      "language_loss": 0.80878794,
      "learning_rate": 3.449291229884591e-06,
      "loss": 0.83215964,
      "num_input_tokens_seen": 47243815,
      "step": 2202,
      "time_per_iteration": 2.735761880874634
    },
    {
      "auxiliary_loss_clip": 0.013586,
      "auxiliary_loss_mlp": 0.01028834,
      "balance_loss_clip": 1.05233216,
      "balance_loss_mlp": 1.02041173,
      "epoch": 0.2648950880779174,
      "flos": 26797799502720.0,
      "grad_norm": 2.3161687143870897,
      "language_loss": 0.86763632,
      "learning_rate": 3.4487543136954887e-06,
      "loss": 0.89151067,
      "num_input_tokens_seen": 47263435,
      "step": 2203,
      "time_per_iteration": 2.7610201835632324
    },
    {
      "auxiliary_loss_clip": 0.01356625,
      "auxiliary_loss_mlp": 0.01031958,
      "balance_loss_clip": 1.05595529,
      "balance_loss_mlp": 1.02379203,
      "epoch": 0.2650153309685565,
      "flos": 28841008838400.0,
      "grad_norm": 2.1005705068334652,
      "language_loss": 0.91323596,
      "learning_rate": 3.448217177731509e-06,
      "loss": 0.93712175,
      "num_input_tokens_seen": 47283920,
      "step": 2204,
      "time_per_iteration": 2.759852647781372
    },
    {
      "auxiliary_loss_clip": 0.01302662,
      "auxiliary_loss_mlp": 0.01029295,
      "balance_loss_clip": 1.05847597,
      "balance_loss_mlp": 1.02124202,
      "epoch": 0.26513557385919556,
      "flos": 20303247271680.0,
      "grad_norm": 2.8633939709614187,
      "language_loss": 0.77860755,
      "learning_rate": 3.4476798220741348e-06,
      "loss": 0.80192715,
      "num_input_tokens_seen": 47302800,
      "step": 2205,
      "time_per_iteration": 2.711859941482544
    },
    {
      "auxiliary_loss_clip": 0.01208512,
      "auxiliary_loss_mlp": 0.01029232,
      "balance_loss_clip": 1.06245565,
      "balance_loss_mlp": 1.02128673,
      "epoch": 0.26525581674983467,
      "flos": 17676274101120.0,
      "grad_norm": 1.8749670780893746,
      "language_loss": 0.78714418,
      "learning_rate": 3.4471422468048826e-06,
      "loss": 0.80952168,
      "num_input_tokens_seen": 47321525,
      "step": 2206,
      "time_per_iteration": 3.5018718242645264
    },
    {
      "auxiliary_loss_clip": 0.01256463,
      "auxiliary_loss_mlp": 0.01036007,
      "balance_loss_clip": 1.05937278,
      "balance_loss_mlp": 1.0279963,
      "epoch": 0.2653760596404738,
      "flos": 26833746038400.0,
      "grad_norm": 3.129493055962416,
      "language_loss": 0.73017782,
      "learning_rate": 3.4466044520053022e-06,
      "loss": 0.75310254,
      "num_input_tokens_seen": 47340530,
      "step": 2207,
      "time_per_iteration": 2.6898462772369385
    },
    {
      "auxiliary_loss_clip": 0.01300226,
      "auxiliary_loss_mlp": 0.0103309,
      "balance_loss_clip": 1.05374718,
      "balance_loss_mlp": 1.02438152,
      "epoch": 0.26549630253111284,
      "flos": 22782160581120.0,
      "grad_norm": 2.1676731050768248,
      "language_loss": 0.60782105,
      "learning_rate": 3.446066437756977e-06,
      "loss": 0.63115418,
      "num_input_tokens_seen": 47359735,
      "step": 2208,
      "time_per_iteration": 2.8188843727111816
    },
    {
      "auxiliary_loss_clip": 0.01301972,
      "auxiliary_loss_mlp": 0.01031407,
      "balance_loss_clip": 1.05732989,
      "balance_loss_mlp": 1.02300835,
      "epoch": 0.26561654542175195,
      "flos": 23550002640000.0,
      "grad_norm": 2.113123205853615,
      "language_loss": 0.75010455,
      "learning_rate": 3.4455282041415224e-06,
      "loss": 0.77343833,
      "num_input_tokens_seen": 47378945,
      "step": 2209,
      "time_per_iteration": 2.679535150527954
    },
    {
      "auxiliary_loss_clip": 0.01355235,
      "auxiliary_loss_mlp": 0.01027896,
      "balance_loss_clip": 1.054726,
      "balance_loss_mlp": 1.02022159,
      "epoch": 0.265736788312391,
      "flos": 26906680604160.0,
      "grad_norm": 2.1465409334470014,
      "language_loss": 0.87365985,
      "learning_rate": 3.4449897512405894e-06,
      "loss": 0.89749116,
      "num_input_tokens_seen": 47398095,
      "step": 2210,
      "time_per_iteration": 2.7511255741119385
    },
    {
      "auxiliary_loss_clip": 0.01460282,
      "auxiliary_loss_mlp": 0.02573596,
      "balance_loss_clip": 1.05364299,
      "balance_loss_mlp": 1.00032461,
      "epoch": 0.2658570312030301,
      "flos": 23477139901440.0,
      "grad_norm": 3.01021503377063,
      "language_loss": 0.75371206,
      "learning_rate": 3.444451079135859e-06,
      "loss": 0.79405081,
      "num_input_tokens_seen": 47417605,
      "step": 2211,
      "time_per_iteration": 2.7451672554016113
    },
    {
      "auxiliary_loss_clip": 0.01400127,
      "auxiliary_loss_mlp": 0.02576239,
      "balance_loss_clip": 1.05143046,
      "balance_loss_mlp": 1.00023222,
      "epoch": 0.2659772740936692,
      "flos": 21866402315520.0,
      "grad_norm": 2.9912683586121274,
      "language_loss": 0.74284518,
      "learning_rate": 3.4439121879090493e-06,
      "loss": 0.78260881,
      "num_input_tokens_seen": 47435385,
      "step": 2212,
      "time_per_iteration": 2.7600314617156982
    },
    {
      "auxiliary_loss_clip": 0.0131266,
      "auxiliary_loss_mlp": 0.01029471,
      "balance_loss_clip": 1.05809152,
      "balance_loss_mlp": 1.02130473,
      "epoch": 0.2660975169843083,
      "flos": 19793100360960.0,
      "grad_norm": 2.3896075704837663,
      "language_loss": 0.83249259,
      "learning_rate": 3.4433730776419082e-06,
      "loss": 0.85591388,
      "num_input_tokens_seen": 47454310,
      "step": 2213,
      "time_per_iteration": 2.6980814933776855
    },
    {
      "auxiliary_loss_clip": 0.0126074,
      "auxiliary_loss_mlp": 0.02578361,
      "balance_loss_clip": 1.05625486,
      "balance_loss_mlp": 1.00024557,
      "epoch": 0.2662177598749474,
      "flos": 29018981750400.0,
      "grad_norm": 4.325716000888862,
      "language_loss": 0.80308843,
      "learning_rate": 3.4428337484162183e-06,
      "loss": 0.84147942,
      "num_input_tokens_seen": 47475120,
      "step": 2214,
      "time_per_iteration": 2.714698553085327
    },
    {
      "auxiliary_loss_clip": 0.01310096,
      "auxiliary_loss_mlp": 0.01031248,
      "balance_loss_clip": 1.05698991,
      "balance_loss_mlp": 1.02292669,
      "epoch": 0.2663380027655865,
      "flos": 21762549118080.0,
      "grad_norm": 2.021624370153554,
      "language_loss": 0.84433997,
      "learning_rate": 3.442294200313797e-06,
      "loss": 0.86775339,
      "num_input_tokens_seen": 47493150,
      "step": 2215,
      "time_per_iteration": 2.694211959838867
    },
    {
      "auxiliary_loss_clip": 0.01108624,
      "auxiliary_loss_mlp": 0.01004433,
      "balance_loss_clip": 1.02962065,
      "balance_loss_mlp": 1.00327051,
      "epoch": 0.26645824565622556,
      "flos": 66980333819520.0,
      "grad_norm": 0.7751258591500861,
      "language_loss": 0.52676392,
      "learning_rate": 3.4417544334164916e-06,
      "loss": 0.54789448,
      "num_input_tokens_seen": 47557295,
      "step": 2216,
      "time_per_iteration": 3.2301928997039795
    },
    {
      "auxiliary_loss_clip": 0.01352753,
      "auxiliary_loss_mlp": 0.01028762,
      "balance_loss_clip": 1.05644512,
      "balance_loss_mlp": 1.02057242,
      "epoch": 0.26657848854686467,
      "flos": 25264198373760.0,
      "grad_norm": 1.677211483779638,
      "language_loss": 0.77777588,
      "learning_rate": 3.4412144478061854e-06,
      "loss": 0.80159104,
      "num_input_tokens_seen": 47579705,
      "step": 2217,
      "time_per_iteration": 2.7869880199432373
    },
    {
      "auxiliary_loss_clip": 0.01548902,
      "auxiliary_loss_mlp": 0.01029101,
      "balance_loss_clip": 1.04621625,
      "balance_loss_mlp": 1.01986861,
      "epoch": 0.2666987314375038,
      "flos": 23696769611520.0,
      "grad_norm": 1.961456495998587,
      "language_loss": 0.75445759,
      "learning_rate": 3.4406742435647925e-06,
      "loss": 0.78023767,
      "num_input_tokens_seen": 47599770,
      "step": 2218,
      "time_per_iteration": 2.9682061672210693
    },
    {
      "auxiliary_loss_clip": 0.0125605,
      "auxiliary_loss_mlp": 0.01028441,
      "balance_loss_clip": 1.06058288,
      "balance_loss_mlp": 1.02078235,
      "epoch": 0.26681897432814283,
      "flos": 27048958375680.0,
      "grad_norm": 1.9920701290668184,
      "language_loss": 0.79104972,
      "learning_rate": 3.440133820774263e-06,
      "loss": 0.81389463,
      "num_input_tokens_seen": 47619580,
      "step": 2219,
      "time_per_iteration": 2.7921273708343506
    },
    {
      "auxiliary_loss_clip": 0.01314106,
      "auxiliary_loss_mlp": 0.01029533,
      "balance_loss_clip": 1.06044245,
      "balance_loss_mlp": 1.02077138,
      "epoch": 0.26693921721878194,
      "flos": 28985944216320.0,
      "grad_norm": 2.622645677879936,
      "language_loss": 0.81705242,
      "learning_rate": 3.439593179516578e-06,
      "loss": 0.84048879,
      "num_input_tokens_seen": 47639490,
      "step": 2220,
      "time_per_iteration": 2.6976125240325928
    },
    {
      "auxiliary_loss_clip": 0.01225672,
      "auxiliary_loss_mlp": 0.0102837,
      "balance_loss_clip": 1.05829585,
      "balance_loss_mlp": 1.01961434,
      "epoch": 0.26705946010942105,
      "flos": 21507834798720.0,
      "grad_norm": 1.71356998757466,
      "language_loss": 0.80897051,
      "learning_rate": 3.4390523198737524e-06,
      "loss": 0.83151096,
      "num_input_tokens_seen": 47658650,
      "step": 2221,
      "time_per_iteration": 3.7354772090911865
    },
    {
      "auxiliary_loss_clip": 0.01213128,
      "auxiliary_loss_mlp": 0.02575314,
      "balance_loss_clip": 1.06489289,
      "balance_loss_mlp": 1.00025105,
      "epoch": 0.2671797030000601,
      "flos": 21471277731840.0,
      "grad_norm": 1.7745544422052872,
      "language_loss": 0.73662317,
      "learning_rate": 3.4385112419278333e-06,
      "loss": 0.77450764,
      "num_input_tokens_seen": 47679875,
      "step": 2222,
      "time_per_iteration": 2.675638198852539
    },
    {
      "auxiliary_loss_clip": 0.01162473,
      "auxiliary_loss_mlp": 0.01002386,
      "balance_loss_clip": 1.02819371,
      "balance_loss_mlp": 1.0012002,
      "epoch": 0.2672999458906992,
      "flos": 64189929767040.0,
      "grad_norm": 0.7959930301610104,
      "language_loss": 0.64838123,
      "learning_rate": 3.4379699457609033e-06,
      "loss": 0.67002988,
      "num_input_tokens_seen": 47737700,
      "step": 2223,
      "time_per_iteration": 4.362829923629761
    },
    {
      "auxiliary_loss_clip": 0.01300583,
      "auxiliary_loss_mlp": 0.01038313,
      "balance_loss_clip": 1.05372369,
      "balance_loss_mlp": 1.02946806,
      "epoch": 0.26742018878133833,
      "flos": 16909042573440.0,
      "grad_norm": 1.8409490324270186,
      "language_loss": 0.8975569,
      "learning_rate": 3.4374284314550755e-06,
      "loss": 0.92094576,
      "num_input_tokens_seen": 47756740,
      "step": 2224,
      "time_per_iteration": 2.6963257789611816
    },
    {
      "auxiliary_loss_clip": 0.01209684,
      "auxiliary_loss_mlp": 0.0103091,
      "balance_loss_clip": 1.06194115,
      "balance_loss_mlp": 1.0228579,
      "epoch": 0.2675404316719774,
      "flos": 20667560964480.0,
      "grad_norm": 2.108961661964907,
      "language_loss": 0.811652,
      "learning_rate": 3.436886699092498e-06,
      "loss": 0.83405799,
      "num_input_tokens_seen": 47775255,
      "step": 2225,
      "time_per_iteration": 2.6519532203674316
    },
    {
      "auxiliary_loss_clip": 0.0121392,
      "auxiliary_loss_mlp": 0.01021699,
      "balance_loss_clip": 1.06305563,
      "balance_loss_mlp": 1.01357532,
      "epoch": 0.2676606745626165,
      "flos": 17485013157120.0,
      "grad_norm": 2.584344598599819,
      "language_loss": 0.71643776,
      "learning_rate": 3.4363447487553502e-06,
      "loss": 0.73879391,
      "num_input_tokens_seen": 47788570,
      "step": 2226,
      "time_per_iteration": 2.5250654220581055
    },
    {
      "auxiliary_loss_clip": 0.01302372,
      "auxiliary_loss_mlp": 0.01028928,
      "balance_loss_clip": 1.05868101,
      "balance_loss_mlp": 1.02060175,
      "epoch": 0.26778091745325555,
      "flos": 27852675143040.0,
      "grad_norm": 4.503296894435826,
      "language_loss": 0.77824223,
      "learning_rate": 3.4358025805258455e-06,
      "loss": 0.80155528,
      "num_input_tokens_seen": 47808275,
      "step": 2227,
      "time_per_iteration": 3.674898386001587
    },
    {
      "auxiliary_loss_clip": 0.01408621,
      "auxiliary_loss_mlp": 0.01033418,
      "balance_loss_clip": 1.05205584,
      "balance_loss_mlp": 1.02504921,
      "epoch": 0.26790116034389466,
      "flos": 20955995176320.0,
      "grad_norm": 2.6743673785916853,
      "language_loss": 0.83585536,
      "learning_rate": 3.435260194486232e-06,
      "loss": 0.8602758,
      "num_input_tokens_seen": 47826245,
      "step": 2228,
      "time_per_iteration": 2.7341690063476562
    },
    {
      "auxiliary_loss_clip": 0.01304135,
      "auxiliary_loss_mlp": 0.01026301,
      "balance_loss_clip": 1.05797482,
      "balance_loss_mlp": 1.01749182,
      "epoch": 0.2680214032345338,
      "flos": 18040659621120.0,
      "grad_norm": 2.940056532592233,
      "language_loss": 0.8286407,
      "learning_rate": 3.4347175907187875e-06,
      "loss": 0.85194498,
      "num_input_tokens_seen": 47843235,
      "step": 2229,
      "time_per_iteration": 2.6963300704956055
    },
    {
      "auxiliary_loss_clip": 0.01255531,
      "auxiliary_loss_mlp": 0.010334,
      "balance_loss_clip": 1.06068635,
      "balance_loss_mlp": 1.02517414,
      "epoch": 0.26814164612517283,
      "flos": 22419427086720.0,
      "grad_norm": 2.042931775374046,
      "language_loss": 0.88015574,
      "learning_rate": 3.4341747693058254e-06,
      "loss": 0.90304512,
      "num_input_tokens_seen": 47861710,
      "step": 2230,
      "time_per_iteration": 2.662877082824707
    },
    {
      "auxiliary_loss_clip": 0.01462907,
      "auxiliary_loss_mlp": 0.01035685,
      "balance_loss_clip": 1.04955673,
      "balance_loss_mlp": 1.02779078,
      "epoch": 0.26826188901581194,
      "flos": 35627371159680.0,
      "grad_norm": 1.844620000122029,
      "language_loss": 0.77217197,
      "learning_rate": 3.4336317303296916e-06,
      "loss": 0.79715788,
      "num_input_tokens_seen": 47882685,
      "step": 2231,
      "time_per_iteration": 3.1450953483581543
    },
    {
      "auxiliary_loss_clip": 0.01252605,
      "auxiliary_loss_mlp": 0.01031132,
      "balance_loss_clip": 1.05693913,
      "balance_loss_mlp": 1.02300191,
      "epoch": 0.26838213190645105,
      "flos": 17639788861440.0,
      "grad_norm": 2.7667070184691696,
      "language_loss": 0.75921911,
      "learning_rate": 3.4330884738727635e-06,
      "loss": 0.78205645,
      "num_input_tokens_seen": 47900860,
      "step": 2232,
      "time_per_iteration": 3.925947427749634
    },
    {
      "auxiliary_loss_clip": 0.0139868,
      "auxiliary_loss_mlp": 0.01032236,
      "balance_loss_clip": 1.05292761,
      "balance_loss_mlp": 1.02353954,
      "epoch": 0.2685023747970901,
      "flos": 22674823764480.0,
      "grad_norm": 2.2646099026373974,
      "language_loss": 0.70839602,
      "learning_rate": 3.4325450000174535e-06,
      "loss": 0.73270518,
      "num_input_tokens_seen": 47917500,
      "step": 2233,
      "time_per_iteration": 2.7861697673797607
    },
    {
      "auxiliary_loss_clip": 0.01402246,
      "auxiliary_loss_mlp": 0.01041333,
      "balance_loss_clip": 1.05507231,
      "balance_loss_mlp": 1.03288722,
      "epoch": 0.2686226176877292,
      "flos": 20120533764480.0,
      "grad_norm": 1.9561881453842593,
      "language_loss": 0.74454737,
      "learning_rate": 3.4320013088462067e-06,
      "loss": 0.76898313,
      "num_input_tokens_seen": 47934860,
      "step": 2234,
      "time_per_iteration": 2.8341996669769287
    },
    {
      "auxiliary_loss_clip": 0.01270666,
      "auxiliary_loss_mlp": 0.01031864,
      "balance_loss_clip": 1.05266333,
      "balance_loss_mlp": 1.02397227,
      "epoch": 0.2687428605783683,
      "flos": 21872040750720.0,
      "grad_norm": 1.5810351264132023,
      "language_loss": 0.81526488,
      "learning_rate": 3.431457400441499e-06,
      "loss": 0.83829021,
      "num_input_tokens_seen": 47955255,
      "step": 2235,
      "time_per_iteration": 2.792424440383911
    },
    {
      "auxiliary_loss_clip": 0.01354092,
      "auxiliary_loss_mlp": 0.01003826,
      "balance_loss_clip": 1.02460051,
      "balance_loss_mlp": 1.00275922,
      "epoch": 0.2688631034690074,
      "flos": 69943320766080.0,
      "grad_norm": 0.9360237582998867,
      "language_loss": 0.60814869,
      "learning_rate": 3.4309132748858424e-06,
      "loss": 0.63172793,
      "num_input_tokens_seen": 48016245,
      "step": 2236,
      "time_per_iteration": 3.421807050704956
    },
    {
      "auxiliary_loss_clip": 0.01252879,
      "auxiliary_loss_mlp": 0.01030009,
      "balance_loss_clip": 1.05873132,
      "balance_loss_mlp": 1.02206326,
      "epoch": 0.2689833463596465,
      "flos": 22856639431680.0,
      "grad_norm": 1.680188141236984,
      "language_loss": 0.84151936,
      "learning_rate": 3.430368932261779e-06,
      "loss": 0.86434829,
      "num_input_tokens_seen": 48036600,
      "step": 2237,
      "time_per_iteration": 3.197848081588745
    },
    {
      "auxiliary_loss_clip": 0.01302308,
      "auxiliary_loss_mlp": 0.01041888,
      "balance_loss_clip": 1.05779815,
      "balance_loss_mlp": 1.03314447,
      "epoch": 0.2691035892502856,
      "flos": 17200242132480.0,
      "grad_norm": 2.5861482592588136,
      "language_loss": 0.75358331,
      "learning_rate": 3.429824372651886e-06,
      "loss": 0.77702528,
      "num_input_tokens_seen": 48054750,
      "step": 2238,
      "time_per_iteration": 2.6434123516082764
    },
    {
      "auxiliary_loss_clip": 0.01410504,
      "auxiliary_loss_mlp": 0.01031163,
      "balance_loss_clip": 1.05348134,
      "balance_loss_mlp": 1.02264571,
      "epoch": 0.26922383214092466,
      "flos": 17747484814080.0,
      "grad_norm": 1.9799642601153071,
      "language_loss": 0.8372665,
      "learning_rate": 3.4292795961387732e-06,
      "loss": 0.86168313,
      "num_input_tokens_seen": 48072650,
      "step": 2239,
      "time_per_iteration": 2.750469207763672
    },
    {
      "auxiliary_loss_clip": 0.01209318,
      "auxiliary_loss_mlp": 0.01027478,
      "balance_loss_clip": 1.06108594,
      "balance_loss_mlp": 1.01903248,
      "epoch": 0.26934407503156377,
      "flos": 16173376122240.0,
      "grad_norm": 2.2431645521100014,
      "language_loss": 0.87833065,
      "learning_rate": 3.4287346028050818e-06,
      "loss": 0.9006986,
      "num_input_tokens_seen": 48088720,
      "step": 2240,
      "time_per_iteration": 2.687189817428589
    },
    {
      "auxiliary_loss_clip": 0.01219999,
      "auxiliary_loss_mlp": 0.01026297,
      "balance_loss_clip": 1.05858684,
      "balance_loss_mlp": 1.0189178,
      "epoch": 0.2694643179222028,
      "flos": 23732895715200.0,
      "grad_norm": 1.6215028833952119,
      "language_loss": 0.79891664,
      "learning_rate": 3.4281893927334866e-06,
      "loss": 0.82137954,
      "num_input_tokens_seen": 48108630,
      "step": 2241,
      "time_per_iteration": 2.7871153354644775
    },
    {
      "auxiliary_loss_clip": 0.01258799,
      "auxiliary_loss_mlp": 0.01031401,
      "balance_loss_clip": 1.06021535,
      "balance_loss_mlp": 1.02278781,
      "epoch": 0.26958456081284193,
      "flos": 24718140840960.0,
      "grad_norm": 2.140392465923096,
      "language_loss": 0.75282848,
      "learning_rate": 3.4276439660066963e-06,
      "loss": 0.77573049,
      "num_input_tokens_seen": 48128330,
      "step": 2242,
      "time_per_iteration": 2.71162748336792
    },
    {
      "auxiliary_loss_clip": 0.01206099,
      "auxiliary_loss_mlp": 0.01035082,
      "balance_loss_clip": 1.05934763,
      "balance_loss_mlp": 1.02723765,
      "epoch": 0.26970480370348104,
      "flos": 18112588606080.0,
      "grad_norm": 2.346018234484839,
      "language_loss": 0.84284461,
      "learning_rate": 3.427098322707452e-06,
      "loss": 0.86525643,
      "num_input_tokens_seen": 48144295,
      "step": 2243,
      "time_per_iteration": 2.5823709964752197
    },
    {
      "auxiliary_loss_clip": 0.0126593,
      "auxiliary_loss_mlp": 0.01032172,
      "balance_loss_clip": 1.06856799,
      "balance_loss_mlp": 1.0236963,
      "epoch": 0.2698250465941201,
      "flos": 10816546250880.0,
      "grad_norm": 2.3044478383670848,
      "language_loss": 0.89533424,
      "learning_rate": 3.426552462918526e-06,
      "loss": 0.91831529,
      "num_input_tokens_seen": 48162230,
      "step": 2244,
      "time_per_iteration": 2.643319606781006
    },
    {
      "auxiliary_loss_clip": 0.01204853,
      "auxiliary_loss_mlp": 0.01041996,
      "balance_loss_clip": 1.06084585,
      "balance_loss_mlp": 1.03420591,
      "epoch": 0.2699452894847592,
      "flos": 17308117653120.0,
      "grad_norm": 3.4533351838510944,
      "language_loss": 0.73375797,
      "learning_rate": 3.426006386722726e-06,
      "loss": 0.75622654,
      "num_input_tokens_seen": 48180290,
      "step": 2245,
      "time_per_iteration": 2.5544040203094482
    },
    {
      "auxiliary_loss_clip": 0.01358604,
      "auxiliary_loss_mlp": 0.01032319,
      "balance_loss_clip": 1.06020951,
      "balance_loss_mlp": 1.02367592,
      "epoch": 0.2700655323753983,
      "flos": 18078150441600.0,
      "grad_norm": 2.285368716320637,
      "language_loss": 0.9237873,
      "learning_rate": 3.4254600942028914e-06,
      "loss": 0.94769651,
      "num_input_tokens_seen": 48198165,
      "step": 2246,
      "time_per_iteration": 2.785475969314575
    },
    {
      "auxiliary_loss_clip": 0.01304244,
      "auxiliary_loss_mlp": 0.0103101,
      "balance_loss_clip": 1.05891109,
      "balance_loss_mlp": 1.02291,
      "epoch": 0.2701857752660374,
      "flos": 18186636493440.0,
      "grad_norm": 2.028490073975747,
      "language_loss": 0.82885307,
      "learning_rate": 3.424913585441893e-06,
      "loss": 0.85220563,
      "num_input_tokens_seen": 48216000,
      "step": 2247,
      "time_per_iteration": 3.983516216278076
    },
    {
      "auxiliary_loss_clip": 0.0125288,
      "auxiliary_loss_mlp": 0.01032683,
      "balance_loss_clip": 1.05833101,
      "balance_loss_mlp": 1.02404642,
      "epoch": 0.2703060181566765,
      "flos": 16319496648960.0,
      "grad_norm": 5.292291579007133,
      "language_loss": 0.87729412,
      "learning_rate": 3.4243668605226374e-06,
      "loss": 0.9001497,
      "num_input_tokens_seen": 48233025,
      "step": 2248,
      "time_per_iteration": 3.4898061752319336
    },
    {
      "auxiliary_loss_clip": 0.01353987,
      "auxiliary_loss_mlp": 0.01029161,
      "balance_loss_clip": 1.05725956,
      "balance_loss_mlp": 1.02070296,
      "epoch": 0.2704262610473156,
      "flos": 19572357329280.0,
      "grad_norm": 5.895027134945218,
      "language_loss": 0.82784092,
      "learning_rate": 3.423819919528061e-06,
      "loss": 0.85167241,
      "num_input_tokens_seen": 48251110,
      "step": 2249,
      "time_per_iteration": 2.7674291133880615
    },
    {
      "auxiliary_loss_clip": 0.01408611,
      "auxiliary_loss_mlp": 0.01032921,
      "balance_loss_clip": 1.05144417,
      "balance_loss_mlp": 1.02413511,
      "epoch": 0.27054650393795465,
      "flos": 20740746925440.0,
      "grad_norm": 2.21814528994253,
      "language_loss": 0.78496587,
      "learning_rate": 3.4232727625411355e-06,
      "loss": 0.80938119,
      "num_input_tokens_seen": 48270215,
      "step": 2250,
      "time_per_iteration": 2.7340168952941895
    },
    {
      "auxiliary_loss_clip": 0.01451073,
      "auxiliary_loss_mlp": 0.0103203,
      "balance_loss_clip": 1.0532763,
      "balance_loss_mlp": 1.02412093,
      "epoch": 0.27066674682859376,
      "flos": 18658322916480.0,
      "grad_norm": 1.8205373090918058,
      "language_loss": 0.86430144,
      "learning_rate": 3.4227253896448626e-06,
      "loss": 0.88913238,
      "num_input_tokens_seen": 48288075,
      "step": 2251,
      "time_per_iteration": 2.8119969367980957
    },
    {
      "auxiliary_loss_clip": 0.01203579,
      "auxiliary_loss_mlp": 0.01029932,
      "balance_loss_clip": 1.05783296,
      "balance_loss_mlp": 1.02165902,
      "epoch": 0.2707869897192329,
      "flos": 23002759958400.0,
      "grad_norm": 3.064669861039408,
      "language_loss": 0.8238315,
      "learning_rate": 3.42217780092228e-06,
      "loss": 0.84616661,
      "num_input_tokens_seen": 48306415,
      "step": 2252,
      "time_per_iteration": 3.5525686740875244
    },
    {
      "auxiliary_loss_clip": 0.0127154,
      "auxiliary_loss_mlp": 0.00999432,
      "balance_loss_clip": 1.02813625,
      "balance_loss_mlp": 0.99826378,
      "epoch": 0.27090723260987193,
      "flos": 58323240293760.0,
      "grad_norm": 0.7969878011177948,
      "language_loss": 0.60292602,
      "learning_rate": 3.421629996456456e-06,
      "loss": 0.62563574,
      "num_input_tokens_seen": 48365035,
      "step": 2253,
      "time_per_iteration": 3.2010695934295654
    },
    {
      "auxiliary_loss_clip": 0.01255509,
      "auxiliary_loss_mlp": 0.0103376,
      "balance_loss_clip": 1.05698931,
      "balance_loss_mlp": 1.02588606,
      "epoch": 0.27102747550051104,
      "flos": 11984540797440.0,
      "grad_norm": 3.0264380539318503,
      "language_loss": 0.82891726,
      "learning_rate": 3.421081976330491e-06,
      "loss": 0.85180998,
      "num_input_tokens_seen": 48383550,
      "step": 2254,
      "time_per_iteration": 2.6774861812591553
    },
    {
      "auxiliary_loss_clip": 0.01302734,
      "auxiliary_loss_mlp": 0.01026405,
      "balance_loss_clip": 1.05590713,
      "balance_loss_mlp": 1.0182451,
      "epoch": 0.27114771839115015,
      "flos": 19900401264000.0,
      "grad_norm": 1.9296032878570766,
      "language_loss": 0.88061094,
      "learning_rate": 3.4205337406275207e-06,
      "loss": 0.90390235,
      "num_input_tokens_seen": 48403670,
      "step": 2255,
      "time_per_iteration": 2.6691739559173584
    },
    {
      "auxiliary_loss_clip": 0.01207446,
      "auxiliary_loss_mlp": 0.01027966,
      "balance_loss_clip": 1.05979681,
      "balance_loss_mlp": 1.01920366,
      "epoch": 0.2712679612817892,
      "flos": 18331966920960.0,
      "grad_norm": 2.5568780004767495,
      "language_loss": 0.75408423,
      "learning_rate": 3.4199852894307114e-06,
      "loss": 0.77643836,
      "num_input_tokens_seen": 48420420,
      "step": 2256,
      "time_per_iteration": 2.6249403953552246
    },
    {
      "auxiliary_loss_clip": 0.01462064,
      "auxiliary_loss_mlp": 0.01039577,
      "balance_loss_clip": 1.05843568,
      "balance_loss_mlp": 1.0310533,
      "epoch": 0.2713882041724283,
      "flos": 24460302038400.0,
      "grad_norm": 2.2771082053113543,
      "language_loss": 0.78607541,
      "learning_rate": 3.419436622823262e-06,
      "loss": 0.81109178,
      "num_input_tokens_seen": 48441140,
      "step": 2257,
      "time_per_iteration": 2.815152406692505
    },
    {
      "auxiliary_loss_clip": 0.0130543,
      "auxiliary_loss_mlp": 0.0102937,
      "balance_loss_clip": 1.05677223,
      "balance_loss_mlp": 1.02108467,
      "epoch": 0.27150844706306737,
      "flos": 23039317025280.0,
      "grad_norm": 1.941425020564697,
      "language_loss": 0.74291301,
      "learning_rate": 3.4188877408884063e-06,
      "loss": 0.76626098,
      "num_input_tokens_seen": 48461845,
      "step": 2258,
      "time_per_iteration": 3.6279125213623047
    },
    {
      "auxiliary_loss_clip": 0.01302597,
      "auxiliary_loss_mlp": 0.01029717,
      "balance_loss_clip": 1.05608165,
      "balance_loss_mlp": 1.02193832,
      "epoch": 0.2716286899537065,
      "flos": 22563644192640.0,
      "grad_norm": 2.6410487979879003,
      "language_loss": 0.64943308,
      "learning_rate": 3.4183386437094088e-06,
      "loss": 0.67275625,
      "num_input_tokens_seen": 48478510,
      "step": 2259,
      "time_per_iteration": 2.7657570838928223
    },
    {
      "auxiliary_loss_clip": 0.01307333,
      "auxiliary_loss_mlp": 0.01028425,
      "balance_loss_clip": 1.05589485,
      "balance_loss_mlp": 1.02043247,
      "epoch": 0.2717489328443456,
      "flos": 13115044523520.0,
      "grad_norm": 2.200527518705088,
      "language_loss": 0.82473743,
      "learning_rate": 3.417789331369565e-06,
      "loss": 0.84809506,
      "num_input_tokens_seen": 48494300,
      "step": 2260,
      "time_per_iteration": 2.629396438598633
    },
    {
      "auxiliary_loss_clip": 0.01210718,
      "auxiliary_loss_mlp": 0.01032266,
      "balance_loss_clip": 1.06191754,
      "balance_loss_mlp": 1.023862,
      "epoch": 0.27186917573498465,
      "flos": 29278688060160.0,
      "grad_norm": 11.862988576049597,
      "language_loss": 0.91308522,
      "learning_rate": 3.4172398039522088e-06,
      "loss": 0.93551505,
      "num_input_tokens_seen": 48515585,
      "step": 2261,
      "time_per_iteration": 2.6797196865081787
    },
    {
      "auxiliary_loss_clip": 0.01258655,
      "auxiliary_loss_mlp": 0.01031603,
      "balance_loss_clip": 1.05889058,
      "balance_loss_mlp": 1.02235258,
      "epoch": 0.27198941862562376,
      "flos": 26032220000640.0,
      "grad_norm": 5.097336061987405,
      "language_loss": 0.79734349,
      "learning_rate": 3.4166900615407e-06,
      "loss": 0.82024616,
      "num_input_tokens_seen": 48533500,
      "step": 2262,
      "time_per_iteration": 2.724379539489746
    },
    {
      "auxiliary_loss_clip": 0.01255626,
      "auxiliary_loss_mlp": 0.01028015,
      "balance_loss_clip": 1.06023502,
      "balance_loss_mlp": 1.0202719,
      "epoch": 0.27210966151626287,
      "flos": 32780983760640.0,
      "grad_norm": 2.174030734026544,
      "language_loss": 0.75576943,
      "learning_rate": 3.416140104218436e-06,
      "loss": 0.77860582,
      "num_input_tokens_seen": 48552865,
      "step": 2263,
      "time_per_iteration": 2.774895429611206
    },
    {
      "auxiliary_loss_clip": 0.01205469,
      "auxiliary_loss_mlp": 0.02520682,
      "balance_loss_clip": 1.02429676,
      "balance_loss_mlp": 0.9997322,
      "epoch": 0.2722299044069019,
      "flos": 65471043219840.0,
      "grad_norm": 0.8407574266846619,
      "language_loss": 0.69649374,
      "learning_rate": 3.4155899320688437e-06,
      "loss": 0.73375523,
      "num_input_tokens_seen": 48618940,
      "step": 2264,
      "time_per_iteration": 3.2911365032196045
    },
    {
      "auxiliary_loss_clip": 0.0145579,
      "auxiliary_loss_mlp": 0.01027826,
      "balance_loss_clip": 1.05636597,
      "balance_loss_mlp": 1.0197612,
      "epoch": 0.27235014729754103,
      "flos": 15334143782400.0,
      "grad_norm": 2.099892700934939,
      "language_loss": 0.74424922,
      "learning_rate": 3.415039545175384e-06,
      "loss": 0.76908535,
      "num_input_tokens_seen": 48634665,
      "step": 2265,
      "time_per_iteration": 2.716639518737793
    },
    {
      "auxiliary_loss_clip": 0.01257575,
      "auxiliary_loss_mlp": 0.01029652,
      "balance_loss_clip": 1.05835247,
      "balance_loss_mlp": 1.02154005,
      "epoch": 0.27247039018818014,
      "flos": 21872363973120.0,
      "grad_norm": 1.9833147999141445,
      "language_loss": 0.65270567,
      "learning_rate": 3.414488943621551e-06,
      "loss": 0.67557794,
      "num_input_tokens_seen": 48653330,
      "step": 2266,
      "time_per_iteration": 2.672616720199585
    },
    {
      "auxiliary_loss_clip": 0.0126816,
      "auxiliary_loss_mlp": 0.01037037,
      "balance_loss_clip": 1.06794882,
      "balance_loss_mlp": 1.02826309,
      "epoch": 0.2725906330788192,
      "flos": 18695490514560.0,
      "grad_norm": 2.2244409880949725,
      "language_loss": 0.73618203,
      "learning_rate": 3.41393812749087e-06,
      "loss": 0.75923395,
      "num_input_tokens_seen": 48671375,
      "step": 2267,
      "time_per_iteration": 2.6146247386932373
    },
    {
      "auxiliary_loss_clip": 0.0130513,
      "auxiliary_loss_mlp": 0.01033633,
      "balance_loss_clip": 1.06307554,
      "balance_loss_mlp": 1.02508616,
      "epoch": 0.2727108759694583,
      "flos": 17886099398400.0,
      "grad_norm": 2.294904595109277,
      "language_loss": 0.71653467,
      "learning_rate": 3.4133870968668984e-06,
      "loss": 0.73992234,
      "num_input_tokens_seen": 48686175,
      "step": 2268,
      "time_per_iteration": 2.831547498703003
    },
    {
      "auxiliary_loss_clip": 0.01305165,
      "auxiliary_loss_mlp": 0.01027742,
      "balance_loss_clip": 1.05695605,
      "balance_loss_mlp": 1.01972497,
      "epoch": 0.2728311188600974,
      "flos": 24461666755200.0,
      "grad_norm": 2.4127143825651167,
      "language_loss": 0.78516936,
      "learning_rate": 3.412835851833229e-06,
      "loss": 0.80849844,
      "num_input_tokens_seen": 48708370,
      "step": 2269,
      "time_per_iteration": 2.6964430809020996
    },
    {
      "auxiliary_loss_clip": 0.01261727,
      "auxiliary_loss_mlp": 0.01030723,
      "balance_loss_clip": 1.06378961,
      "balance_loss_mlp": 1.02195501,
      "epoch": 0.2729513617507365,
      "flos": 30993314757120.0,
      "grad_norm": 4.566889251546843,
      "language_loss": 0.78230941,
      "learning_rate": 3.4122843924734834e-06,
      "loss": 0.80523396,
      "num_input_tokens_seen": 48730670,
      "step": 2270,
      "time_per_iteration": 2.7146248817443848
    },
    {
      "auxiliary_loss_clip": 0.01303149,
      "auxiliary_loss_mlp": 0.01031736,
      "balance_loss_clip": 1.05495811,
      "balance_loss_mlp": 1.02194333,
      "epoch": 0.2730716046413756,
      "flos": 19094637421440.0,
      "grad_norm": 1.8370133314273813,
      "language_loss": 0.88351017,
      "learning_rate": 3.411732718871319e-06,
      "loss": 0.90685904,
      "num_input_tokens_seen": 48746510,
      "step": 2271,
      "time_per_iteration": 2.636880874633789
    },
    {
      "auxiliary_loss_clip": 0.01209648,
      "auxiliary_loss_mlp": 0.01031893,
      "balance_loss_clip": 1.06534481,
      "balance_loss_mlp": 1.02411449,
      "epoch": 0.27319184753201464,
      "flos": 26944566474240.0,
      "grad_norm": 1.9148348850873054,
      "language_loss": 0.78907281,
      "learning_rate": 3.4111808311104227e-06,
      "loss": 0.81148815,
      "num_input_tokens_seen": 48768825,
      "step": 2272,
      "time_per_iteration": 3.5168445110321045
    },
    {
      "auxiliary_loss_clip": 0.0131471,
      "auxiliary_loss_mlp": 0.01027018,
      "balance_loss_clip": 1.05677986,
      "balance_loss_mlp": 1.01817298,
      "epoch": 0.27331209042265375,
      "flos": 31759828012800.0,
      "grad_norm": 2.0663198542256955,
      "language_loss": 0.69673967,
      "learning_rate": 3.410628729274517e-06,
      "loss": 0.72015691,
      "num_input_tokens_seen": 48790345,
      "step": 2273,
      "time_per_iteration": 2.8559274673461914
    },
    {
      "auxiliary_loss_clip": 0.01304721,
      "auxiliary_loss_mlp": 0.02573047,
      "balance_loss_clip": 1.05755782,
      "balance_loss_mlp": 1.00029016,
      "epoch": 0.27343233331329286,
      "flos": 25739081107200.0,
      "grad_norm": 2.2425006950187187,
      "language_loss": 0.82739151,
      "learning_rate": 3.4100764134473546e-06,
      "loss": 0.86616921,
      "num_input_tokens_seen": 48809630,
      "step": 2274,
      "time_per_iteration": 3.6453490257263184
    },
    {
      "auxiliary_loss_clip": 0.01212133,
      "auxiliary_loss_mlp": 0.01030012,
      "balance_loss_clip": 1.06493711,
      "balance_loss_mlp": 1.02190626,
      "epoch": 0.2735525762039319,
      "flos": 24389414547840.0,
      "grad_norm": 2.4079408620392106,
      "language_loss": 0.85147882,
      "learning_rate": 3.4095238837127215e-06,
      "loss": 0.87390018,
      "num_input_tokens_seen": 48828770,
      "step": 2275,
      "time_per_iteration": 2.7708542346954346
    },
    {
      "auxiliary_loss_clip": 0.01352815,
      "auxiliary_loss_mlp": 0.01023646,
      "balance_loss_clip": 1.05451572,
      "balance_loss_mlp": 1.01553965,
      "epoch": 0.27367281909457103,
      "flos": 14465357527680.0,
      "grad_norm": 6.552964479297912,
      "language_loss": 0.80012208,
      "learning_rate": 3.4089711401544355e-06,
      "loss": 0.82388669,
      "num_input_tokens_seen": 48846365,
      "step": 2276,
      "time_per_iteration": 2.7555465698242188
    },
    {
      "auxiliary_loss_clip": 0.01259444,
      "auxiliary_loss_mlp": 0.0103378,
      "balance_loss_clip": 1.05923212,
      "balance_loss_mlp": 1.02514911,
      "epoch": 0.27379306198521014,
      "flos": 23476996247040.0,
      "grad_norm": 2.795441716519085,
      "language_loss": 0.67652833,
      "learning_rate": 3.4084181828563486e-06,
      "loss": 0.69946063,
      "num_input_tokens_seen": 48863085,
      "step": 2277,
      "time_per_iteration": 2.623030662536621
    },
    {
      "auxiliary_loss_clip": 0.01400126,
      "auxiliary_loss_mlp": 0.0103957,
      "balance_loss_clip": 1.05082119,
      "balance_loss_mlp": 1.03123116,
      "epoch": 0.2739133048758492,
      "flos": 17458152762240.0,
      "grad_norm": 1.8178333614470847,
      "language_loss": 0.70790237,
      "learning_rate": 3.4078650119023428e-06,
      "loss": 0.73229933,
      "num_input_tokens_seen": 48881400,
      "step": 2278,
      "time_per_iteration": 2.7602438926696777
    },
    {
      "auxiliary_loss_clip": 0.01375692,
      "auxiliary_loss_mlp": 0.01032397,
      "balance_loss_clip": 1.05107987,
      "balance_loss_mlp": 1.02329564,
      "epoch": 0.2740335477664883,
      "flos": 19273113123840.0,
      "grad_norm": 4.535963779593559,
      "language_loss": 0.74466717,
      "learning_rate": 3.4073116273763337e-06,
      "loss": 0.76874804,
      "num_input_tokens_seen": 48895845,
      "step": 2279,
      "time_per_iteration": 3.667773723602295
    },
    {
      "auxiliary_loss_clip": 0.01309626,
      "auxiliary_loss_mlp": 0.01036408,
      "balance_loss_clip": 1.05731463,
      "balance_loss_mlp": 1.0270853,
      "epoch": 0.2741537906571274,
      "flos": 26104723603200.0,
      "grad_norm": 2.148122388502644,
      "language_loss": 0.81014013,
      "learning_rate": 3.40675802936227e-06,
      "loss": 0.8336004,
      "num_input_tokens_seen": 48916630,
      "step": 2280,
      "time_per_iteration": 2.760514497756958
    },
    {
      "auxiliary_loss_clip": 0.01302452,
      "auxiliary_loss_mlp": 0.01035613,
      "balance_loss_clip": 1.05764866,
      "balance_loss_mlp": 1.02717853,
      "epoch": 0.27427403354776647,
      "flos": 34164190644480.0,
      "grad_norm": 2.4742830875087085,
      "language_loss": 0.71738565,
      "learning_rate": 3.4062042179441318e-06,
      "loss": 0.74076623,
      "num_input_tokens_seen": 48937100,
      "step": 2281,
      "time_per_iteration": 2.750361204147339
    },
    {
      "auxiliary_loss_clip": 0.01257185,
      "auxiliary_loss_mlp": 0.01031635,
      "balance_loss_clip": 1.0621258,
      "balance_loss_mlp": 1.02323115,
      "epoch": 0.2743942764384056,
      "flos": 18766988536320.0,
      "grad_norm": 2.4317219392131295,
      "language_loss": 0.8054533,
      "learning_rate": 3.4056501932059314e-06,
      "loss": 0.82834148,
      "num_input_tokens_seen": 48955175,
      "step": 2282,
      "time_per_iteration": 2.6658520698547363
    },
    {
      "auxiliary_loss_clip": 0.0110398,
      "auxiliary_loss_mlp": 0.01008248,
      "balance_loss_clip": 1.02697122,
      "balance_loss_mlp": 1.00701416,
      "epoch": 0.2745145193290447,
      "flos": 64904048058240.0,
      "grad_norm": 0.7670006561349062,
      "language_loss": 0.58118641,
      "learning_rate": 3.405095955231715e-06,
      "loss": 0.60230869,
      "num_input_tokens_seen": 49006830,
      "step": 2283,
      "time_per_iteration": 3.120814085006714
    },
    {
      "auxiliary_loss_clip": 0.01259286,
      "auxiliary_loss_mlp": 0.01031585,
      "balance_loss_clip": 1.0567596,
      "balance_loss_mlp": 1.02352643,
      "epoch": 0.27463476221968375,
      "flos": 16136926796160.0,
      "grad_norm": 2.6055711400925707,
      "language_loss": 0.94478101,
      "learning_rate": 3.4045415041055585e-06,
      "loss": 0.96768975,
      "num_input_tokens_seen": 49022470,
      "step": 2284,
      "time_per_iteration": 3.5337467193603516
    },
    {
      "auxiliary_loss_clip": 0.01308847,
      "auxiliary_loss_mlp": 0.01026607,
      "balance_loss_clip": 1.05852675,
      "balance_loss_mlp": 1.01736224,
      "epoch": 0.27475500511032286,
      "flos": 10376712213120.0,
      "grad_norm": 2.275972546818995,
      "language_loss": 0.77737677,
      "learning_rate": 3.4039868399115728e-06,
      "loss": 0.8007313,
      "num_input_tokens_seen": 49037110,
      "step": 2285,
      "time_per_iteration": 2.7079529762268066
    },
    {
      "auxiliary_loss_clip": 0.01459151,
      "auxiliary_loss_mlp": 0.01028782,
      "balance_loss_clip": 1.058079,
      "balance_loss_mlp": 1.02051532,
      "epoch": 0.27487524800096197,
      "flos": 17311062568320.0,
      "grad_norm": 1.8024362534546061,
      "language_loss": 0.80537164,
      "learning_rate": 3.4034319627339003e-06,
      "loss": 0.83025098,
      "num_input_tokens_seen": 49053975,
      "step": 2286,
      "time_per_iteration": 2.7310242652893066
    },
    {
      "auxiliary_loss_clip": 0.01309945,
      "auxiliary_loss_mlp": 0.0103139,
      "balance_loss_clip": 1.05929422,
      "balance_loss_mlp": 1.02293825,
      "epoch": 0.274995490891601,
      "flos": 27120205002240.0,
      "grad_norm": 2.787052680496394,
      "language_loss": 0.69610429,
      "learning_rate": 3.402876872656715e-06,
      "loss": 0.71951771,
      "num_input_tokens_seen": 49072295,
      "step": 2287,
      "time_per_iteration": 2.723740339279175
    },
    {
      "auxiliary_loss_clip": 0.01305445,
      "auxiliary_loss_mlp": 0.01028865,
      "balance_loss_clip": 1.06001043,
      "balance_loss_mlp": 1.02013922,
      "epoch": 0.27511573378224013,
      "flos": 23436093634560.0,
      "grad_norm": 2.2250891606119563,
      "language_loss": 0.89689851,
      "learning_rate": 3.402321569764223e-06,
      "loss": 0.92024171,
      "num_input_tokens_seen": 49091600,
      "step": 2288,
      "time_per_iteration": 2.705022096633911
    },
    {
      "auxiliary_loss_clip": 0.01403404,
      "auxiliary_loss_mlp": 0.02574682,
      "balance_loss_clip": 1.05406153,
      "balance_loss_mlp": 1.00021195,
      "epoch": 0.2752359766728792,
      "flos": 16722019434240.0,
      "grad_norm": 2.2690495251972256,
      "language_loss": 0.83429074,
      "learning_rate": 3.4017660541406635e-06,
      "loss": 0.8740716,
      "num_input_tokens_seen": 49107665,
      "step": 2289,
      "time_per_iteration": 2.8192358016967773
    },
    {
      "auxiliary_loss_clip": 0.01225469,
      "auxiliary_loss_mlp": 0.01034102,
      "balance_loss_clip": 1.05422103,
      "balance_loss_mlp": 1.02563787,
      "epoch": 0.2753562195635183,
      "flos": 25297738698240.0,
      "grad_norm": 1.729412416867722,
      "language_loss": 0.74231815,
      "learning_rate": 3.4012103258703092e-06,
      "loss": 0.76491392,
      "num_input_tokens_seen": 49126420,
      "step": 2290,
      "time_per_iteration": 2.691594123840332
    },
    {
      "auxiliary_loss_clip": 0.01359011,
      "auxiliary_loss_mlp": 0.01035597,
      "balance_loss_clip": 1.05726647,
      "balance_loss_mlp": 1.02627528,
      "epoch": 0.2754764624541574,
      "flos": 27338972785920.0,
      "grad_norm": 2.291411373900019,
      "language_loss": 0.82928795,
      "learning_rate": 3.4006543850374616e-06,
      "loss": 0.85323405,
      "num_input_tokens_seen": 49141470,
      "step": 2291,
      "time_per_iteration": 2.7444300651550293
    },
    {
      "auxiliary_loss_clip": 0.01261731,
      "auxiliary_loss_mlp": 0.010346,
      "balance_loss_clip": 1.05927229,
      "balance_loss_mlp": 1.02623785,
      "epoch": 0.27559670534479647,
      "flos": 17238379397760.0,
      "grad_norm": 1.9279865146350252,
      "language_loss": 0.74859905,
      "learning_rate": 3.400098231726458e-06,
      "loss": 0.77156234,
      "num_input_tokens_seen": 49158570,
      "step": 2292,
      "time_per_iteration": 2.625044584274292
    },
    {
      "auxiliary_loss_clip": 0.01361285,
      "auxiliary_loss_mlp": 0.01035542,
      "balance_loss_clip": 1.05387044,
      "balance_loss_mlp": 1.02661312,
      "epoch": 0.2757169482354356,
      "flos": 21939085486080.0,
      "grad_norm": 1.8380433769050215,
      "language_loss": 0.87242019,
      "learning_rate": 3.3995418660216657e-06,
      "loss": 0.89638841,
      "num_input_tokens_seen": 49176025,
      "step": 2293,
      "time_per_iteration": 2.6758062839508057
    },
    {
      "auxiliary_loss_clip": 0.01217116,
      "auxiliary_loss_mlp": 0.01030759,
      "balance_loss_clip": 1.06385732,
      "balance_loss_mlp": 1.02186596,
      "epoch": 0.2758371911260747,
      "flos": 20850669521280.0,
      "grad_norm": 3.223407809559123,
      "language_loss": 0.80401707,
      "learning_rate": 3.3989852880074848e-06,
      "loss": 0.82649589,
      "num_input_tokens_seen": 49197455,
      "step": 2294,
      "time_per_iteration": 2.6269705295562744
    },
    {
      "auxiliary_loss_clip": 0.01224672,
      "auxiliary_loss_mlp": 0.01005571,
      "balance_loss_clip": 1.03851163,
      "balance_loss_mlp": 1.00430763,
      "epoch": 0.27595743401671374,
      "flos": 69269063592960.0,
      "grad_norm": 0.7387363283164082,
      "language_loss": 0.60610211,
      "learning_rate": 3.398428497768348e-06,
      "loss": 0.62840456,
      "num_input_tokens_seen": 49262625,
      "step": 2295,
      "time_per_iteration": 3.3732237815856934
    },
    {
      "auxiliary_loss_clip": 0.0136203,
      "auxiliary_loss_mlp": 0.01039415,
      "balance_loss_clip": 1.05175531,
      "balance_loss_mlp": 1.03123176,
      "epoch": 0.27607767690735285,
      "flos": 21215019127680.0,
      "grad_norm": 2.016646926325579,
      "language_loss": 0.72628248,
      "learning_rate": 3.3978714953887205e-06,
      "loss": 0.75029695,
      "num_input_tokens_seen": 49282380,
      "step": 2296,
      "time_per_iteration": 2.7861533164978027
    },
    {
      "auxiliary_loss_clip": 0.01389584,
      "auxiliary_loss_mlp": 0.01031234,
      "balance_loss_clip": 1.04646075,
      "balance_loss_mlp": 1.02306199,
      "epoch": 0.27619791979799196,
      "flos": 24825334003200.0,
      "grad_norm": 2.1314316643615028,
      "language_loss": 0.86146021,
      "learning_rate": 3.397314280953098e-06,
      "loss": 0.8856684,
      "num_input_tokens_seen": 49303205,
      "step": 2297,
      "time_per_iteration": 2.7623417377471924
    },
    {
      "auxiliary_loss_clip": 0.01303483,
      "auxiliary_loss_mlp": 0.01030742,
      "balance_loss_clip": 1.05599487,
      "balance_loss_mlp": 1.02287972,
      "epoch": 0.276318162688631,
      "flos": 24753548672640.0,
      "grad_norm": 2.192641948407135,
      "language_loss": 0.80399495,
      "learning_rate": 3.3967568545460108e-06,
      "loss": 0.82733721,
      "num_input_tokens_seen": 49322745,
      "step": 2298,
      "time_per_iteration": 3.685112953186035
    },
    {
      "auxiliary_loss_clip": 0.0125449,
      "auxiliary_loss_mlp": 0.01029564,
      "balance_loss_clip": 1.05833173,
      "balance_loss_mlp": 1.02084351,
      "epoch": 0.27643840557927013,
      "flos": 18150007599360.0,
      "grad_norm": 1.9323695049660583,
      "language_loss": 0.80535918,
      "learning_rate": 3.3961992162520185e-06,
      "loss": 0.82819974,
      "num_input_tokens_seen": 49341370,
      "step": 2299,
      "time_per_iteration": 2.6405513286590576
    },
    {
      "auxiliary_loss_clip": 0.01255391,
      "auxiliary_loss_mlp": 0.01032465,
      "balance_loss_clip": 1.05762959,
      "balance_loss_mlp": 1.02444804,
      "epoch": 0.27655864846990924,
      "flos": 24823933372800.0,
      "grad_norm": 2.1535239087182534,
      "language_loss": 0.72019011,
      "learning_rate": 3.3956413661557156e-06,
      "loss": 0.7430687,
      "num_input_tokens_seen": 49361545,
      "step": 2300,
      "time_per_iteration": 2.6626012325286865
    },
    {
      "auxiliary_loss_clip": 0.01361281,
      "auxiliary_loss_mlp": 0.01032317,
      "balance_loss_clip": 1.05319476,
      "balance_loss_mlp": 1.02356696,
      "epoch": 0.2766788913605483,
      "flos": 20266582464000.0,
      "grad_norm": 2.1986072132945034,
      "language_loss": 0.66475987,
      "learning_rate": 3.3950833043417273e-06,
      "loss": 0.68869591,
      "num_input_tokens_seen": 49379690,
      "step": 2301,
      "time_per_iteration": 3.6959290504455566
    },
    {
      "auxiliary_loss_clip": 0.01262723,
      "auxiliary_loss_mlp": 0.01033516,
      "balance_loss_clip": 1.06317627,
      "balance_loss_mlp": 1.02501321,
      "epoch": 0.2767991342511874,
      "flos": 21470272151040.0,
      "grad_norm": 2.42803580789738,
      "language_loss": 0.73736453,
      "learning_rate": 3.3945250308947105e-06,
      "loss": 0.76032686,
      "num_input_tokens_seen": 49395995,
      "step": 2302,
      "time_per_iteration": 2.8495635986328125
    },
    {
      "auxiliary_loss_clip": 0.01162888,
      "auxiliary_loss_mlp": 0.01001522,
      "balance_loss_clip": 1.03273416,
      "balance_loss_mlp": 1.00038326,
      "epoch": 0.2769193771418265,
      "flos": 66002627571840.0,
      "grad_norm": 1.2825802560359032,
      "language_loss": 0.68265045,
      "learning_rate": 3.3939665458993556e-06,
      "loss": 0.7042945,
      "num_input_tokens_seen": 49450415,
      "step": 2303,
      "time_per_iteration": 3.1996092796325684
    },
    {
      "auxiliary_loss_clip": 0.01361572,
      "auxiliary_loss_mlp": 0.01034867,
      "balance_loss_clip": 1.05411601,
      "balance_loss_mlp": 1.02639771,
      "epoch": 0.27703962003246557,
      "flos": 20704441253760.0,
      "grad_norm": 2.4114730479094764,
      "language_loss": 0.77016473,
      "learning_rate": 3.3934078494403843e-06,
      "loss": 0.79412901,
      "num_input_tokens_seen": 49469990,
      "step": 2304,
      "time_per_iteration": 3.633247137069702
    },
    {
      "auxiliary_loss_clip": 0.01419895,
      "auxiliary_loss_mlp": 0.02577151,
      "balance_loss_clip": 1.05080283,
      "balance_loss_mlp": 1.00028372,
      "epoch": 0.2771598629231047,
      "flos": 22929897219840.0,
      "grad_norm": 2.2849158312904625,
      "language_loss": 0.81152499,
      "learning_rate": 3.3928489416025495e-06,
      "loss": 0.85149544,
      "num_input_tokens_seen": 49490835,
      "step": 2305,
      "time_per_iteration": 2.870325803756714
    },
    {
      "auxiliary_loss_clip": 0.0130234,
      "auxiliary_loss_mlp": 0.01033287,
      "balance_loss_clip": 1.05662942,
      "balance_loss_mlp": 1.0239892,
      "epoch": 0.27728010581374374,
      "flos": 18369457741440.0,
      "grad_norm": 2.379660543833536,
      "language_loss": 0.79250735,
      "learning_rate": 3.392289822470638e-06,
      "loss": 0.81586361,
      "num_input_tokens_seen": 49508815,
      "step": 2306,
      "time_per_iteration": 2.960207939147949
    },
    {
      "auxiliary_loss_clip": 0.01306111,
      "auxiliary_loss_mlp": 0.01034784,
      "balance_loss_clip": 1.05655837,
      "balance_loss_mlp": 1.02596283,
      "epoch": 0.27740034870438285,
      "flos": 19427637432960.0,
      "grad_norm": 2.1727557766948538,
      "language_loss": 0.75675231,
      "learning_rate": 3.3917304921294674e-06,
      "loss": 0.78016126,
      "num_input_tokens_seen": 49526980,
      "step": 2307,
      "time_per_iteration": 2.8224289417266846
    },
    {
      "auxiliary_loss_clip": 0.01258817,
      "auxiliary_loss_mlp": 0.01031764,
      "balance_loss_clip": 1.05819726,
      "balance_loss_mlp": 1.02248955,
      "epoch": 0.27752059159502196,
      "flos": 21614776565760.0,
      "grad_norm": 1.5237439060168039,
      "language_loss": 0.80743128,
      "learning_rate": 3.3911709506638876e-06,
      "loss": 0.83033705,
      "num_input_tokens_seen": 49546290,
      "step": 2308,
      "time_per_iteration": 2.633056402206421
    },
    {
      "auxiliary_loss_clip": 0.01346405,
      "auxiliary_loss_mlp": 0.02576368,
      "balance_loss_clip": 1.05051768,
      "balance_loss_mlp": 1.00025272,
      "epoch": 0.277640834485661,
      "flos": 26608011016320.0,
      "grad_norm": 3.6449421104696915,
      "language_loss": 0.80955911,
      "learning_rate": 3.390611198158781e-06,
      "loss": 0.84878677,
      "num_input_tokens_seen": 49564165,
      "step": 2309,
      "time_per_iteration": 2.7712318897247314
    },
    {
      "auxiliary_loss_clip": 0.0121057,
      "auxiliary_loss_mlp": 0.01034601,
      "balance_loss_clip": 1.06045449,
      "balance_loss_mlp": 1.02553558,
      "epoch": 0.2777610773763001,
      "flos": 19492814661120.0,
      "grad_norm": 2.394708770784709,
      "language_loss": 0.90106833,
      "learning_rate": 3.3900512346990612e-06,
      "loss": 0.92352009,
      "num_input_tokens_seen": 49580155,
      "step": 2310,
      "time_per_iteration": 3.7179393768310547
    },
    {
      "auxiliary_loss_clip": 0.01404057,
      "auxiliary_loss_mlp": 0.01030298,
      "balance_loss_clip": 1.04620528,
      "balance_loss_mlp": 1.02119064,
      "epoch": 0.27788132026693924,
      "flos": 38290650001920.0,
      "grad_norm": 1.9441428677055217,
      "language_loss": 0.65949106,
      "learning_rate": 3.389491060369674e-06,
      "loss": 0.68383461,
      "num_input_tokens_seen": 49605830,
      "step": 2311,
      "time_per_iteration": 2.9027559757232666
    },
    {
      "auxiliary_loss_clip": 0.01398195,
      "auxiliary_loss_mlp": 0.01036584,
      "balance_loss_clip": 1.05253983,
      "balance_loss_mlp": 1.02786398,
      "epoch": 0.2780015631575783,
      "flos": 22382546797440.0,
      "grad_norm": 2.1910172031858632,
      "language_loss": 0.90131491,
      "learning_rate": 3.388930675255598e-06,
      "loss": 0.9256627,
      "num_input_tokens_seen": 49625680,
      "step": 2312,
      "time_per_iteration": 2.8384487628936768
    },
    {
      "auxiliary_loss_clip": 0.01310313,
      "auxiliary_loss_mlp": 0.01035113,
      "balance_loss_clip": 1.05699801,
      "balance_loss_mlp": 1.02591586,
      "epoch": 0.2781218060482174,
      "flos": 12203200840320.0,
      "grad_norm": 2.428165474267086,
      "language_loss": 0.79657626,
      "learning_rate": 3.388370079441843e-06,
      "loss": 0.82003045,
      "num_input_tokens_seen": 49641195,
      "step": 2313,
      "time_per_iteration": 2.6483089923858643
    },
    {
      "auxiliary_loss_clip": 0.01354721,
      "auxiliary_loss_mlp": 0.01037969,
      "balance_loss_clip": 1.05797541,
      "balance_loss_mlp": 1.02970743,
      "epoch": 0.2782420489388565,
      "flos": 18107632529280.0,
      "grad_norm": 2.314469058331186,
      "language_loss": 0.9267109,
      "learning_rate": 3.3878092730134505e-06,
      "loss": 0.95063776,
      "num_input_tokens_seen": 49659180,
      "step": 2314,
      "time_per_iteration": 2.7238245010375977
    },
    {
      "auxiliary_loss_clip": 0.0125303,
      "auxiliary_loss_mlp": 0.01031078,
      "balance_loss_clip": 1.05626953,
      "balance_loss_mlp": 1.02211344,
      "epoch": 0.27836229182949557,
      "flos": 18514752255360.0,
      "grad_norm": 1.8742101149903125,
      "language_loss": 0.80758774,
      "learning_rate": 3.3872482560554947e-06,
      "loss": 0.83042884,
      "num_input_tokens_seen": 49677955,
      "step": 2315,
      "time_per_iteration": 2.624131441116333
    },
    {
      "auxiliary_loss_clip": 0.01164463,
      "auxiliary_loss_mlp": 0.00999785,
      "balance_loss_clip": 1.03176749,
      "balance_loss_mlp": 0.99856913,
      "epoch": 0.2784825347201347,
      "flos": 67079230940160.0,
      "grad_norm": 0.7954572740252815,
      "language_loss": 0.56912738,
      "learning_rate": 3.386687028653082e-06,
      "loss": 0.59076989,
      "num_input_tokens_seen": 49740800,
      "step": 2316,
      "time_per_iteration": 3.293079376220703
    },
    {
      "auxiliary_loss_clip": 0.01407443,
      "auxiliary_loss_mlp": 0.01034271,
      "balance_loss_clip": 1.05556273,
      "balance_loss_mlp": 1.0252943,
      "epoch": 0.2786027776107738,
      "flos": 22631119891200.0,
      "grad_norm": 2.3016803554896197,
      "language_loss": 0.85223299,
      "learning_rate": 3.386125590891349e-06,
      "loss": 0.87665009,
      "num_input_tokens_seen": 49757675,
      "step": 2317,
      "time_per_iteration": 2.7876710891723633
    },
    {
      "auxiliary_loss_clip": 0.01298577,
      "auxiliary_loss_mlp": 0.01028252,
      "balance_loss_clip": 1.05310607,
      "balance_loss_mlp": 1.01974022,
      "epoch": 0.27872302050141284,
      "flos": 15778826156160.0,
      "grad_norm": 3.4011118587612637,
      "language_loss": 0.83013904,
      "learning_rate": 3.3855639428554657e-06,
      "loss": 0.85340738,
      "num_input_tokens_seen": 49775205,
      "step": 2318,
      "time_per_iteration": 2.671074151992798
    },
    {
      "auxiliary_loss_clip": 0.01345252,
      "auxiliary_loss_mlp": 0.01032531,
      "balance_loss_clip": 1.05258155,
      "balance_loss_mlp": 1.02437758,
      "epoch": 0.27884326339205195,
      "flos": 22126970551680.0,
      "grad_norm": 3.497007402191324,
      "language_loss": 0.80490738,
      "learning_rate": 3.385002084630635e-06,
      "loss": 0.82868522,
      "num_input_tokens_seen": 49794175,
      "step": 2319,
      "time_per_iteration": 2.716866970062256
    },
    {
      "auxiliary_loss_clip": 0.01261586,
      "auxiliary_loss_mlp": 0.0103508,
      "balance_loss_clip": 1.05813766,
      "balance_loss_mlp": 1.02619934,
      "epoch": 0.278963506282691,
      "flos": 20558715776640.0,
      "grad_norm": 2.4598548283145116,
      "language_loss": 0.84977657,
      "learning_rate": 3.384440016302088e-06,
      "loss": 0.87274325,
      "num_input_tokens_seen": 49812850,
      "step": 2320,
      "time_per_iteration": 2.609208345413208
    },
    {
      "auxiliary_loss_clip": 0.01252737,
      "auxiliary_loss_mlp": 0.0103666,
      "balance_loss_clip": 1.05728042,
      "balance_loss_mlp": 1.027511,
      "epoch": 0.2790837491733301,
      "flos": 21942928241280.0,
      "grad_norm": 2.286317959728214,
      "language_loss": 0.62319183,
      "learning_rate": 3.3838777379550923e-06,
      "loss": 0.64608586,
      "num_input_tokens_seen": 49832295,
      "step": 2321,
      "time_per_iteration": 2.721670389175415
    },
    {
      "auxiliary_loss_clip": 0.01219754,
      "auxiliary_loss_mlp": 0.0103904,
      "balance_loss_clip": 1.05737388,
      "balance_loss_mlp": 1.03128529,
      "epoch": 0.27920399206396923,
      "flos": 26286790665600.0,
      "grad_norm": 3.0517686408914924,
      "language_loss": 0.78709197,
      "learning_rate": 3.383315249674944e-06,
      "loss": 0.80967993,
      "num_input_tokens_seen": 49850860,
      "step": 2322,
      "time_per_iteration": 2.7195305824279785
    },
    {
      "auxiliary_loss_clip": 0.01356609,
      "auxiliary_loss_mlp": 0.01035896,
      "balance_loss_clip": 1.05670846,
      "balance_loss_mlp": 1.02718163,
      "epoch": 0.2793242349546083,
      "flos": 25400981364480.0,
      "grad_norm": 2.139985678189235,
      "language_loss": 0.86035407,
      "learning_rate": 3.3827525515469715e-06,
      "loss": 0.88427913,
      "num_input_tokens_seen": 49865765,
      "step": 2323,
      "time_per_iteration": 2.7428858280181885
    },
    {
      "auxiliary_loss_clip": 0.01348653,
      "auxiliary_loss_mlp": 0.01025979,
      "balance_loss_clip": 1.05197883,
      "balance_loss_mlp": 1.01735973,
      "epoch": 0.2794444778452474,
      "flos": 20850346298880.0,
      "grad_norm": 1.9901661670421955,
      "language_loss": 0.70654619,
      "learning_rate": 3.3821896436565367e-06,
      "loss": 0.7302925,
      "num_input_tokens_seen": 49885425,
      "step": 2324,
      "time_per_iteration": 3.6036741733551025
    },
    {
      "auxiliary_loss_clip": 0.01259656,
      "auxiliary_loss_mlp": 0.01033121,
      "balance_loss_clip": 1.06063962,
      "balance_loss_mlp": 1.02511573,
      "epoch": 0.2795647207358865,
      "flos": 21576244250880.0,
      "grad_norm": 1.759896654550285,
      "language_loss": 0.70373416,
      "learning_rate": 3.381626526089032e-06,
      "loss": 0.72666192,
      "num_input_tokens_seen": 49904990,
      "step": 2325,
      "time_per_iteration": 2.6391217708587646
    },
    {
      "auxiliary_loss_clip": 0.0130509,
      "auxiliary_loss_mlp": 0.0103079,
      "balance_loss_clip": 1.05282092,
      "balance_loss_mlp": 1.02202177,
      "epoch": 0.27968496362652556,
      "flos": 21471744608640.0,
      "grad_norm": 1.9565186118586222,
      "language_loss": 0.78650677,
      "learning_rate": 3.3810631989298815e-06,
      "loss": 0.80986559,
      "num_input_tokens_seen": 49924600,
      "step": 2326,
      "time_per_iteration": 3.573195219039917
    },
    {
      "auxiliary_loss_clip": 0.01410541,
      "auxiliary_loss_mlp": 0.01039027,
      "balance_loss_clip": 1.05362976,
      "balance_loss_mlp": 1.02992511,
      "epoch": 0.2798052065171647,
      "flos": 23258695340160.0,
      "grad_norm": 2.5454961563517635,
      "language_loss": 0.84228992,
      "learning_rate": 3.3804996622645423e-06,
      "loss": 0.86678565,
      "num_input_tokens_seen": 49942600,
      "step": 2327,
      "time_per_iteration": 2.7171247005462646
    },
    {
      "auxiliary_loss_clip": 0.01205614,
      "auxiliary_loss_mlp": 0.01036762,
      "balance_loss_clip": 1.05858064,
      "balance_loss_mlp": 1.02807164,
      "epoch": 0.2799254494078038,
      "flos": 21539328048000.0,
      "grad_norm": 2.557392275389734,
      "language_loss": 0.89461386,
      "learning_rate": 3.3799359161785015e-06,
      "loss": 0.91703761,
      "num_input_tokens_seen": 49962250,
      "step": 2328,
      "time_per_iteration": 2.7149624824523926
    },
    {
      "auxiliary_loss_clip": 0.0125541,
      "auxiliary_loss_mlp": 0.01034846,
      "balance_loss_clip": 1.05908549,
      "balance_loss_mlp": 1.02589357,
      "epoch": 0.28004569229844284,
      "flos": 26393912000640.0,
      "grad_norm": 1.575765000828965,
      "language_loss": 0.85622519,
      "learning_rate": 3.3793719607572798e-06,
      "loss": 0.87912774,
      "num_input_tokens_seen": 49983215,
      "step": 2329,
      "time_per_iteration": 2.6982479095458984
    },
    {
      "auxiliary_loss_clip": 0.01295713,
      "auxiliary_loss_mlp": 0.0103595,
      "balance_loss_clip": 1.05271423,
      "balance_loss_mlp": 1.02753329,
      "epoch": 0.28016593518908195,
      "flos": 33547676584320.0,
      "grad_norm": 3.05355724607244,
      "language_loss": 0.77224147,
      "learning_rate": 3.378807796086428e-06,
      "loss": 0.79555809,
      "num_input_tokens_seen": 50006075,
      "step": 2330,
      "time_per_iteration": 2.7917513847351074
    },
    {
      "auxiliary_loss_clip": 0.01211686,
      "auxiliary_loss_mlp": 0.01034843,
      "balance_loss_clip": 1.06192243,
      "balance_loss_mlp": 1.02668953,
      "epoch": 0.28028617807972106,
      "flos": 15340823712000.0,
      "grad_norm": 2.5074126165458073,
      "language_loss": 0.77485156,
      "learning_rate": 3.37824342225153e-06,
      "loss": 0.79731679,
      "num_input_tokens_seen": 50022495,
      "step": 2331,
      "time_per_iteration": 3.6638550758361816
    },
    {
      "auxiliary_loss_clip": 0.01409096,
      "auxiliary_loss_mlp": 0.0103385,
      "balance_loss_clip": 1.05766916,
      "balance_loss_mlp": 1.02556455,
      "epoch": 0.2804064209703601,
      "flos": 25520277409920.0,
      "grad_norm": 2.1546448683821646,
      "language_loss": 0.77889991,
      "learning_rate": 3.3776788393382006e-06,
      "loss": 0.80332935,
      "num_input_tokens_seen": 50041975,
      "step": 2332,
      "time_per_iteration": 2.812812566757202
    },
    {
      "auxiliary_loss_clip": 0.01209542,
      "auxiliary_loss_mlp": 0.01031885,
      "balance_loss_clip": 1.06174195,
      "balance_loss_mlp": 1.02347517,
      "epoch": 0.2805266638609992,
      "flos": 29351766280320.0,
      "grad_norm": 3.7943022488935325,
      "language_loss": 0.76549011,
      "learning_rate": 3.3771140474320872e-06,
      "loss": 0.78790438,
      "num_input_tokens_seen": 50061925,
      "step": 2333,
      "time_per_iteration": 2.6724226474761963
    },
    {
      "auxiliary_loss_clip": 0.01362761,
      "auxiliary_loss_mlp": 0.01032231,
      "balance_loss_clip": 1.05763173,
      "balance_loss_mlp": 1.02324283,
      "epoch": 0.28064690675163834,
      "flos": 21463735875840.0,
      "grad_norm": 3.005314974448947,
      "language_loss": 0.79882467,
      "learning_rate": 3.3765490466188664e-06,
      "loss": 0.82277453,
      "num_input_tokens_seen": 50079325,
      "step": 2334,
      "time_per_iteration": 2.69590425491333
    },
    {
      "auxiliary_loss_clip": 0.01356357,
      "auxiliary_loss_mlp": 0.0103719,
      "balance_loss_clip": 1.05336332,
      "balance_loss_mlp": 1.02836204,
      "epoch": 0.2807671496422774,
      "flos": 20995640812800.0,
      "grad_norm": 2.2271239243823366,
      "language_loss": 0.73470414,
      "learning_rate": 3.3759838369842508e-06,
      "loss": 0.75863957,
      "num_input_tokens_seen": 50097400,
      "step": 2335,
      "time_per_iteration": 2.684208393096924
    },
    {
      "auxiliary_loss_clip": 0.01366047,
      "auxiliary_loss_mlp": 0.01031833,
      "balance_loss_clip": 1.06024361,
      "balance_loss_mlp": 1.02313113,
      "epoch": 0.2808873925329165,
      "flos": 21506577822720.0,
      "grad_norm": 2.06975658292684,
      "language_loss": 0.72830254,
      "learning_rate": 3.375418418613981e-06,
      "loss": 0.75228131,
      "num_input_tokens_seen": 50116425,
      "step": 2336,
      "time_per_iteration": 3.583998441696167
    },
    {
      "auxiliary_loss_clip": 0.01310142,
      "auxiliary_loss_mlp": 0.01037178,
      "balance_loss_clip": 1.06020069,
      "balance_loss_mlp": 1.02824295,
      "epoch": 0.28100763542355556,
      "flos": 16070815814400.0,
      "grad_norm": 2.257843173432874,
      "language_loss": 0.83678031,
      "learning_rate": 3.374852791593831e-06,
      "loss": 0.86025351,
      "num_input_tokens_seen": 50132625,
      "step": 2337,
      "time_per_iteration": 2.6838948726654053
    },
    {
      "auxiliary_loss_clip": 0.01328723,
      "auxiliary_loss_mlp": 0.01031246,
      "balance_loss_clip": 1.0542109,
      "balance_loss_mlp": 1.02272224,
      "epoch": 0.28112787831419467,
      "flos": 19062605468160.0,
      "grad_norm": 3.4944370699742398,
      "language_loss": 0.53974295,
      "learning_rate": 3.374286956009605e-06,
      "loss": 0.56334263,
      "num_input_tokens_seen": 50151190,
      "step": 2338,
      "time_per_iteration": 2.7462823390960693
    },
    {
      "auxiliary_loss_clip": 0.01256665,
      "auxiliary_loss_mlp": 0.01036326,
      "balance_loss_clip": 1.06091714,
      "balance_loss_mlp": 1.02814198,
      "epoch": 0.2812481212048338,
      "flos": 12823629482880.0,
      "grad_norm": 2.7179538726934354,
      "language_loss": 0.75567663,
      "learning_rate": 3.3737209119471405e-06,
      "loss": 0.77860653,
      "num_input_tokens_seen": 50167700,
      "step": 2339,
      "time_per_iteration": 2.6496145725250244
    },
    {
      "auxiliary_loss_clip": 0.01263882,
      "auxiliary_loss_mlp": 0.0103609,
      "balance_loss_clip": 1.05994201,
      "balance_loss_mlp": 1.0263803,
      "epoch": 0.28136836409547283,
      "flos": 15633064765440.0,
      "grad_norm": 2.4830097974249385,
      "language_loss": 0.63886046,
      "learning_rate": 3.373154659492306e-06,
      "loss": 0.66186011,
      "num_input_tokens_seen": 50185840,
      "step": 2340,
      "time_per_iteration": 2.660494089126587
    },
    {
      "auxiliary_loss_clip": 0.01224447,
      "auxiliary_loss_mlp": 0.01029482,
      "balance_loss_clip": 1.05896068,
      "balance_loss_mlp": 1.02156615,
      "epoch": 0.28148860698611194,
      "flos": 19933726106880.0,
      "grad_norm": 2.311958879624698,
      "language_loss": 0.8536588,
      "learning_rate": 3.3725881987310016e-06,
      "loss": 0.87619805,
      "num_input_tokens_seen": 50203375,
      "step": 2341,
      "time_per_iteration": 2.635190010070801
    },
    {
      "auxiliary_loss_clip": 0.01304577,
      "auxiliary_loss_mlp": 0.01029679,
      "balance_loss_clip": 1.05497563,
      "balance_loss_mlp": 1.02102447,
      "epoch": 0.28160884987675106,
      "flos": 17457219008640.0,
      "grad_norm": 2.0073503627286753,
      "language_loss": 0.87880731,
      "learning_rate": 3.372021529749159e-06,
      "loss": 0.90214992,
      "num_input_tokens_seen": 50222435,
      "step": 2342,
      "time_per_iteration": 2.7199554443359375
    },
    {
      "auxiliary_loss_clip": 0.01371502,
      "auxiliary_loss_mlp": 0.01028736,
      "balance_loss_clip": 1.05550575,
      "balance_loss_mlp": 1.02043939,
      "epoch": 0.2817290927673901,
      "flos": 16834743290880.0,
      "grad_norm": 2.1129819879504996,
      "language_loss": 0.92188132,
      "learning_rate": 3.3714546526327405e-06,
      "loss": 0.94588369,
      "num_input_tokens_seen": 50240435,
      "step": 2343,
      "time_per_iteration": 2.7650246620178223
    },
    {
      "auxiliary_loss_clip": 0.01359067,
      "auxiliary_loss_mlp": 0.01035228,
      "balance_loss_clip": 1.05386853,
      "balance_loss_mlp": 1.02663291,
      "epoch": 0.2818493356580292,
      "flos": 15414081500160.0,
      "grad_norm": 2.056495612434856,
      "language_loss": 0.88264632,
      "learning_rate": 3.3708875674677423e-06,
      "loss": 0.90658921,
      "num_input_tokens_seen": 50258410,
      "step": 2344,
      "time_per_iteration": 2.731826066970825
    },
    {
      "auxiliary_loss_clip": 0.01233475,
      "auxiliary_loss_mlp": 0.01031737,
      "balance_loss_clip": 1.06134665,
      "balance_loss_mlp": 1.02305818,
      "epoch": 0.28196957854866833,
      "flos": 20412451595520.0,
      "grad_norm": 2.0971311990259847,
      "language_loss": 0.83533764,
      "learning_rate": 3.37032027434019e-06,
      "loss": 0.85798979,
      "num_input_tokens_seen": 50277930,
      "step": 2345,
      "time_per_iteration": 2.6527647972106934
    },
    {
      "auxiliary_loss_clip": 0.01268151,
      "auxiliary_loss_mlp": 0.01034628,
      "balance_loss_clip": 1.0587728,
      "balance_loss_mlp": 1.02426863,
      "epoch": 0.2820898214393074,
      "flos": 19973120348160.0,
      "grad_norm": 207.15586357011608,
      "language_loss": 0.83190668,
      "learning_rate": 3.369752773336141e-06,
      "loss": 0.85493445,
      "num_input_tokens_seen": 50297410,
      "step": 2346,
      "time_per_iteration": 2.6819732189178467
    },
    {
      "auxiliary_loss_clip": 0.01307652,
      "auxiliary_loss_mlp": 0.01033354,
      "balance_loss_clip": 1.05686867,
      "balance_loss_mlp": 1.02435422,
      "epoch": 0.2822100643299465,
      "flos": 22528308188160.0,
      "grad_norm": 1.979928458911732,
      "language_loss": 0.78548181,
      "learning_rate": 3.3691850645416864e-06,
      "loss": 0.80889189,
      "num_input_tokens_seen": 50317120,
      "step": 2347,
      "time_per_iteration": 2.680032253265381
    },
    {
      "auxiliary_loss_clip": 0.01264289,
      "auxiliary_loss_mlp": 0.01034928,
      "balance_loss_clip": 1.05971336,
      "balance_loss_mlp": 1.02629781,
      "epoch": 0.2823303072205856,
      "flos": 11546682007680.0,
      "grad_norm": 2.6206834110930455,
      "language_loss": 0.83364558,
      "learning_rate": 3.368617148042945e-06,
      "loss": 0.85663772,
      "num_input_tokens_seen": 50334790,
      "step": 2348,
      "time_per_iteration": 2.666668176651001
    },
    {
      "auxiliary_loss_clip": 0.01305033,
      "auxiliary_loss_mlp": 0.01032551,
      "balance_loss_clip": 1.05292845,
      "balance_loss_mlp": 1.02354455,
      "epoch": 0.28245055011122466,
      "flos": 18259894281600.0,
      "grad_norm": 1.925541276046068,
      "language_loss": 0.84518659,
      "learning_rate": 3.368049023926071e-06,
      "loss": 0.86856246,
      "num_input_tokens_seen": 50353785,
      "step": 2349,
      "time_per_iteration": 2.654344081878662
    },
    {
      "auxiliary_loss_clip": 0.01260321,
      "auxiliary_loss_mlp": 0.01028092,
      "balance_loss_clip": 1.06191218,
      "balance_loss_mlp": 1.01944363,
      "epoch": 0.2825707930018638,
      "flos": 24608110504320.0,
      "grad_norm": 1.554198648494667,
      "language_loss": 0.83542854,
      "learning_rate": 3.3674806922772476e-06,
      "loss": 0.85831273,
      "num_input_tokens_seen": 50374670,
      "step": 2350,
      "time_per_iteration": 2.6809418201446533
    },
    {
      "auxiliary_loss_clip": 0.01273901,
      "auxiliary_loss_mlp": 0.01031755,
      "balance_loss_clip": 1.05596769,
      "balance_loss_mlp": 1.02360129,
      "epoch": 0.28269103589250283,
      "flos": 25226994862080.0,
      "grad_norm": 1.7361797599195932,
      "language_loss": 0.75082511,
      "learning_rate": 3.3669121531826904e-06,
      "loss": 0.77388167,
      "num_input_tokens_seen": 50395650,
      "step": 2351,
      "time_per_iteration": 3.707632303237915
    },
    {
      "auxiliary_loss_clip": 0.01356351,
      "auxiliary_loss_mlp": 0.01041806,
      "balance_loss_clip": 1.06139255,
      "balance_loss_mlp": 1.03282404,
      "epoch": 0.28281127878314194,
      "flos": 19281552819840.0,
      "grad_norm": 2.1884511172061925,
      "language_loss": 0.83159912,
      "learning_rate": 3.366343406728647e-06,
      "loss": 0.85558057,
      "num_input_tokens_seen": 50415100,
      "step": 2352,
      "time_per_iteration": 2.685370683670044
    },
    {
      "auxiliary_loss_clip": 0.01253362,
      "auxiliary_loss_mlp": 0.01034276,
      "balance_loss_clip": 1.05531955,
      "balance_loss_mlp": 1.02541256,
      "epoch": 0.28293152167378105,
      "flos": 23878405710720.0,
      "grad_norm": 2.2497039379319173,
      "language_loss": 0.68605727,
      "learning_rate": 3.3657744530013946e-06,
      "loss": 0.70893365,
      "num_input_tokens_seen": 50434335,
      "step": 2353,
      "time_per_iteration": 3.537548303604126
    },
    {
      "auxiliary_loss_clip": 0.01264944,
      "auxiliary_loss_mlp": 0.01038289,
      "balance_loss_clip": 1.06162953,
      "balance_loss_mlp": 1.02986097,
      "epoch": 0.2830517645644201,
      "flos": 43866965928960.0,
      "grad_norm": 2.28629676460402,
      "language_loss": 0.71539569,
      "learning_rate": 3.3652052920872437e-06,
      "loss": 0.738428,
      "num_input_tokens_seen": 50457200,
      "step": 2354,
      "time_per_iteration": 2.845522880554199
    },
    {
      "auxiliary_loss_clip": 0.01310438,
      "auxiliary_loss_mlp": 0.01028966,
      "balance_loss_clip": 1.05853653,
      "balance_loss_mlp": 1.0205977,
      "epoch": 0.2831720074550592,
      "flos": 26651750803200.0,
      "grad_norm": 2.061811263023273,
      "language_loss": 0.85616267,
      "learning_rate": 3.3646359240725355e-06,
      "loss": 0.87955666,
      "num_input_tokens_seen": 50476390,
      "step": 2355,
      "time_per_iteration": 2.688249111175537
    },
    {
      "auxiliary_loss_clip": 0.01260225,
      "auxiliary_loss_mlp": 0.02575047,
      "balance_loss_clip": 1.06037068,
      "balance_loss_mlp": 1.00029922,
      "epoch": 0.2832922503456983,
      "flos": 31029979564800.0,
      "grad_norm": 2.2623496620773973,
      "language_loss": 0.67490947,
      "learning_rate": 3.364066349043643e-06,
      "loss": 0.71326214,
      "num_input_tokens_seen": 50497595,
      "step": 2356,
      "time_per_iteration": 3.640286445617676
    },
    {
      "auxiliary_loss_clip": 0.01305896,
      "auxiliary_loss_mlp": 0.0102891,
      "balance_loss_clip": 1.05691886,
      "balance_loss_mlp": 1.02064848,
      "epoch": 0.2834124932363374,
      "flos": 20405699838720.0,
      "grad_norm": 1.6873160645560967,
      "language_loss": 0.82143593,
      "learning_rate": 3.363496567086969e-06,
      "loss": 0.84478396,
      "num_input_tokens_seen": 50514690,
      "step": 2357,
      "time_per_iteration": 2.6608974933624268
    },
    {
      "auxiliary_loss_clip": 0.01211635,
      "auxiliary_loss_mlp": 0.01029738,
      "balance_loss_clip": 1.06339359,
      "balance_loss_mlp": 1.02142954,
      "epoch": 0.2835327361269765,
      "flos": 39384848056320.0,
      "grad_norm": 2.01952887933366,
      "language_loss": 0.76015496,
      "learning_rate": 3.3629265782889506e-06,
      "loss": 0.78256869,
      "num_input_tokens_seen": 50536515,
      "step": 2358,
      "time_per_iteration": 2.7880594730377197
    },
    {
      "auxiliary_loss_clip": 0.01353771,
      "auxiliary_loss_mlp": 0.01033769,
      "balance_loss_clip": 1.05313921,
      "balance_loss_mlp": 1.02540386,
      "epoch": 0.2836529790176156,
      "flos": 30261598801920.0,
      "grad_norm": 2.3759832690772273,
      "language_loss": 0.72106922,
      "learning_rate": 3.362356382736054e-06,
      "loss": 0.74494469,
      "num_input_tokens_seen": 50557120,
      "step": 2359,
      "time_per_iteration": 2.774721384048462
    },
    {
      "auxiliary_loss_clip": 0.01270986,
      "auxiliary_loss_mlp": 0.01028556,
      "balance_loss_clip": 1.05237794,
      "balance_loss_mlp": 1.02022934,
      "epoch": 0.28377322190825466,
      "flos": 12677796264960.0,
      "grad_norm": 2.9278535885640187,
      "language_loss": 0.90567487,
      "learning_rate": 3.361785980514777e-06,
      "loss": 0.92867029,
      "num_input_tokens_seen": 50573320,
      "step": 2360,
      "time_per_iteration": 2.7116382122039795
    },
    {
      "auxiliary_loss_clip": 0.0145461,
      "auxiliary_loss_mlp": 0.0103851,
      "balance_loss_clip": 1.05715036,
      "balance_loss_mlp": 1.02909255,
      "epoch": 0.28389346479889377,
      "flos": 18296666830080.0,
      "grad_norm": 2.0317083771438935,
      "language_loss": 0.76494479,
      "learning_rate": 3.361215371711649e-06,
      "loss": 0.78987598,
      "num_input_tokens_seen": 50592415,
      "step": 2361,
      "time_per_iteration": 2.8017146587371826
    },
    {
      "auxiliary_loss_clip": 0.01347486,
      "auxiliary_loss_mlp": 0.01027308,
      "balance_loss_clip": 1.05471969,
      "balance_loss_mlp": 1.01906443,
      "epoch": 0.2840137076895329,
      "flos": 20406992728320.0,
      "grad_norm": 3.662069296918815,
      "language_loss": 0.83143044,
      "learning_rate": 3.3606445564132326e-06,
      "loss": 0.85517836,
      "num_input_tokens_seen": 50609710,
      "step": 2362,
      "time_per_iteration": 3.669661521911621
    },
    {
      "auxiliary_loss_clip": 0.01211894,
      "auxiliary_loss_mlp": 0.02571786,
      "balance_loss_clip": 1.0636127,
      "balance_loss_mlp": 1.00019491,
      "epoch": 0.28413395058017193,
      "flos": 20048030161920.0,
      "grad_norm": 2.3548569319525687,
      "language_loss": 0.81973541,
      "learning_rate": 3.360073534706118e-06,
      "loss": 0.8575722,
      "num_input_tokens_seen": 50626865,
      "step": 2363,
      "time_per_iteration": 2.625143527984619
    },
    {
      "auxiliary_loss_clip": 0.01313099,
      "auxiliary_loss_mlp": 0.01042116,
      "balance_loss_clip": 1.06110239,
      "balance_loss_mlp": 1.03316879,
      "epoch": 0.28425419347081105,
      "flos": 37663613256960.0,
      "grad_norm": 2.277143994979008,
      "language_loss": 0.76396263,
      "learning_rate": 3.35950230667693e-06,
      "loss": 0.78751481,
      "num_input_tokens_seen": 50648560,
      "step": 2364,
      "time_per_iteration": 2.826960325241089
    },
    {
      "auxiliary_loss_clip": 0.01262126,
      "auxiliary_loss_mlp": 0.01030701,
      "balance_loss_clip": 1.05997765,
      "balance_loss_mlp": 1.02282763,
      "epoch": 0.28437443636145016,
      "flos": 13845072539520.0,
      "grad_norm": 2.251863704565061,
      "language_loss": 0.86404681,
      "learning_rate": 3.358930872412323e-06,
      "loss": 0.88697505,
      "num_input_tokens_seen": 50665725,
      "step": 2365,
      "time_per_iteration": 2.618797779083252
    },
    {
      "auxiliary_loss_clip": 0.01259107,
      "auxiliary_loss_mlp": 0.01031576,
      "balance_loss_clip": 1.06147075,
      "balance_loss_mlp": 1.02328515,
      "epoch": 0.2844946792520892,
      "flos": 22747794243840.0,
      "grad_norm": 1.7900463829841267,
      "language_loss": 0.81065035,
      "learning_rate": 3.3583592319989825e-06,
      "loss": 0.83355719,
      "num_input_tokens_seen": 50685095,
      "step": 2366,
      "time_per_iteration": 2.6055939197540283
    },
    {
      "auxiliary_loss_clip": 0.01270014,
      "auxiliary_loss_mlp": 0.01031849,
      "balance_loss_clip": 1.06261873,
      "balance_loss_mlp": 1.0231111,
      "epoch": 0.2846149221427283,
      "flos": 32415987709440.0,
      "grad_norm": 2.1612976724064503,
      "language_loss": 0.68862122,
      "learning_rate": 3.357787385523627e-06,
      "loss": 0.71163988,
      "num_input_tokens_seen": 50706500,
      "step": 2367,
      "time_per_iteration": 2.803755521774292
    },
    {
      "auxiliary_loss_clip": 0.01461903,
      "auxiliary_loss_mlp": 0.01029467,
      "balance_loss_clip": 1.05118942,
      "balance_loss_mlp": 1.02070475,
      "epoch": 0.2847351650333674,
      "flos": 28475976873600.0,
      "grad_norm": 3.756720248948288,
      "language_loss": 0.82468861,
      "learning_rate": 3.3572153330730048e-06,
      "loss": 0.84960234,
      "num_input_tokens_seen": 50727595,
      "step": 2368,
      "time_per_iteration": 2.909616708755493
    },
    {
      "auxiliary_loss_clip": 0.01270167,
      "auxiliary_loss_mlp": 0.0099973,
      "balance_loss_clip": 1.03343463,
      "balance_loss_mlp": 0.99841255,
      "epoch": 0.2848554079240065,
      "flos": 55753399704960.0,
      "grad_norm": 1.707263767815247,
      "language_loss": 0.64647889,
      "learning_rate": 3.3566430747338956e-06,
      "loss": 0.66917789,
      "num_input_tokens_seen": 50782800,
      "step": 2369,
      "time_per_iteration": 3.2818734645843506
    },
    {
      "auxiliary_loss_clip": 0.01260064,
      "auxiliary_loss_mlp": 0.01033179,
      "balance_loss_clip": 1.05754399,
      "balance_loss_mlp": 1.02373815,
      "epoch": 0.2849756508146456,
      "flos": 11836875985920.0,
      "grad_norm": 2.1115862046720064,
      "language_loss": 0.8691352,
      "learning_rate": 3.35607061059311e-06,
      "loss": 0.89206767,
      "num_input_tokens_seen": 50797730,
      "step": 2370,
      "time_per_iteration": 2.690619707107544
    },
    {
      "auxiliary_loss_clip": 0.0121048,
      "auxiliary_loss_mlp": 0.01030834,
      "balance_loss_clip": 1.0635438,
      "balance_loss_mlp": 1.02238238,
      "epoch": 0.28509589370528465,
      "flos": 25155209531520.0,
      "grad_norm": 3.154948941316112,
      "language_loss": 0.75024807,
      "learning_rate": 3.3554979407374917e-06,
      "loss": 0.77266133,
      "num_input_tokens_seen": 50819840,
      "step": 2371,
      "time_per_iteration": 2.6628963947296143
    },
    {
      "auxiliary_loss_clip": 0.01259944,
      "auxiliary_loss_mlp": 0.01029477,
      "balance_loss_clip": 1.05869114,
      "balance_loss_mlp": 1.02143085,
      "epoch": 0.28521613659592376,
      "flos": 19974808287360.0,
      "grad_norm": 1.8169060607204155,
      "language_loss": 0.7394588,
      "learning_rate": 3.3549250652539134e-06,
      "loss": 0.76235294,
      "num_input_tokens_seen": 50838935,
      "step": 2372,
      "time_per_iteration": 2.6996805667877197
    },
    {
      "auxiliary_loss_clip": 0.01308177,
      "auxiliary_loss_mlp": 0.01033386,
      "balance_loss_clip": 1.0544796,
      "balance_loss_mlp": 1.02387333,
      "epoch": 0.2853363794865629,
      "flos": 23367971491200.0,
      "grad_norm": 1.897202019720728,
      "language_loss": 0.81528211,
      "learning_rate": 3.3543519842292794e-06,
      "loss": 0.83869773,
      "num_input_tokens_seen": 50858590,
      "step": 2373,
      "time_per_iteration": 2.688646078109741
    },
    {
      "auxiliary_loss_clip": 0.01212356,
      "auxiliary_loss_mlp": 0.02574624,
      "balance_loss_clip": 1.06385255,
      "balance_loss_mlp": 1.00024021,
      "epoch": 0.28545662237720193,
      "flos": 19861940776320.0,
      "grad_norm": 3.595441459188916,
      "language_loss": 0.83491933,
      "learning_rate": 3.353778697750527e-06,
      "loss": 0.87278914,
      "num_input_tokens_seen": 50876995,
      "step": 2374,
      "time_per_iteration": 2.6900582313537598
    },
    {
      "auxiliary_loss_clip": 0.0130175,
      "auxiliary_loss_mlp": 0.01028975,
      "balance_loss_clip": 1.05490732,
      "balance_loss_mlp": 1.02042723,
      "epoch": 0.28557686526784104,
      "flos": 23879016241920.0,
      "grad_norm": 5.455733538389054,
      "language_loss": 0.89568418,
      "learning_rate": 3.353205205904622e-06,
      "loss": 0.91899145,
      "num_input_tokens_seen": 50896105,
      "step": 2375,
      "time_per_iteration": 2.7007546424865723
    },
    {
      "auxiliary_loss_clip": 0.01303484,
      "auxiliary_loss_mlp": 0.01033649,
      "balance_loss_clip": 1.05763245,
      "balance_loss_mlp": 1.02474439,
      "epoch": 0.28569710815848015,
      "flos": 44890384233600.0,
      "grad_norm": 1.9745515813161378,
      "language_loss": 0.71873832,
      "learning_rate": 3.3526315087785637e-06,
      "loss": 0.74210954,
      "num_input_tokens_seen": 50917220,
      "step": 2376,
      "time_per_iteration": 2.8914403915405273
    },
    {
      "auxiliary_loss_clip": 0.01392683,
      "auxiliary_loss_mlp": 0.01034391,
      "balance_loss_clip": 1.0530982,
      "balance_loss_mlp": 1.02595675,
      "epoch": 0.2858173510491192,
      "flos": 26829759628800.0,
      "grad_norm": 1.9256781989474228,
      "language_loss": 0.8103832,
      "learning_rate": 3.3520576064593805e-06,
      "loss": 0.83465385,
      "num_input_tokens_seen": 50937175,
      "step": 2377,
      "time_per_iteration": 3.6553120613098145
    },
    {
      "auxiliary_loss_clip": 0.0126304,
      "auxiliary_loss_mlp": 0.01036505,
      "balance_loss_clip": 1.0605613,
      "balance_loss_mlp": 1.02798748,
      "epoch": 0.2859375939397583,
      "flos": 23148916398720.0,
      "grad_norm": 1.6546307272636671,
      "language_loss": 0.81805903,
      "learning_rate": 3.3514834990341337e-06,
      "loss": 0.8410545,
      "num_input_tokens_seen": 50957500,
      "step": 2378,
      "time_per_iteration": 2.6668803691864014
    },
    {
      "auxiliary_loss_clip": 0.0131261,
      "auxiliary_loss_mlp": 0.01033564,
      "balance_loss_clip": 1.05759442,
      "balance_loss_mlp": 1.02528524,
      "epoch": 0.2860578368303974,
      "flos": 12129799397760.0,
      "grad_norm": 2.926940839876949,
      "language_loss": 0.9271307,
      "learning_rate": 3.3509091865899144e-06,
      "loss": 0.95059246,
      "num_input_tokens_seen": 50972690,
      "step": 2379,
      "time_per_iteration": 3.524513006210327
    },
    {
      "auxiliary_loss_clip": 0.01210681,
      "auxiliary_loss_mlp": 0.01034632,
      "balance_loss_clip": 1.06133366,
      "balance_loss_mlp": 1.02554846,
      "epoch": 0.2861780797210365,
      "flos": 19938035738880.0,
      "grad_norm": 1.813035788219474,
      "language_loss": 0.70472074,
      "learning_rate": 3.350334669213846e-06,
      "loss": 0.72717386,
      "num_input_tokens_seen": 50990095,
      "step": 2380,
      "time_per_iteration": 2.6124229431152344
    },
    {
      "auxiliary_loss_clip": 0.01259296,
      "auxiliary_loss_mlp": 0.01035051,
      "balance_loss_clip": 1.06303334,
      "balance_loss_mlp": 1.02726126,
      "epoch": 0.2862983226116756,
      "flos": 27563127609600.0,
      "grad_norm": 3.7202335314661665,
      "language_loss": 0.75558496,
      "learning_rate": 3.3497599469930816e-06,
      "loss": 0.77852839,
      "num_input_tokens_seen": 51008305,
      "step": 2381,
      "time_per_iteration": 2.6928834915161133
    },
    {
      "auxiliary_loss_clip": 0.01207298,
      "auxiliary_loss_mlp": 0.01033607,
      "balance_loss_clip": 1.05916739,
      "balance_loss_mlp": 1.02513075,
      "epoch": 0.28641856550231465,
      "flos": 22053964158720.0,
      "grad_norm": 2.541910336602393,
      "language_loss": 0.83269465,
      "learning_rate": 3.349185020014807e-06,
      "loss": 0.85510367,
      "num_input_tokens_seen": 51025570,
      "step": 2382,
      "time_per_iteration": 2.6738297939300537
    },
    {
      "auxiliary_loss_clip": 0.01260768,
      "auxiliary_loss_mlp": 0.01035973,
      "balance_loss_clip": 1.05834126,
      "balance_loss_mlp": 1.02694893,
      "epoch": 0.28653880839295376,
      "flos": 22378775869440.0,
      "grad_norm": 1.8744138453808687,
      "language_loss": 0.74899179,
      "learning_rate": 3.348609888366237e-06,
      "loss": 0.77195925,
      "num_input_tokens_seen": 51044585,
      "step": 2383,
      "time_per_iteration": 3.6198854446411133
    },
    {
      "auxiliary_loss_clip": 0.01454885,
      "auxiliary_loss_mlp": 0.01032316,
      "balance_loss_clip": 1.05268824,
      "balance_loss_mlp": 1.0242995,
      "epoch": 0.28665905128359287,
      "flos": 23367971491200.0,
      "grad_norm": 2.0969070526246494,
      "language_loss": 0.62688816,
      "learning_rate": 3.348034552134619e-06,
      "loss": 0.65176022,
      "num_input_tokens_seen": 51063990,
      "step": 2384,
      "time_per_iteration": 2.865286111831665
    },
    {
      "auxiliary_loss_clip": 0.01403091,
      "auxiliary_loss_mlp": 0.01034169,
      "balance_loss_clip": 1.05743384,
      "balance_loss_mlp": 1.02609313,
      "epoch": 0.2867792941742319,
      "flos": 20881695893760.0,
      "grad_norm": 2.7576370384801936,
      "language_loss": 0.84305668,
      "learning_rate": 3.3474590114072316e-06,
      "loss": 0.86742932,
      "num_input_tokens_seen": 51081990,
      "step": 2385,
      "time_per_iteration": 2.777210235595703
    },
    {
      "auxiliary_loss_clip": 0.01347508,
      "auxiliary_loss_mlp": 0.01036977,
      "balance_loss_clip": 1.0555563,
      "balance_loss_mlp": 1.0283637,
      "epoch": 0.28689953706487104,
      "flos": 20664005518080.0,
      "grad_norm": 2.2219577042455394,
      "language_loss": 0.82996273,
      "learning_rate": 3.3468832662713836e-06,
      "loss": 0.85380757,
      "num_input_tokens_seen": 51100235,
      "step": 2386,
      "time_per_iteration": 2.7487494945526123
    },
    {
      "auxiliary_loss_clip": 0.01353168,
      "auxiliary_loss_mlp": 0.01031964,
      "balance_loss_clip": 1.0576055,
      "balance_loss_mlp": 1.0232321,
      "epoch": 0.28701977995551015,
      "flos": 12675533708160.0,
      "grad_norm": 2.524083249134626,
      "language_loss": 0.84134609,
      "learning_rate": 3.346307316814415e-06,
      "loss": 0.86519742,
      "num_input_tokens_seen": 51115405,
      "step": 2387,
      "time_per_iteration": 2.6626346111297607
    },
    {
      "auxiliary_loss_clip": 0.0125739,
      "auxiliary_loss_mlp": 0.01032143,
      "balance_loss_clip": 1.06147504,
      "balance_loss_mlp": 1.02348232,
      "epoch": 0.2871400228461492,
      "flos": 21252366293760.0,
      "grad_norm": 1.9854695061457712,
      "language_loss": 0.7589494,
      "learning_rate": 3.3457311631236965e-06,
      "loss": 0.78184474,
      "num_input_tokens_seen": 51136390,
      "step": 2388,
      "time_per_iteration": 3.5655486583709717
    },
    {
      "auxiliary_loss_clip": 0.01298833,
      "auxiliary_loss_mlp": 0.0103661,
      "balance_loss_clip": 1.05470157,
      "balance_loss_mlp": 1.02771139,
      "epoch": 0.2872602657367883,
      "flos": 25119262995840.0,
      "grad_norm": 1.935403461645632,
      "language_loss": 0.84262788,
      "learning_rate": 3.345154805286631e-06,
      "loss": 0.86598235,
      "num_input_tokens_seen": 51156650,
      "step": 2389,
      "time_per_iteration": 2.7266483306884766
    },
    {
      "auxiliary_loss_clip": 0.01253462,
      "auxiliary_loss_mlp": 0.01026231,
      "balance_loss_clip": 1.05624461,
      "balance_loss_mlp": 1.01826763,
      "epoch": 0.2873805086274274,
      "flos": 16646606830080.0,
      "grad_norm": 2.2601901057986913,
      "language_loss": 0.76521611,
      "learning_rate": 3.344578243390651e-06,
      "loss": 0.78801304,
      "num_input_tokens_seen": 51172210,
      "step": 2390,
      "time_per_iteration": 2.628401041030884
    },
    {
      "auxiliary_loss_clip": 0.01304331,
      "auxiliary_loss_mlp": 0.01034489,
      "balance_loss_clip": 1.05862379,
      "balance_loss_mlp": 1.0265733,
      "epoch": 0.2875007515180665,
      "flos": 17420123237760.0,
      "grad_norm": 2.256712608582304,
      "language_loss": 0.78350097,
      "learning_rate": 3.3440014775232206e-06,
      "loss": 0.80688918,
      "num_input_tokens_seen": 51190265,
      "step": 2391,
      "time_per_iteration": 2.682675838470459
    },
    {
      "auxiliary_loss_clip": 0.01271624,
      "auxiliary_loss_mlp": 0.01032977,
      "balance_loss_clip": 1.0566982,
      "balance_loss_mlp": 1.02525198,
      "epoch": 0.2876209944087056,
      "flos": 23434190213760.0,
      "grad_norm": 2.274558894428822,
      "language_loss": 0.71503782,
      "learning_rate": 3.343424507771834e-06,
      "loss": 0.73808384,
      "num_input_tokens_seen": 51208475,
      "step": 2392,
      "time_per_iteration": 2.722459316253662
    },
    {
      "auxiliary_loss_clip": 0.01351691,
      "auxiliary_loss_mlp": 0.01030223,
      "balance_loss_clip": 1.05509448,
      "balance_loss_mlp": 1.02261114,
      "epoch": 0.2877412372993447,
      "flos": 13735509079680.0,
      "grad_norm": 2.0733116085087766,
      "language_loss": 0.86842895,
      "learning_rate": 3.342847334224018e-06,
      "loss": 0.89224815,
      "num_input_tokens_seen": 51225875,
      "step": 2393,
      "time_per_iteration": 2.6888420581817627
    },
    {
      "auxiliary_loss_clip": 0.01163217,
      "auxiliary_loss_mlp": 0.01003986,
      "balance_loss_clip": 1.03518271,
      "balance_loss_mlp": 1.0027287,
      "epoch": 0.28786148018998375,
      "flos": 58079695104000.0,
      "grad_norm": 0.9562889300801051,
      "language_loss": 0.6238997,
      "learning_rate": 3.342269956967329e-06,
      "loss": 0.64557171,
      "num_input_tokens_seen": 51287780,
      "step": 2394,
      "time_per_iteration": 3.3532893657684326
    },
    {
      "auxiliary_loss_clip": 0.012604,
      "auxiliary_loss_mlp": 0.01031677,
      "balance_loss_clip": 1.05892479,
      "balance_loss_mlp": 1.0225991,
      "epoch": 0.28798172308062286,
      "flos": 23435052140160.0,
      "grad_norm": 2.7244210861774816,
      "language_loss": 0.71975374,
      "learning_rate": 3.341692376089355e-06,
      "loss": 0.74267447,
      "num_input_tokens_seen": 51303335,
      "step": 2395,
      "time_per_iteration": 2.677497386932373
    },
    {
      "auxiliary_loss_clip": 0.0125377,
      "auxiliary_loss_mlp": 0.01026823,
      "balance_loss_clip": 1.05914521,
      "balance_loss_mlp": 1.01900911,
      "epoch": 0.288101965971262,
      "flos": 25110033200640.0,
      "grad_norm": 3.1194070683483996,
      "language_loss": 0.84588116,
      "learning_rate": 3.3411145916777146e-06,
      "loss": 0.86868709,
      "num_input_tokens_seen": 51317495,
      "step": 2396,
      "time_per_iteration": 2.7279052734375
    },
    {
      "auxiliary_loss_clip": 0.01301978,
      "auxiliary_loss_mlp": 0.01030855,
      "balance_loss_clip": 1.05400419,
      "balance_loss_mlp": 1.02277887,
      "epoch": 0.28822220886190103,
      "flos": 16252559654400.0,
      "grad_norm": 2.327500724067234,
      "language_loss": 0.91304159,
      "learning_rate": 3.3405366038200566e-06,
      "loss": 0.9363699,
      "num_input_tokens_seen": 51336430,
      "step": 2397,
      "time_per_iteration": 2.6768481731414795
    },
    {
      "auxiliary_loss_clip": 0.01307516,
      "auxiliary_loss_mlp": 0.01036862,
      "balance_loss_clip": 1.0595144,
      "balance_loss_mlp": 1.02882755,
      "epoch": 0.28834245175254014,
      "flos": 24535642815360.0,
      "grad_norm": 2.8259063526231087,
      "language_loss": 0.85119009,
      "learning_rate": 3.3399584126040617e-06,
      "loss": 0.87463391,
      "num_input_tokens_seen": 51355930,
      "step": 2398,
      "time_per_iteration": 2.800168752670288
    },
    {
      "auxiliary_loss_clip": 0.01206942,
      "auxiliary_loss_mlp": 0.02567821,
      "balance_loss_clip": 1.05947375,
      "balance_loss_mlp": 1.0000639,
      "epoch": 0.2884626946431792,
      "flos": 24571445696640.0,
      "grad_norm": 2.1625489360353978,
      "language_loss": 0.9134798,
      "learning_rate": 3.339380018117441e-06,
      "loss": 0.95122743,
      "num_input_tokens_seen": 51376765,
      "step": 2399,
      "time_per_iteration": 2.633810520172119
    },
    {
      "auxiliary_loss_clip": 0.01250424,
      "auxiliary_loss_mlp": 0.0102952,
      "balance_loss_clip": 1.05866611,
      "balance_loss_mlp": 1.02130055,
      "epoch": 0.2885829375338183,
      "flos": 16544657053440.0,
      "grad_norm": 2.989422020357337,
      "language_loss": 0.77830386,
      "learning_rate": 3.3388014204479366e-06,
      "loss": 0.80110323,
      "num_input_tokens_seen": 51394570,
      "step": 2400,
      "time_per_iteration": 2.682804822921753
    },
    {
      "auxiliary_loss_clip": 0.0120944,
      "auxiliary_loss_mlp": 0.0103598,
      "balance_loss_clip": 1.06025994,
      "balance_loss_mlp": 1.02748668,
      "epoch": 0.2887031804244574,
      "flos": 24061226958720.0,
      "grad_norm": 2.183695888338757,
      "language_loss": 0.91801858,
      "learning_rate": 3.338222619683321e-06,
      "loss": 0.94047284,
      "num_input_tokens_seen": 51414535,
      "step": 2401,
      "time_per_iteration": 2.6036267280578613
    },
    {
      "auxiliary_loss_clip": 0.01307859,
      "auxiliary_loss_mlp": 0.01035453,
      "balance_loss_clip": 1.05959463,
      "balance_loss_mlp": 1.02647018,
      "epoch": 0.2888234233150965,
      "flos": 23330696152320.0,
      "grad_norm": 2.7442425722314154,
      "language_loss": 0.73687589,
      "learning_rate": 3.337643615911398e-06,
      "loss": 0.76030898,
      "num_input_tokens_seen": 51434160,
      "step": 2402,
      "time_per_iteration": 2.7123374938964844
    },
    {
      "auxiliary_loss_clip": 0.01256941,
      "auxiliary_loss_mlp": 0.01033596,
      "balance_loss_clip": 1.05711436,
      "balance_loss_mlp": 1.02516818,
      "epoch": 0.2889436662057356,
      "flos": 22272767856000.0,
      "grad_norm": 2.385954787163811,
      "language_loss": 0.79055029,
      "learning_rate": 3.3370644092200026e-06,
      "loss": 0.8134557,
      "num_input_tokens_seen": 51451435,
      "step": 2403,
      "time_per_iteration": 3.621412754058838
    },
    {
      "auxiliary_loss_clip": 0.01343834,
      "auxiliary_loss_mlp": 0.01031211,
      "balance_loss_clip": 1.04866958,
      "balance_loss_mlp": 1.02266419,
      "epoch": 0.2890639090963747,
      "flos": 21616931381760.0,
      "grad_norm": 2.6476164565864417,
      "language_loss": 0.78391576,
      "learning_rate": 3.3364849996969985e-06,
      "loss": 0.80766618,
      "num_input_tokens_seen": 51471455,
      "step": 2404,
      "time_per_iteration": 2.7525649070739746
    },
    {
      "auxiliary_loss_clip": 0.01250737,
      "auxiliary_loss_mlp": 0.01032179,
      "balance_loss_clip": 1.05620909,
      "balance_loss_mlp": 1.02400088,
      "epoch": 0.28918415198701375,
      "flos": 28585540333440.0,
      "grad_norm": 2.349834334502594,
      "language_loss": 0.8526274,
      "learning_rate": 3.335905387430283e-06,
      "loss": 0.87545657,
      "num_input_tokens_seen": 51492890,
      "step": 2405,
      "time_per_iteration": 3.736158847808838
    },
    {
      "auxiliary_loss_clip": 0.01311459,
      "auxiliary_loss_mlp": 0.01035642,
      "balance_loss_clip": 1.05561781,
      "balance_loss_mlp": 1.0279175,
      "epoch": 0.28930439487765286,
      "flos": 21944688007680.0,
      "grad_norm": 2.5435120340794826,
      "language_loss": 0.82905412,
      "learning_rate": 3.335325572507782e-06,
      "loss": 0.85252517,
      "num_input_tokens_seen": 51513390,
      "step": 2406,
      "time_per_iteration": 2.6724612712860107
    },
    {
      "auxiliary_loss_clip": 0.01208026,
      "auxiliary_loss_mlp": 0.02569586,
      "balance_loss_clip": 1.06212628,
      "balance_loss_mlp": 1.00017536,
      "epoch": 0.28942463776829197,
      "flos": 19281911955840.0,
      "grad_norm": 1.6248628898873405,
      "language_loss": 0.74211591,
      "learning_rate": 3.3347455550174537e-06,
      "loss": 0.77989203,
      "num_input_tokens_seen": 51532730,
      "step": 2407,
      "time_per_iteration": 2.616062641143799
    },
    {
      "auxiliary_loss_clip": 0.01351545,
      "auxiliary_loss_mlp": 0.01029258,
      "balance_loss_clip": 1.05212402,
      "balance_loss_mlp": 1.02122331,
      "epoch": 0.289544880658931,
      "flos": 14645700737280.0,
      "grad_norm": 2.1518753140418774,
      "language_loss": 0.6820091,
      "learning_rate": 3.3341653350472864e-06,
      "loss": 0.70581716,
      "num_input_tokens_seen": 51549560,
      "step": 2408,
      "time_per_iteration": 2.7221713066101074
    },
    {
      "auxiliary_loss_clip": 0.01214993,
      "auxiliary_loss_mlp": 0.01032799,
      "balance_loss_clip": 1.06145084,
      "balance_loss_mlp": 1.0240196,
      "epoch": 0.28966512354957014,
      "flos": 28621881918720.0,
      "grad_norm": 2.4882509980367034,
      "language_loss": 0.69239891,
      "learning_rate": 3.333584912685298e-06,
      "loss": 0.71487689,
      "num_input_tokens_seen": 51568180,
      "step": 2409,
      "time_per_iteration": 3.7170522212982178
    },
    {
      "auxiliary_loss_clip": 0.01173071,
      "auxiliary_loss_mlp": 0.01007986,
      "balance_loss_clip": 1.03722954,
      "balance_loss_mlp": 1.00676382,
      "epoch": 0.28978536644020925,
      "flos": 64711784511360.0,
      "grad_norm": 0.9119556327156105,
      "language_loss": 0.55592555,
      "learning_rate": 3.3330042880195385e-06,
      "loss": 0.57773614,
      "num_input_tokens_seen": 51622530,
      "step": 2410,
      "time_per_iteration": 3.3272042274475098
    },
    {
      "auxiliary_loss_clip": 0.01306234,
      "auxiliary_loss_mlp": 0.01028455,
      "balance_loss_clip": 1.05594039,
      "balance_loss_mlp": 1.02102244,
      "epoch": 0.2899056093308483,
      "flos": 18624638937600.0,
      "grad_norm": 1.9099424653121524,
      "language_loss": 0.78462094,
      "learning_rate": 3.3324234611380888e-06,
      "loss": 0.8079679,
      "num_input_tokens_seen": 51641260,
      "step": 2411,
      "time_per_iteration": 2.68977952003479
    },
    {
      "auxiliary_loss_clip": 0.01351504,
      "auxiliary_loss_mlp": 0.01031015,
      "balance_loss_clip": 1.05662751,
      "balance_loss_mlp": 1.02329063,
      "epoch": 0.2900258522214874,
      "flos": 22893735202560.0,
      "grad_norm": 1.869506455980278,
      "language_loss": 0.81522155,
      "learning_rate": 3.3318424321290596e-06,
      "loss": 0.83904672,
      "num_input_tokens_seen": 51660975,
      "step": 2412,
      "time_per_iteration": 2.876106023788452
    },
    {
      "auxiliary_loss_clip": 0.01267726,
      "auxiliary_loss_mlp": 0.01002062,
      "balance_loss_clip": 1.03589785,
      "balance_loss_mlp": 1.00097084,
      "epoch": 0.2901460951121265,
      "flos": 71106036013440.0,
      "grad_norm": 0.8359375878246211,
      "language_loss": 0.59887516,
      "learning_rate": 3.3312612010805917e-06,
      "loss": 0.62157303,
      "num_input_tokens_seen": 51720550,
      "step": 2413,
      "time_per_iteration": 4.2374587059021
    },
    {
      "auxiliary_loss_clip": 0.01296039,
      "auxiliary_loss_mlp": 0.01036621,
      "balance_loss_clip": 1.05311751,
      "balance_loss_mlp": 1.02772784,
      "epoch": 0.2902663380027656,
      "flos": 32160986081280.0,
      "grad_norm": 1.63221163937246,
      "language_loss": 0.6993258,
      "learning_rate": 3.330679768080858e-06,
      "loss": 0.72265244,
      "num_input_tokens_seen": 51744435,
      "step": 2414,
      "time_per_iteration": 2.755849838256836
    },
    {
      "auxiliary_loss_clip": 0.01254559,
      "auxiliary_loss_mlp": 0.01031678,
      "balance_loss_clip": 1.05999231,
      "balance_loss_mlp": 1.02385187,
      "epoch": 0.2903865808934047,
      "flos": 29351658539520.0,
      "grad_norm": 2.163055314071408,
      "language_loss": 0.83366537,
      "learning_rate": 3.3300981332180627e-06,
      "loss": 0.85652775,
      "num_input_tokens_seen": 51763640,
      "step": 2415,
      "time_per_iteration": 2.7158291339874268
    },
    {
      "auxiliary_loss_clip": 0.01270838,
      "auxiliary_loss_mlp": 0.01030098,
      "balance_loss_clip": 1.05453253,
      "balance_loss_mlp": 1.02206945,
      "epoch": 0.29050682378404374,
      "flos": 17089026647040.0,
      "grad_norm": 5.231211085944058,
      "language_loss": 0.80469882,
      "learning_rate": 3.3295162965804373e-06,
      "loss": 0.82770818,
      "num_input_tokens_seen": 51782135,
      "step": 2416,
      "time_per_iteration": 2.748145818710327
    },
    {
      "auxiliary_loss_clip": 0.01351335,
      "auxiliary_loss_mlp": 0.01029746,
      "balance_loss_clip": 1.05564749,
      "balance_loss_mlp": 1.02238178,
      "epoch": 0.29062706667468285,
      "flos": 17858233422720.0,
      "grad_norm": 2.5712714367722693,
      "language_loss": 0.78742975,
      "learning_rate": 3.328934258256247e-06,
      "loss": 0.81124055,
      "num_input_tokens_seen": 51800200,
      "step": 2417,
      "time_per_iteration": 2.660404682159424
    },
    {
      "auxiliary_loss_clip": 0.01253389,
      "auxiliary_loss_mlp": 0.01029307,
      "balance_loss_clip": 1.05529714,
      "balance_loss_mlp": 1.02090263,
      "epoch": 0.29074730956532197,
      "flos": 24279815174400.0,
      "grad_norm": 2.1448073749761094,
      "language_loss": 0.67352986,
      "learning_rate": 3.3283520183337856e-06,
      "loss": 0.69635689,
      "num_input_tokens_seen": 51819905,
      "step": 2418,
      "time_per_iteration": 2.689288854598999
    },
    {
      "auxiliary_loss_clip": 0.01300763,
      "auxiliary_loss_mlp": 0.01033053,
      "balance_loss_clip": 1.05520678,
      "balance_loss_mlp": 1.02537632,
      "epoch": 0.290867552455961,
      "flos": 22340961826560.0,
      "grad_norm": 3.273820632962703,
      "language_loss": 0.69004464,
      "learning_rate": 3.3277695769013797e-06,
      "loss": 0.71338272,
      "num_input_tokens_seen": 51839350,
      "step": 2419,
      "time_per_iteration": 2.653068780899048
    },
    {
      "auxiliary_loss_clip": 0.01254279,
      "auxiliary_loss_mlp": 0.01032847,
      "balance_loss_clip": 1.05846548,
      "balance_loss_mlp": 1.02435923,
      "epoch": 0.29098779534660013,
      "flos": 23186155824000.0,
      "grad_norm": 2.186299594273936,
      "language_loss": 0.77552795,
      "learning_rate": 3.327186934047385e-06,
      "loss": 0.79839915,
      "num_input_tokens_seen": 51858045,
      "step": 2420,
      "time_per_iteration": 2.6707561016082764
    },
    {
      "auxiliary_loss_clip": 0.01294903,
      "auxiliary_loss_mlp": 0.01027991,
      "balance_loss_clip": 1.04833269,
      "balance_loss_mlp": 1.02010274,
      "epoch": 0.29110803823723924,
      "flos": 15304194817920.0,
      "grad_norm": 2.1981556299533125,
      "language_loss": 0.65724701,
      "learning_rate": 3.3266040898601877e-06,
      "loss": 0.68047595,
      "num_input_tokens_seen": 51875880,
      "step": 2421,
      "time_per_iteration": 2.651920795440674
    },
    {
      "auxiliary_loss_clip": 0.01313774,
      "auxiliary_loss_mlp": 0.01037704,
      "balance_loss_clip": 1.05054164,
      "balance_loss_mlp": 1.02912748,
      "epoch": 0.2912282811278783,
      "flos": 22595352923520.0,
      "grad_norm": 2.20870533271388,
      "language_loss": 0.78191763,
      "learning_rate": 3.3260210444282045e-06,
      "loss": 0.80543244,
      "num_input_tokens_seen": 51893835,
      "step": 2422,
      "time_per_iteration": 2.7745282649993896
    },
    {
      "auxiliary_loss_clip": 0.01254838,
      "auxiliary_loss_mlp": 0.01032051,
      "balance_loss_clip": 1.05926752,
      "balance_loss_mlp": 1.0240401,
      "epoch": 0.2913485240185174,
      "flos": 24497900599680.0,
      "grad_norm": 2.4012708744566376,
      "language_loss": 0.73402047,
      "learning_rate": 3.325437797839883e-06,
      "loss": 0.75688934,
      "num_input_tokens_seen": 51912205,
      "step": 2423,
      "time_per_iteration": 2.629438877105713
    },
    {
      "auxiliary_loss_clip": 0.0120481,
      "auxiliary_loss_mlp": 0.01031156,
      "balance_loss_clip": 1.0579977,
      "balance_loss_mlp": 1.02333546,
      "epoch": 0.2914687669091565,
      "flos": 17931024334080.0,
      "grad_norm": 2.406913145609172,
      "language_loss": 0.75208116,
      "learning_rate": 3.3248543501837015e-06,
      "loss": 0.77444077,
      "num_input_tokens_seen": 51929410,
      "step": 2424,
      "time_per_iteration": 2.616438388824463
    },
    {
      "auxiliary_loss_clip": 0.01321202,
      "auxiliary_loss_mlp": 0.01031305,
      "balance_loss_clip": 1.05708981,
      "balance_loss_mlp": 1.02345562,
      "epoch": 0.2915890097997956,
      "flos": 22529313768960.0,
      "grad_norm": 2.8621430404638675,
      "language_loss": 0.77269328,
      "learning_rate": 3.3242707015481684e-06,
      "loss": 0.7962184,
      "num_input_tokens_seen": 51949345,
      "step": 2425,
      "time_per_iteration": 2.7535741329193115
    },
    {
      "auxiliary_loss_clip": 0.0130298,
      "auxiliary_loss_mlp": 0.01031506,
      "balance_loss_clip": 1.0521909,
      "balance_loss_mlp": 1.02364993,
      "epoch": 0.2917092526904347,
      "flos": 13845216193920.0,
      "grad_norm": 3.976127153118563,
      "language_loss": 0.812105,
      "learning_rate": 3.323686852021823e-06,
      "loss": 0.83544993,
      "num_input_tokens_seen": 51966855,
      "step": 2426,
      "time_per_iteration": 2.805311679840088
    },
    {
      "auxiliary_loss_clip": 0.01355602,
      "auxiliary_loss_mlp": 0.01025635,
      "balance_loss_clip": 1.0496192,
      "balance_loss_mlp": 1.0181551,
      "epoch": 0.2918294955810738,
      "flos": 22674859678080.0,
      "grad_norm": 2.6977594989689053,
      "language_loss": 0.80201054,
      "learning_rate": 3.323102801693235e-06,
      "loss": 0.82582295,
      "num_input_tokens_seen": 51985620,
      "step": 2427,
      "time_per_iteration": 2.7254743576049805
    },
    {
      "auxiliary_loss_clip": 0.01250341,
      "auxiliary_loss_mlp": 0.01029683,
      "balance_loss_clip": 1.05422735,
      "balance_loss_mlp": 1.0210464,
      "epoch": 0.29194973847171285,
      "flos": 23438284364160.0,
      "grad_norm": 11.310680429474695,
      "language_loss": 0.80773312,
      "learning_rate": 3.322518550651003e-06,
      "loss": 0.83053333,
      "num_input_tokens_seen": 52004930,
      "step": 2428,
      "time_per_iteration": 2.684202194213867
    },
    {
      "auxiliary_loss_clip": 0.01309655,
      "auxiliary_loss_mlp": 0.01032411,
      "balance_loss_clip": 1.05501306,
      "balance_loss_mlp": 1.02461433,
      "epoch": 0.29206998136235196,
      "flos": 21909064694400.0,
      "grad_norm": 1.7061678209372848,
      "language_loss": 0.81243908,
      "learning_rate": 3.3219340989837586e-06,
      "loss": 0.83585978,
      "num_input_tokens_seen": 52024920,
      "step": 2429,
      "time_per_iteration": 3.579921245574951
    },
    {
      "auxiliary_loss_clip": 0.01304012,
      "auxiliary_loss_mlp": 0.01031041,
      "balance_loss_clip": 1.05576372,
      "balance_loss_mlp": 1.02371836,
      "epoch": 0.292190224252991,
      "flos": 23215925220480.0,
      "grad_norm": 1.916619280491183,
      "language_loss": 0.80449361,
      "learning_rate": 3.3213494467801625e-06,
      "loss": 0.82784414,
      "num_input_tokens_seen": 52044095,
      "step": 2430,
      "time_per_iteration": 3.5244944095611572
    },
    {
      "auxiliary_loss_clip": 0.01411851,
      "auxiliary_loss_mlp": 0.01028277,
      "balance_loss_clip": 1.04508066,
      "balance_loss_mlp": 1.019557,
      "epoch": 0.2923104671436301,
      "flos": 20740818752640.0,
      "grad_norm": 2.587714043096797,
      "language_loss": 0.71406895,
      "learning_rate": 3.3207645941289063e-06,
      "loss": 0.73847026,
      "num_input_tokens_seen": 52062440,
      "step": 2431,
      "time_per_iteration": 2.9059460163116455
    },
    {
      "auxiliary_loss_clip": 0.01252332,
      "auxiliary_loss_mlp": 0.02568644,
      "balance_loss_clip": 1.05757523,
      "balance_loss_mlp": 1.00003886,
      "epoch": 0.29243071003426924,
      "flos": 35809114999680.0,
      "grad_norm": 2.346601556102325,
      "language_loss": 0.80270302,
      "learning_rate": 3.320179541118711e-06,
      "loss": 0.84091276,
      "num_input_tokens_seen": 52084940,
      "step": 2432,
      "time_per_iteration": 2.862241744995117
    },
    {
      "auxiliary_loss_clip": 0.01075381,
      "auxiliary_loss_mlp": 0.01006319,
      "balance_loss_clip": 1.03682065,
      "balance_loss_mlp": 1.00507879,
      "epoch": 0.2925509529249083,
      "flos": 58081598524800.0,
      "grad_norm": 1.0194224122888937,
      "language_loss": 0.60326058,
      "learning_rate": 3.3195942878383293e-06,
      "loss": 0.6240775,
      "num_input_tokens_seen": 52141040,
      "step": 2433,
      "time_per_iteration": 3.2145872116088867
    },
    {
      "auxiliary_loss_clip": 0.01253266,
      "auxiliary_loss_mlp": 0.01025052,
      "balance_loss_clip": 1.05741715,
      "balance_loss_mlp": 1.01710665,
      "epoch": 0.2926711958155474,
      "flos": 21397122103680.0,
      "grad_norm": 2.1502253986746966,
      "language_loss": 0.77897894,
      "learning_rate": 3.319008834376543e-06,
      "loss": 0.8017621,
      "num_input_tokens_seen": 52160730,
      "step": 2434,
      "time_per_iteration": 3.568903923034668
    },
    {
      "auxiliary_loss_clip": 0.01269546,
      "auxiliary_loss_mlp": 0.01030801,
      "balance_loss_clip": 1.04984081,
      "balance_loss_mlp": 1.02230096,
      "epoch": 0.2927914387061865,
      "flos": 23185796688000.0,
      "grad_norm": 3.8761370728434033,
      "language_loss": 0.88481569,
      "learning_rate": 3.3184231808221654e-06,
      "loss": 0.90781927,
      "num_input_tokens_seen": 52175055,
      "step": 2435,
      "time_per_iteration": 2.6711018085479736
    },
    {
      "auxiliary_loss_clip": 0.01365458,
      "auxiliary_loss_mlp": 0.0103472,
      "balance_loss_clip": 1.06053388,
      "balance_loss_mlp": 1.02603555,
      "epoch": 0.29291168159682557,
      "flos": 22455553190400.0,
      "grad_norm": 2.15432491440137,
      "language_loss": 0.63195324,
      "learning_rate": 3.3178373272640394e-06,
      "loss": 0.65595508,
      "num_input_tokens_seen": 52194150,
      "step": 2436,
      "time_per_iteration": 2.7754809856414795
    },
    {
      "auxiliary_loss_clip": 0.01203674,
      "auxiliary_loss_mlp": 0.01034619,
      "balance_loss_clip": 1.05931497,
      "balance_loss_mlp": 1.02672172,
      "epoch": 0.2930319244874647,
      "flos": 21170632896000.0,
      "grad_norm": 4.017419385215984,
      "language_loss": 0.85272753,
      "learning_rate": 3.3172512737910387e-06,
      "loss": 0.87511051,
      "num_input_tokens_seen": 52211660,
      "step": 2437,
      "time_per_iteration": 2.626894474029541
    },
    {
      "auxiliary_loss_clip": 0.01255811,
      "auxiliary_loss_mlp": 0.01031308,
      "balance_loss_clip": 1.05628252,
      "balance_loss_mlp": 1.02359533,
      "epoch": 0.2931521673781038,
      "flos": 31357843931520.0,
      "grad_norm": 2.80152413885797,
      "language_loss": 0.88492543,
      "learning_rate": 3.3166650204920674e-06,
      "loss": 0.90779662,
      "num_input_tokens_seen": 52232830,
      "step": 2438,
      "time_per_iteration": 2.706667184829712
    },
    {
      "auxiliary_loss_clip": 0.01255804,
      "auxiliary_loss_mlp": 0.01033676,
      "balance_loss_clip": 1.05957317,
      "balance_loss_mlp": 1.02434242,
      "epoch": 0.29327241026874284,
      "flos": 24200990778240.0,
      "grad_norm": 1.825812275592468,
      "language_loss": 0.81544787,
      "learning_rate": 3.316078567456059e-06,
      "loss": 0.83834267,
      "num_input_tokens_seen": 52250670,
      "step": 2439,
      "time_per_iteration": 3.8635995388031006
    },
    {
      "auxiliary_loss_clip": 0.01366805,
      "auxiliary_loss_mlp": 0.01034128,
      "balance_loss_clip": 1.05289316,
      "balance_loss_mlp": 1.02569962,
      "epoch": 0.29339265315938196,
      "flos": 24242611662720.0,
      "grad_norm": 1.8296922965760747,
      "language_loss": 0.75683904,
      "learning_rate": 3.3154919147719786e-06,
      "loss": 0.78084838,
      "num_input_tokens_seen": 52271685,
      "step": 2440,
      "time_per_iteration": 2.792072057723999
    },
    {
      "auxiliary_loss_clip": 0.01254509,
      "auxiliary_loss_mlp": 0.01031767,
      "balance_loss_clip": 1.05673003,
      "balance_loss_mlp": 1.02305245,
      "epoch": 0.29351289605002107,
      "flos": 16946641134720.0,
      "grad_norm": 2.088935258625219,
      "language_loss": 0.86834347,
      "learning_rate": 3.31490506252882e-06,
      "loss": 0.8912062,
      "num_input_tokens_seen": 52291065,
      "step": 2441,
      "time_per_iteration": 2.6605446338653564
    },
    {
      "auxiliary_loss_clip": 0.0134605,
      "auxiliary_loss_mlp": 0.01026281,
      "balance_loss_clip": 1.04967177,
      "balance_loss_mlp": 1.0188272,
      "epoch": 0.2936331389406601,
      "flos": 19829082810240.0,
      "grad_norm": 1.8811715965764475,
      "language_loss": 0.84255892,
      "learning_rate": 3.31431801081561e-06,
      "loss": 0.86628222,
      "num_input_tokens_seen": 52310000,
      "step": 2442,
      "time_per_iteration": 2.7363483905792236
    },
    {
      "auxiliary_loss_clip": 0.01211598,
      "auxiliary_loss_mlp": 0.01000193,
      "balance_loss_clip": 1.03096986,
      "balance_loss_mlp": 0.9989236,
      "epoch": 0.29375338183129923,
      "flos": 71416844398080.0,
      "grad_norm": 0.9154966025159244,
      "language_loss": 0.678949,
      "learning_rate": 3.313730759721402e-06,
      "loss": 0.70106685,
      "num_input_tokens_seen": 52372930,
      "step": 2443,
      "time_per_iteration": 3.3419809341430664
    },
    {
      "auxiliary_loss_clip": 0.0129955,
      "auxiliary_loss_mlp": 0.01034667,
      "balance_loss_clip": 1.05783296,
      "balance_loss_mlp": 1.02667403,
      "epoch": 0.29387362472193834,
      "flos": 22054502862720.0,
      "grad_norm": 2.219466330196458,
      "language_loss": 0.8668437,
      "learning_rate": 3.313143309335282e-06,
      "loss": 0.89018583,
      "num_input_tokens_seen": 52391420,
      "step": 2444,
      "time_per_iteration": 2.7035715579986572
    },
    {
      "auxiliary_loss_clip": 0.01261146,
      "auxiliary_loss_mlp": 0.01029797,
      "balance_loss_clip": 1.05520105,
      "balance_loss_mlp": 1.02183414,
      "epoch": 0.2939938676125774,
      "flos": 22966418373120.0,
      "grad_norm": 3.765547075673411,
      "language_loss": 0.8502965,
      "learning_rate": 3.3125556597463665e-06,
      "loss": 0.8732059,
      "num_input_tokens_seen": 52410725,
      "step": 2445,
      "time_per_iteration": 2.7154808044433594
    },
    {
      "auxiliary_loss_clip": 0.01252574,
      "auxiliary_loss_mlp": 0.01031486,
      "balance_loss_clip": 1.05861521,
      "balance_loss_mlp": 1.02359748,
      "epoch": 0.2941141105032165,
      "flos": 31358705857920.0,
      "grad_norm": 3.8981235628687614,
      "language_loss": 0.65745568,
      "learning_rate": 3.311967811043801e-06,
      "loss": 0.6802963,
      "num_input_tokens_seen": 52432645,
      "step": 2446,
      "time_per_iteration": 2.760721445083618
    },
    {
      "auxiliary_loss_clip": 0.01249699,
      "auxiliary_loss_mlp": 0.0103217,
      "balance_loss_clip": 1.05751657,
      "balance_loss_mlp": 1.02467155,
      "epoch": 0.29423435339385556,
      "flos": 23222138273280.0,
      "grad_norm": 10.187299110923338,
      "language_loss": 0.81920373,
      "learning_rate": 3.3113797633167617e-06,
      "loss": 0.84202242,
      "num_input_tokens_seen": 52450940,
      "step": 2447,
      "time_per_iteration": 2.653059720993042
    },
    {
      "auxiliary_loss_clip": 0.01201703,
      "auxiliary_loss_mlp": 0.01028978,
      "balance_loss_clip": 1.05721068,
      "balance_loss_mlp": 1.02100253,
      "epoch": 0.2943545962844947,
      "flos": 26864054138880.0,
      "grad_norm": 3.486715775575015,
      "language_loss": 0.69353735,
      "learning_rate": 3.310791516654455e-06,
      "loss": 0.71584421,
      "num_input_tokens_seen": 52468000,
      "step": 2448,
      "time_per_iteration": 2.660759687423706
    },
    {
      "auxiliary_loss_clip": 0.01357948,
      "auxiliary_loss_mlp": 0.01030317,
      "balance_loss_clip": 1.05452061,
      "balance_loss_mlp": 1.02196598,
      "epoch": 0.2944748391751338,
      "flos": 20231677422720.0,
      "grad_norm": 2.714349753141441,
      "language_loss": 0.79711479,
      "learning_rate": 3.3102030711461177e-06,
      "loss": 0.82099742,
      "num_input_tokens_seen": 52487575,
      "step": 2449,
      "time_per_iteration": 2.709632158279419
    },
    {
      "auxiliary_loss_clip": 0.01353611,
      "auxiliary_loss_mlp": 0.01026865,
      "balance_loss_clip": 1.05403042,
      "balance_loss_mlp": 1.01890779,
      "epoch": 0.29459508206577284,
      "flos": 15960965045760.0,
      "grad_norm": 2.0562641476112575,
      "language_loss": 0.68445849,
      "learning_rate": 3.3096144268810156e-06,
      "loss": 0.70826328,
      "num_input_tokens_seen": 52506335,
      "step": 2450,
      "time_per_iteration": 2.7327301502227783
    },
    {
      "auxiliary_loss_clip": 0.01248858,
      "auxiliary_loss_mlp": 0.01028231,
      "balance_loss_clip": 1.05485666,
      "balance_loss_mlp": 1.0195049,
      "epoch": 0.29471532495641195,
      "flos": 20412882558720.0,
      "grad_norm": 1.9264981714129963,
      "language_loss": 0.72845453,
      "learning_rate": 3.3090255839484462e-06,
      "loss": 0.75122541,
      "num_input_tokens_seen": 52524330,
      "step": 2451,
      "time_per_iteration": 2.6395602226257324
    },
    {
      "auxiliary_loss_clip": 0.01301071,
      "auxiliary_loss_mlp": 0.01029952,
      "balance_loss_clip": 1.05174387,
      "balance_loss_mlp": 1.02175641,
      "epoch": 0.29483556784705106,
      "flos": 20376576887040.0,
      "grad_norm": 1.8184192361154947,
      "language_loss": 0.85588551,
      "learning_rate": 3.3084365424377366e-06,
      "loss": 0.87919575,
      "num_input_tokens_seen": 52543095,
      "step": 2452,
      "time_per_iteration": 2.705089569091797
    },
    {
      "auxiliary_loss_clip": 0.01340948,
      "auxiliary_loss_mlp": 0.01003451,
      "balance_loss_clip": 1.04611838,
      "balance_loss_mlp": 1.00206232,
      "epoch": 0.2949558107376901,
      "flos": 68555660595840.0,
      "grad_norm": 0.7220568289583451,
      "language_loss": 0.55939472,
      "learning_rate": 3.307847302438245e-06,
      "loss": 0.58283877,
      "num_input_tokens_seen": 52597075,
      "step": 2453,
      "time_per_iteration": 3.182859420776367
    },
    {
      "auxiliary_loss_clip": 0.01305241,
      "auxiliary_loss_mlp": 0.01034611,
      "balance_loss_clip": 1.04686463,
      "balance_loss_mlp": 1.02613497,
      "epoch": 0.2950760536283292,
      "flos": 16107085572480.0,
      "grad_norm": 2.3658732403103726,
      "language_loss": 0.78011262,
      "learning_rate": 3.3072578640393562e-06,
      "loss": 0.80351114,
      "num_input_tokens_seen": 52614410,
      "step": 2454,
      "time_per_iteration": 2.715099573135376
    },
    {
      "auxiliary_loss_clip": 0.01304298,
      "auxiliary_loss_mlp": 0.01036315,
      "balance_loss_clip": 1.05603266,
      "balance_loss_mlp": 1.02828074,
      "epoch": 0.29519629651896834,
      "flos": 20483626394880.0,
      "grad_norm": 1.804049434426277,
      "language_loss": 0.79673111,
      "learning_rate": 3.3066682273304886e-06,
      "loss": 0.8201372,
      "num_input_tokens_seen": 52632055,
      "step": 2455,
      "time_per_iteration": 3.8757946491241455
    },
    {
      "auxiliary_loss_clip": 0.01259958,
      "auxiliary_loss_mlp": 0.02575341,
      "balance_loss_clip": 1.0588727,
      "balance_loss_mlp": 1.00002587,
      "epoch": 0.2953165394096074,
      "flos": 18916484941440.0,
      "grad_norm": 3.386523748038248,
      "language_loss": 0.79164529,
      "learning_rate": 3.3060783924010904e-06,
      "loss": 0.82999825,
      "num_input_tokens_seen": 52649980,
      "step": 2456,
      "time_per_iteration": 3.559276819229126
    },
    {
      "auxiliary_loss_clip": 0.01349447,
      "auxiliary_loss_mlp": 0.01030962,
      "balance_loss_clip": 1.05418038,
      "balance_loss_mlp": 1.02219999,
      "epoch": 0.2954367823002465,
      "flos": 20624467622400.0,
      "grad_norm": 3.13126806580031,
      "language_loss": 0.85169208,
      "learning_rate": 3.3054883593406387e-06,
      "loss": 0.87549615,
      "num_input_tokens_seen": 52664730,
      "step": 2457,
      "time_per_iteration": 2.7350945472717285
    },
    {
      "auxiliary_loss_clip": 0.01303937,
      "auxiliary_loss_mlp": 0.01030029,
      "balance_loss_clip": 1.05377865,
      "balance_loss_mlp": 1.02146363,
      "epoch": 0.2955570251908856,
      "flos": 31175525473920.0,
      "grad_norm": 2.106240031928633,
      "language_loss": 0.65391326,
      "learning_rate": 3.3048981282386404e-06,
      "loss": 0.67725295,
      "num_input_tokens_seen": 52686040,
      "step": 2458,
      "time_per_iteration": 2.776604652404785
    },
    {
      "auxiliary_loss_clip": 0.01340723,
      "auxiliary_loss_mlp": 0.01034718,
      "balance_loss_clip": 1.0522213,
      "balance_loss_mlp": 1.02711272,
      "epoch": 0.29567726808152467,
      "flos": 21650328051840.0,
      "grad_norm": 2.4547814874609255,
      "language_loss": 0.82926625,
      "learning_rate": 3.304307699184634e-06,
      "loss": 0.85302067,
      "num_input_tokens_seen": 52704630,
      "step": 2459,
      "time_per_iteration": 2.680156707763672
    },
    {
      "auxiliary_loss_clip": 0.01215441,
      "auxiliary_loss_mlp": 0.01033895,
      "balance_loss_clip": 1.05735648,
      "balance_loss_mlp": 1.02642632,
      "epoch": 0.2957975109721638,
      "flos": 24243868638720.0,
      "grad_norm": 1.7290338264729899,
      "language_loss": 0.78723174,
      "learning_rate": 3.3037170722681866e-06,
      "loss": 0.80972505,
      "num_input_tokens_seen": 52725465,
      "step": 2460,
      "time_per_iteration": 3.629185676574707
    },
    {
      "auxiliary_loss_clip": 0.01343995,
      "auxiliary_loss_mlp": 0.01027899,
      "balance_loss_clip": 1.05328965,
      "balance_loss_mlp": 1.01958394,
      "epoch": 0.29591775386280283,
      "flos": 13479717352320.0,
      "grad_norm": 1.8536665677194815,
      "language_loss": 0.6813581,
      "learning_rate": 3.3031262475788956e-06,
      "loss": 0.70507711,
      "num_input_tokens_seen": 52742405,
      "step": 2461,
      "time_per_iteration": 2.6990041732788086
    },
    {
      "auxiliary_loss_clip": 0.01297691,
      "auxiliary_loss_mlp": 0.01035096,
      "balance_loss_clip": 1.05298018,
      "balance_loss_mlp": 1.02662587,
      "epoch": 0.29603799675344195,
      "flos": 17749783284480.0,
      "grad_norm": 2.0856619155517677,
      "language_loss": 0.73358285,
      "learning_rate": 3.3025352252063897e-06,
      "loss": 0.75691074,
      "num_input_tokens_seen": 52761100,
      "step": 2462,
      "time_per_iteration": 2.6421267986297607
    },
    {
      "auxiliary_loss_clip": 0.01256727,
      "auxiliary_loss_mlp": 0.01039984,
      "balance_loss_clip": 1.06235647,
      "balance_loss_mlp": 1.0314188,
      "epoch": 0.29615823964408106,
      "flos": 22783920347520.0,
      "grad_norm": 1.7974645321249672,
      "language_loss": 0.75161755,
      "learning_rate": 3.3019440052403252e-06,
      "loss": 0.77458465,
      "num_input_tokens_seen": 52780965,
      "step": 2463,
      "time_per_iteration": 2.6743111610412598
    },
    {
      "auxiliary_loss_clip": 0.01304447,
      "auxiliary_loss_mlp": 0.01034634,
      "balance_loss_clip": 1.05686355,
      "balance_loss_mlp": 1.02711225,
      "epoch": 0.2962784825347201,
      "flos": 23514199758720.0,
      "grad_norm": 2.4199596772604988,
      "language_loss": 0.71068013,
      "learning_rate": 3.30135258777039e-06,
      "loss": 0.73407096,
      "num_input_tokens_seen": 52800335,
      "step": 2464,
      "time_per_iteration": 2.6731889247894287
    },
    {
      "auxiliary_loss_clip": 0.01256014,
      "auxiliary_loss_mlp": 0.02571602,
      "balance_loss_clip": 1.05464983,
      "balance_loss_mlp": 0.99999762,
      "epoch": 0.2963987254253592,
      "flos": 16362769559040.0,
      "grad_norm": 2.083521418647369,
      "language_loss": 0.70998043,
      "learning_rate": 3.3007609728863024e-06,
      "loss": 0.74825656,
      "num_input_tokens_seen": 52818425,
      "step": 2465,
      "time_per_iteration": 3.527679920196533
    },
    {
      "auxiliary_loss_clip": 0.01438717,
      "auxiliary_loss_mlp": 0.01025238,
      "balance_loss_clip": 1.05315316,
      "balance_loss_mlp": 1.01722074,
      "epoch": 0.29651896831599833,
      "flos": 33472263980160.0,
      "grad_norm": 1.8865575624881366,
      "language_loss": 0.73082364,
      "learning_rate": 3.300169160677809e-06,
      "loss": 0.75546312,
      "num_input_tokens_seen": 52842340,
      "step": 2466,
      "time_per_iteration": 2.9511027336120605
    },
    {
      "auxiliary_loss_clip": 0.01362479,
      "auxiliary_loss_mlp": 0.01027533,
      "balance_loss_clip": 1.05762804,
      "balance_loss_mlp": 1.01855063,
      "epoch": 0.2966392112066374,
      "flos": 23805363404160.0,
      "grad_norm": 2.3211277102515133,
      "language_loss": 0.78067714,
      "learning_rate": 3.2995771512346878e-06,
      "loss": 0.80457723,
      "num_input_tokens_seen": 52860690,
      "step": 2467,
      "time_per_iteration": 2.862144708633423
    },
    {
      "auxiliary_loss_clip": 0.01208443,
      "auxiliary_loss_mlp": 0.02571869,
      "balance_loss_clip": 1.06017685,
      "balance_loss_mlp": 1.00010633,
      "epoch": 0.2967594540972765,
      "flos": 19938466702080.0,
      "grad_norm": 2.051564601187901,
      "language_loss": 0.7325623,
      "learning_rate": 3.298984944646746e-06,
      "loss": 0.77036548,
      "num_input_tokens_seen": 52879370,
      "step": 2468,
      "time_per_iteration": 2.654362201690674
    },
    {
      "auxiliary_loss_clip": 0.01261453,
      "auxiliary_loss_mlp": 0.0257048,
      "balance_loss_clip": 1.06211972,
      "balance_loss_mlp": 1.00002682,
      "epoch": 0.2968796969879156,
      "flos": 23732823888000.0,
      "grad_norm": 2.0329103936784425,
      "language_loss": 0.81709933,
      "learning_rate": 3.298392541003822e-06,
      "loss": 0.85541862,
      "num_input_tokens_seen": 52898775,
      "step": 2469,
      "time_per_iteration": 2.65533709526062
    },
    {
      "auxiliary_loss_clip": 0.01303707,
      "auxiliary_loss_mlp": 0.01032427,
      "balance_loss_clip": 1.05725658,
      "balance_loss_mlp": 1.02460694,
      "epoch": 0.29699993987855466,
      "flos": 22893699288960.0,
      "grad_norm": 1.9553620473417779,
      "language_loss": 0.89524221,
      "learning_rate": 3.2977999403957806e-06,
      "loss": 0.91860354,
      "num_input_tokens_seen": 52917535,
      "step": 2470,
      "time_per_iteration": 2.7032053470611572
    },
    {
      "auxiliary_loss_clip": 0.01208218,
      "auxiliary_loss_mlp": 0.01034554,
      "balance_loss_clip": 1.06241369,
      "balance_loss_mlp": 1.02577972,
      "epoch": 0.2971201827691938,
      "flos": 33832555349760.0,
      "grad_norm": 2.0950553785318635,
      "language_loss": 0.67345089,
      "learning_rate": 3.2972071429125207e-06,
      "loss": 0.69587862,
      "num_input_tokens_seen": 52938755,
      "step": 2471,
      "time_per_iteration": 2.6803512573242188
    },
    {
      "auxiliary_loss_clip": 0.01347581,
      "auxiliary_loss_mlp": 0.01028727,
      "balance_loss_clip": 1.05556238,
      "balance_loss_mlp": 1.02034688,
      "epoch": 0.2972404256598329,
      "flos": 22054359208320.0,
      "grad_norm": 4.897434976667871,
      "language_loss": 0.88538748,
      "learning_rate": 3.2966141486439682e-06,
      "loss": 0.9091506,
      "num_input_tokens_seen": 52957945,
      "step": 2472,
      "time_per_iteration": 2.7468693256378174
    },
    {
      "auxiliary_loss_clip": 0.01453438,
      "auxiliary_loss_mlp": 0.01039338,
      "balance_loss_clip": 1.04729176,
      "balance_loss_mlp": 1.02949166,
      "epoch": 0.29736066855047194,
      "flos": 31978595796480.0,
      "grad_norm": 2.2395417854149335,
      "language_loss": 0.64162189,
      "learning_rate": 3.29602095768008e-06,
      "loss": 0.66654968,
      "num_input_tokens_seen": 52978460,
      "step": 2473,
      "time_per_iteration": 2.8340892791748047
    },
    {
      "auxiliary_loss_clip": 0.01298893,
      "auxiliary_loss_mlp": 0.0102693,
      "balance_loss_clip": 1.05829072,
      "balance_loss_mlp": 1.01893675,
      "epoch": 0.29748091144111105,
      "flos": 33510401245440.0,
      "grad_norm": 2.116260541091447,
      "language_loss": 0.64084899,
      "learning_rate": 3.2954275701108437e-06,
      "loss": 0.66410726,
      "num_input_tokens_seen": 52999640,
      "step": 2474,
      "time_per_iteration": 2.7778310775756836
    },
    {
      "auxiliary_loss_clip": 0.01398495,
      "auxiliary_loss_mlp": 0.01031097,
      "balance_loss_clip": 1.05008888,
      "balance_loss_mlp": 1.02278805,
      "epoch": 0.29760115433175016,
      "flos": 41283373409280.0,
      "grad_norm": 5.502083954575838,
      "language_loss": 0.68843186,
      "learning_rate": 3.294833986026275e-06,
      "loss": 0.71272779,
      "num_input_tokens_seen": 53022880,
      "step": 2475,
      "time_per_iteration": 2.8946938514709473
    },
    {
      "auxiliary_loss_clip": 0.01347512,
      "auxiliary_loss_mlp": 0.01037675,
      "balance_loss_clip": 1.05435324,
      "balance_loss_mlp": 1.02934265,
      "epoch": 0.2977213972223892,
      "flos": 24493339572480.0,
      "grad_norm": 1.9460918721614684,
      "language_loss": 0.85320818,
      "learning_rate": 3.29424020551642e-06,
      "loss": 0.87706012,
      "num_input_tokens_seen": 53041515,
      "step": 2476,
      "time_per_iteration": 2.7560973167419434
    },
    {
      "auxiliary_loss_clip": 0.01210844,
      "auxiliary_loss_mlp": 0.0103831,
      "balance_loss_clip": 1.0616672,
      "balance_loss_mlp": 1.02939892,
      "epoch": 0.2978416401130283,
      "flos": 21285116519040.0,
      "grad_norm": 4.477142794746547,
      "language_loss": 0.72506094,
      "learning_rate": 3.2936462286713546e-06,
      "loss": 0.74755245,
      "num_input_tokens_seen": 53059865,
      "step": 2477,
      "time_per_iteration": 2.6023271083831787
    },
    {
      "auxiliary_loss_clip": 0.01256731,
      "auxiliary_loss_mlp": 0.0103181,
      "balance_loss_clip": 1.05957937,
      "balance_loss_mlp": 1.02342987,
      "epoch": 0.2979618830036674,
      "flos": 25772154554880.0,
      "grad_norm": 2.2341980462468207,
      "language_loss": 0.77776021,
      "learning_rate": 3.2930520555811846e-06,
      "loss": 0.80064559,
      "num_input_tokens_seen": 53079490,
      "step": 2478,
      "time_per_iteration": 2.628906488418579
    },
    {
      "auxiliary_loss_clip": 0.01488495,
      "auxiliary_loss_mlp": 0.0258021,
      "balance_loss_clip": 1.04853261,
      "balance_loss_mlp": 0.99995112,
      "epoch": 0.2980821258943065,
      "flos": 23476996247040.0,
      "grad_norm": 2.5242023200673245,
      "language_loss": 0.80195558,
      "learning_rate": 3.292457686336046e-06,
      "loss": 0.84264266,
      "num_input_tokens_seen": 53098810,
      "step": 2479,
      "time_per_iteration": 2.859530210494995
    },
    {
      "auxiliary_loss_clip": 0.01217471,
      "auxiliary_loss_mlp": 0.01000185,
      "balance_loss_clip": 1.03311467,
      "balance_loss_mlp": 0.99883759,
      "epoch": 0.2982023687849456,
      "flos": 69752314195200.0,
      "grad_norm": 0.881713457947621,
      "language_loss": 0.61207497,
      "learning_rate": 3.291863121026105e-06,
      "loss": 0.63425159,
      "num_input_tokens_seen": 53162590,
      "step": 2480,
      "time_per_iteration": 3.286374092102051
    },
    {
      "auxiliary_loss_clip": 0.0125904,
      "auxiliary_loss_mlp": 0.01029346,
      "balance_loss_clip": 1.06095779,
      "balance_loss_mlp": 1.02157688,
      "epoch": 0.29832261167558466,
      "flos": 29825930741760.0,
      "grad_norm": 2.268814445818535,
      "language_loss": 0.76849103,
      "learning_rate": 3.2912683597415547e-06,
      "loss": 0.79137492,
      "num_input_tokens_seen": 53186675,
      "step": 2481,
      "time_per_iteration": 3.9402389526367188
    },
    {
      "auxiliary_loss_clip": 0.01357073,
      "auxiliary_loss_mlp": 0.01032733,
      "balance_loss_clip": 1.05665433,
      "balance_loss_mlp": 1.02431059,
      "epoch": 0.29844285456622377,
      "flos": 33910158683520.0,
      "grad_norm": 2.1179375329462973,
      "language_loss": 0.78100395,
      "learning_rate": 3.2906734025726213e-06,
      "loss": 0.80490202,
      "num_input_tokens_seen": 53205940,
      "step": 2482,
      "time_per_iteration": 3.753671884536743
    },
    {
      "auxiliary_loss_clip": 0.01261961,
      "auxiliary_loss_mlp": 0.0103332,
      "balance_loss_clip": 1.06038713,
      "balance_loss_mlp": 1.02450407,
      "epoch": 0.2985630974568629,
      "flos": 23876933253120.0,
      "grad_norm": 4.4167740202633485,
      "language_loss": 0.87793529,
      "learning_rate": 3.290078249609559e-06,
      "loss": 0.90088809,
      "num_input_tokens_seen": 53225360,
      "step": 2483,
      "time_per_iteration": 2.6557047367095947
    },
    {
      "auxiliary_loss_clip": 0.01254998,
      "auxiliary_loss_mlp": 0.01034621,
      "balance_loss_clip": 1.06177342,
      "balance_loss_mlp": 1.02628851,
      "epoch": 0.29868334034750194,
      "flos": 21799106184960.0,
      "grad_norm": 2.2316522580127383,
      "language_loss": 0.87978947,
      "learning_rate": 3.2894829009426514e-06,
      "loss": 0.90268564,
      "num_input_tokens_seen": 53243195,
      "step": 2484,
      "time_per_iteration": 2.6276180744171143
    },
    {
      "auxiliary_loss_clip": 0.01254783,
      "auxiliary_loss_mlp": 0.010402,
      "balance_loss_clip": 1.06084168,
      "balance_loss_mlp": 1.0318675,
      "epoch": 0.29880358323814105,
      "flos": 25666649331840.0,
      "grad_norm": 2.8031807555751262,
      "language_loss": 0.7790221,
      "learning_rate": 3.288887356662213e-06,
      "loss": 0.80197191,
      "num_input_tokens_seen": 53264530,
      "step": 2485,
      "time_per_iteration": 2.6706464290618896
    },
    {
      "auxiliary_loss_clip": 0.01152771,
      "auxiliary_loss_mlp": 0.01001962,
      "balance_loss_clip": 1.03014588,
      "balance_loss_mlp": 1.00082958,
      "epoch": 0.29892382612878016,
      "flos": 71005846003200.0,
      "grad_norm": 0.7685493840554839,
      "language_loss": 0.5964849,
      "learning_rate": 3.288291616858588e-06,
      "loss": 0.61803222,
      "num_input_tokens_seen": 53319920,
      "step": 2486,
      "time_per_iteration": 4.017135143280029
    },
    {
      "auxiliary_loss_clip": 0.01396067,
      "auxiliary_loss_mlp": 0.01032833,
      "balance_loss_clip": 1.05553865,
      "balance_loss_mlp": 1.02408886,
      "epoch": 0.2990440690194192,
      "flos": 25481134563840.0,
      "grad_norm": 1.8919952901795025,
      "language_loss": 0.77090538,
      "learning_rate": 3.287695681622149e-06,
      "loss": 0.79519439,
      "num_input_tokens_seen": 53339270,
      "step": 2487,
      "time_per_iteration": 2.791639566421509
    },
    {
      "auxiliary_loss_clip": 0.01308975,
      "auxiliary_loss_mlp": 0.01031219,
      "balance_loss_clip": 1.05706346,
      "balance_loss_mlp": 1.02361941,
      "epoch": 0.2991643119100583,
      "flos": 23732357011200.0,
      "grad_norm": 2.4551575997281345,
      "language_loss": 0.81034553,
      "learning_rate": 3.2870995510432982e-06,
      "loss": 0.83374751,
      "num_input_tokens_seen": 53357750,
      "step": 2488,
      "time_per_iteration": 2.690185070037842
    },
    {
      "auxiliary_loss_clip": 0.01247138,
      "auxiliary_loss_mlp": 0.01028304,
      "balance_loss_clip": 1.05607939,
      "balance_loss_mlp": 1.020257,
      "epoch": 0.29928455480069743,
      "flos": 27417545786880.0,
      "grad_norm": 2.904165233449176,
      "language_loss": 0.77247298,
      "learning_rate": 3.2865032252124697e-06,
      "loss": 0.79522741,
      "num_input_tokens_seen": 53378265,
      "step": 2489,
      "time_per_iteration": 2.6864163875579834
    },
    {
      "auxiliary_loss_clip": 0.01303876,
      "auxiliary_loss_mlp": 0.01035246,
      "balance_loss_clip": 1.05391526,
      "balance_loss_mlp": 1.02705657,
      "epoch": 0.2994047976913365,
      "flos": 33692935184640.0,
      "grad_norm": 1.5118183032976868,
      "language_loss": 0.77688301,
      "learning_rate": 3.2859067042201243e-06,
      "loss": 0.80027425,
      "num_input_tokens_seen": 53400305,
      "step": 2490,
      "time_per_iteration": 2.797985553741455
    },
    {
      "auxiliary_loss_clip": 0.01435099,
      "auxiliary_loss_mlp": 0.01030913,
      "balance_loss_clip": 1.04864085,
      "balance_loss_mlp": 1.02300358,
      "epoch": 0.2995250405819756,
      "flos": 16763963541120.0,
      "grad_norm": 2.3209571078770823,
      "language_loss": 0.78406274,
      "learning_rate": 3.2853099881567544e-06,
      "loss": 0.80872285,
      "num_input_tokens_seen": 53418705,
      "step": 2491,
      "time_per_iteration": 3.682551860809326
    },
    {
      "auxiliary_loss_clip": 0.01199319,
      "auxiliary_loss_mlp": 0.01039006,
      "balance_loss_clip": 1.05780339,
      "balance_loss_mlp": 1.03145373,
      "epoch": 0.29964528347261465,
      "flos": 22963976248320.0,
      "grad_norm": 2.2236695296931135,
      "language_loss": 0.79516304,
      "learning_rate": 3.284713077112881e-06,
      "loss": 0.81754631,
      "num_input_tokens_seen": 53438135,
      "step": 2492,
      "time_per_iteration": 2.659133195877075
    },
    {
      "auxiliary_loss_clip": 0.01269293,
      "auxiliary_loss_mlp": 0.0103359,
      "balance_loss_clip": 1.05770266,
      "balance_loss_mlp": 1.02464616,
      "epoch": 0.29976552636325376,
      "flos": 16938021870720.0,
      "grad_norm": 2.8579652341768003,
      "language_loss": 0.86505914,
      "learning_rate": 3.284115971179056e-06,
      "loss": 0.88808799,
      "num_input_tokens_seen": 53452165,
      "step": 2493,
      "time_per_iteration": 2.689410924911499
    },
    {
      "auxiliary_loss_clip": 0.01369761,
      "auxiliary_loss_mlp": 0.01033308,
      "balance_loss_clip": 1.05650926,
      "balance_loss_mlp": 1.02487409,
      "epoch": 0.2998857692538929,
      "flos": 17056455989760.0,
      "grad_norm": 1.8087710971189521,
      "language_loss": 0.78676724,
      "learning_rate": 3.283518670445859e-06,
      "loss": 0.81079799,
      "num_input_tokens_seen": 53470075,
      "step": 2494,
      "time_per_iteration": 2.850482940673828
    },
    {
      "auxiliary_loss_clip": 0.01197069,
      "auxiliary_loss_mlp": 0.02521291,
      "balance_loss_clip": 1.02768946,
      "balance_loss_mlp": 0.99991953,
      "epoch": 0.30000601214453193,
      "flos": 68831528025600.0,
      "grad_norm": 0.6946226507398222,
      "language_loss": 0.54346359,
      "learning_rate": 3.2829211750038995e-06,
      "loss": 0.58064717,
      "num_input_tokens_seen": 53538705,
      "step": 2495,
      "time_per_iteration": 3.340308666229248
    },
    {
      "auxiliary_loss_clip": 0.01347263,
      "auxiliary_loss_mlp": 0.01035103,
      "balance_loss_clip": 1.05280006,
      "balance_loss_mlp": 1.02680624,
      "epoch": 0.30012625503517104,
      "flos": 17603267708160.0,
      "grad_norm": 1.911317884690068,
      "language_loss": 0.89259964,
      "learning_rate": 3.2823234849438183e-06,
      "loss": 0.91642332,
      "num_input_tokens_seen": 53556740,
      "step": 2496,
      "time_per_iteration": 2.6934638023376465
    },
    {
      "auxiliary_loss_clip": 0.01216341,
      "auxiliary_loss_mlp": 0.0103408,
      "balance_loss_clip": 1.05638075,
      "balance_loss_mlp": 1.02612281,
      "epoch": 0.30024649792581015,
      "flos": 21252581775360.0,
      "grad_norm": 3.4069635290799916,
      "language_loss": 0.75624901,
      "learning_rate": 3.2817256003562836e-06,
      "loss": 0.77875328,
      "num_input_tokens_seen": 53577115,
      "step": 2497,
      "time_per_iteration": 2.74157452583313
    },
    {
      "auxiliary_loss_clip": 0.01452031,
      "auxiliary_loss_mlp": 0.01030845,
      "balance_loss_clip": 1.05287492,
      "balance_loss_mlp": 1.0227741,
      "epoch": 0.3003667408164492,
      "flos": 23003262748800.0,
      "grad_norm": 1.9557652572541206,
      "language_loss": 0.66164654,
      "learning_rate": 3.281127521331995e-06,
      "loss": 0.68647534,
      "num_input_tokens_seen": 53598295,
      "step": 2498,
      "time_per_iteration": 2.881704330444336
    },
    {
      "auxiliary_loss_clip": 0.01103138,
      "auxiliary_loss_mlp": 0.00999436,
      "balance_loss_clip": 1.02936113,
      "balance_loss_mlp": 0.99827921,
      "epoch": 0.3004869837070883,
      "flos": 64232340750720.0,
      "grad_norm": 0.8881729053248144,
      "language_loss": 0.60671192,
      "learning_rate": 3.2805292479616798e-06,
      "loss": 0.62773764,
      "num_input_tokens_seen": 53657160,
      "step": 2499,
      "time_per_iteration": 3.351998805999756
    },
    {
      "auxiliary_loss_clip": 0.0130446,
      "auxiliary_loss_mlp": 0.01031314,
      "balance_loss_clip": 1.05677617,
      "balance_loss_mlp": 1.02336848,
      "epoch": 0.30060722659772743,
      "flos": 26248653400320.0,
      "grad_norm": 2.511704882822431,
      "language_loss": 0.91458797,
      "learning_rate": 3.2799307803360955e-06,
      "loss": 0.93794572,
      "num_input_tokens_seen": 53673090,
      "step": 2500,
      "time_per_iteration": 2.7682089805603027
    },
    {
      "auxiliary_loss_clip": 0.01200783,
      "auxiliary_loss_mlp": 0.01027148,
      "balance_loss_clip": 1.05844462,
      "balance_loss_mlp": 1.01936972,
      "epoch": 0.3007274694883665,
      "flos": 24970879912320.0,
      "grad_norm": 1.4755093028622863,
      "language_loss": 0.81418228,
      "learning_rate": 3.27933211854603e-06,
      "loss": 0.83646166,
      "num_input_tokens_seen": 53692145,
      "step": 2501,
      "time_per_iteration": 2.669107675552368
    },
    {
      "auxiliary_loss_clip": 0.01307636,
      "auxiliary_loss_mlp": 0.0102934,
      "balance_loss_clip": 1.0588845,
      "balance_loss_mlp": 1.0210433,
      "epoch": 0.3008477123790056,
      "flos": 17055845458560.0,
      "grad_norm": 1.6281712331530631,
      "language_loss": 0.87167209,
      "learning_rate": 3.278733262682299e-06,
      "loss": 0.89504188,
      "num_input_tokens_seen": 53710000,
      "step": 2502,
      "time_per_iteration": 2.7508397102355957
    },
    {
      "auxiliary_loss_clip": 0.01203281,
      "auxiliary_loss_mlp": 0.01030033,
      "balance_loss_clip": 1.05846834,
      "balance_loss_mlp": 1.02175426,
      "epoch": 0.3009679552696447,
      "flos": 21506398254720.0,
      "grad_norm": 2.2394848529085176,
      "language_loss": 0.82618964,
      "learning_rate": 3.2781342128357484e-06,
      "loss": 0.84852278,
      "num_input_tokens_seen": 53729355,
      "step": 2503,
      "time_per_iteration": 2.667433023452759
    },
    {
      "auxiliary_loss_clip": 0.01352447,
      "auxiliary_loss_mlp": 0.01029606,
      "balance_loss_clip": 1.05354846,
      "balance_loss_mlp": 1.02241778,
      "epoch": 0.30108819816028376,
      "flos": 21134004001920.0,
      "grad_norm": 3.772289450812382,
      "language_loss": 0.8050971,
      "learning_rate": 3.2775349690972547e-06,
      "loss": 0.82891762,
      "num_input_tokens_seen": 53743505,
      "step": 2504,
      "time_per_iteration": 2.788083076477051
    },
    {
      "auxiliary_loss_clip": 0.01146112,
      "auxiliary_loss_mlp": 0.00997916,
      "balance_loss_clip": 1.02470422,
      "balance_loss_mlp": 0.99671239,
      "epoch": 0.30120844105092287,
      "flos": 71126434938240.0,
      "grad_norm": 0.7649986665243198,
      "language_loss": 0.5181092,
      "learning_rate": 3.276935531557722e-06,
      "loss": 0.53954947,
      "num_input_tokens_seen": 53808725,
      "step": 2505,
      "time_per_iteration": 3.3167076110839844
    },
    {
      "auxiliary_loss_clip": 0.01406507,
      "auxiliary_loss_mlp": 0.01028151,
      "balance_loss_clip": 1.05279267,
      "balance_loss_mlp": 1.0191927,
      "epoch": 0.301328683941562,
      "flos": 20264571302400.0,
      "grad_norm": 2.3968177886661217,
      "language_loss": 0.80119503,
      "learning_rate": 3.2763359003080837e-06,
      "loss": 0.82554162,
      "num_input_tokens_seen": 53825680,
      "step": 2506,
      "time_per_iteration": 2.738403081893921
    },
    {
      "auxiliary_loss_clip": 0.01207133,
      "auxiliary_loss_mlp": 0.01001026,
      "balance_loss_clip": 1.02871192,
      "balance_loss_mlp": 0.99989951,
      "epoch": 0.30144892683220104,
      "flos": 70648212240000.0,
      "grad_norm": 0.8511917690897763,
      "language_loss": 0.62428123,
      "learning_rate": 3.2757360754393047e-06,
      "loss": 0.64636278,
      "num_input_tokens_seen": 53889750,
      "step": 2507,
      "time_per_iteration": 4.1711344718933105
    },
    {
      "auxiliary_loss_clip": 0.01250951,
      "auxiliary_loss_mlp": 0.01026075,
      "balance_loss_clip": 1.05664456,
      "balance_loss_mlp": 1.01781368,
      "epoch": 0.30156916972284015,
      "flos": 22820549241600.0,
      "grad_norm": 2.666689012782986,
      "language_loss": 0.64373064,
      "learning_rate": 3.2751360570423767e-06,
      "loss": 0.66650093,
      "num_input_tokens_seen": 53908135,
      "step": 2508,
      "time_per_iteration": 3.6207292079925537
    },
    {
      "auxiliary_loss_clip": 0.01301315,
      "auxiliary_loss_mlp": 0.01029921,
      "balance_loss_clip": 1.05598044,
      "balance_loss_mlp": 1.0214541,
      "epoch": 0.3016894126134792,
      "flos": 29899188529920.0,
      "grad_norm": 2.206503614236403,
      "language_loss": 0.75951266,
      "learning_rate": 3.2745358452083236e-06,
      "loss": 0.78282499,
      "num_input_tokens_seen": 53931035,
      "step": 2509,
      "time_per_iteration": 2.7514190673828125
    },
    {
      "auxiliary_loss_clip": 0.01251166,
      "auxiliary_loss_mlp": 0.01036707,
      "balance_loss_clip": 1.05731988,
      "balance_loss_mlp": 1.02875543,
      "epoch": 0.3018096555041183,
      "flos": 21546331200000.0,
      "grad_norm": 9.707340904701688,
      "language_loss": 0.82509643,
      "learning_rate": 3.2739354400281955e-06,
      "loss": 0.84797513,
      "num_input_tokens_seen": 53952255,
      "step": 2510,
      "time_per_iteration": 2.6871273517608643
    },
    {
      "auxiliary_loss_clip": 0.01260205,
      "auxiliary_loss_mlp": 0.02518282,
      "balance_loss_clip": 1.02779889,
      "balance_loss_mlp": 0.99988461,
      "epoch": 0.3019298983947574,
      "flos": 59136294597120.0,
      "grad_norm": 0.8617400421587595,
      "language_loss": 0.63636732,
      "learning_rate": 3.2733348415930744e-06,
      "loss": 0.6741522,
      "num_input_tokens_seen": 54014125,
      "step": 2511,
      "time_per_iteration": 3.2972218990325928
    },
    {
      "auxiliary_loss_clip": 0.01349903,
      "auxiliary_loss_mlp": 0.01031814,
      "balance_loss_clip": 1.05601168,
      "balance_loss_mlp": 1.0235287,
      "epoch": 0.3020501412853965,
      "flos": 34423070941440.0,
      "grad_norm": 2.0978051565477887,
      "language_loss": 0.80559886,
      "learning_rate": 3.27273404999407e-06,
      "loss": 0.82941604,
      "num_input_tokens_seen": 54036345,
      "step": 2512,
      "time_per_iteration": 3.6827056407928467
    },
    {
      "auxiliary_loss_clip": 0.01212008,
      "auxiliary_loss_mlp": 0.01003841,
      "balance_loss_clip": 1.02590609,
      "balance_loss_mlp": 1.0027802,
      "epoch": 0.3021703841760356,
      "flos": 71008288128000.0,
      "grad_norm": 0.8059738882529697,
      "language_loss": 0.60482377,
      "learning_rate": 3.272133065322322e-06,
      "loss": 0.62698221,
      "num_input_tokens_seen": 54094615,
      "step": 2513,
      "time_per_iteration": 3.2287724018096924
    },
    {
      "auxiliary_loss_clip": 0.01196908,
      "auxiliary_loss_mlp": 0.01025792,
      "balance_loss_clip": 1.05489612,
      "balance_loss_mlp": 1.01777554,
      "epoch": 0.3022906270666747,
      "flos": 21510528318720.0,
      "grad_norm": 1.9067913819456626,
      "language_loss": 0.79661226,
      "learning_rate": 3.271531887669e-06,
      "loss": 0.81883931,
      "num_input_tokens_seen": 54114675,
      "step": 2514,
      "time_per_iteration": 2.6113736629486084
    },
    {
      "auxiliary_loss_clip": 0.01402489,
      "auxiliary_loss_mlp": 0.01028027,
      "balance_loss_clip": 1.05030942,
      "balance_loss_mlp": 1.01959896,
      "epoch": 0.30241086995731375,
      "flos": 31132001168640.0,
      "grad_norm": 2.4120813343208782,
      "language_loss": 0.63635933,
      "learning_rate": 3.2709305171253015e-06,
      "loss": 0.66066444,
      "num_input_tokens_seen": 54134795,
      "step": 2515,
      "time_per_iteration": 2.816473960876465
    },
    {
      "auxiliary_loss_clip": 0.01253352,
      "auxiliary_loss_mlp": 0.01030459,
      "balance_loss_clip": 1.0596801,
      "balance_loss_mlp": 1.02235842,
      "epoch": 0.30253111284795287,
      "flos": 23511542152320.0,
      "grad_norm": 1.9291426341541447,
      "language_loss": 0.78230619,
      "learning_rate": 3.2703289537824536e-06,
      "loss": 0.80514431,
      "num_input_tokens_seen": 54154595,
      "step": 2516,
      "time_per_iteration": 2.659670829772949
    },
    {
      "auxiliary_loss_clip": 0.01405767,
      "auxiliary_loss_mlp": 0.01028469,
      "balance_loss_clip": 1.05189574,
      "balance_loss_mlp": 1.01975524,
      "epoch": 0.302651355738592,
      "flos": 18725367651840.0,
      "grad_norm": 2.581097721729028,
      "language_loss": 0.78297418,
      "learning_rate": 3.269727197731714e-06,
      "loss": 0.80731654,
      "num_input_tokens_seen": 54167360,
      "step": 2517,
      "time_per_iteration": 3.618957757949829
    },
    {
      "auxiliary_loss_clip": 0.01388501,
      "auxiliary_loss_mlp": 0.01025371,
      "balance_loss_clip": 1.05275977,
      "balance_loss_mlp": 1.01732492,
      "epoch": 0.30277159862923103,
      "flos": 22418888382720.0,
      "grad_norm": 1.7379422836747218,
      "language_loss": 0.78293955,
      "learning_rate": 3.269125249064367e-06,
      "loss": 0.80707824,
      "num_input_tokens_seen": 54187055,
      "step": 2518,
      "time_per_iteration": 2.740097999572754
    },
    {
      "auxiliary_loss_clip": 0.01204889,
      "auxiliary_loss_mlp": 0.01032093,
      "balance_loss_clip": 1.05744731,
      "balance_loss_mlp": 1.02411234,
      "epoch": 0.30289184151987014,
      "flos": 22273126992000.0,
      "grad_norm": 2.5117126580021503,
      "language_loss": 0.83436584,
      "learning_rate": 3.2685231078717297e-06,
      "loss": 0.85673571,
      "num_input_tokens_seen": 54207245,
      "step": 2519,
      "time_per_iteration": 2.6215996742248535
    },
    {
      "auxiliary_loss_clip": 0.01345159,
      "auxiliary_loss_mlp": 0.02570921,
      "balance_loss_clip": 1.05313766,
      "balance_loss_mlp": 0.9999668,
      "epoch": 0.30301208441050925,
      "flos": 25225594231680.0,
      "grad_norm": 2.065102545242766,
      "language_loss": 0.75668621,
      "learning_rate": 3.267920774245145e-06,
      "loss": 0.795847,
      "num_input_tokens_seen": 54226650,
      "step": 2520,
      "time_per_iteration": 2.760965585708618
    },
    {
      "auxiliary_loss_clip": 0.0125347,
      "auxiliary_loss_mlp": 0.01035081,
      "balance_loss_clip": 1.06022155,
      "balance_loss_mlp": 1.02682567,
      "epoch": 0.3031323273011483,
      "flos": 23039245198080.0,
      "grad_norm": 2.0336153876492133,
      "language_loss": 0.8483541,
      "learning_rate": 3.2673182482759876e-06,
      "loss": 0.87123954,
      "num_input_tokens_seen": 54245765,
      "step": 2521,
      "time_per_iteration": 2.7909414768218994
    },
    {
      "auxiliary_loss_clip": 0.01250476,
      "auxiliary_loss_mlp": 0.01029358,
      "balance_loss_clip": 1.05764914,
      "balance_loss_mlp": 1.0215435,
      "epoch": 0.3032525701917874,
      "flos": 18876695650560.0,
      "grad_norm": 2.1333381636288866,
      "language_loss": 0.65975022,
      "learning_rate": 3.266715530055659e-06,
      "loss": 0.68254858,
      "num_input_tokens_seen": 54263915,
      "step": 2522,
      "time_per_iteration": 2.645923137664795
    },
    {
      "auxiliary_loss_clip": 0.01246726,
      "auxiliary_loss_mlp": 0.01029353,
      "balance_loss_clip": 1.05373263,
      "balance_loss_mlp": 1.020895,
      "epoch": 0.30337281308242653,
      "flos": 17782641250560.0,
      "grad_norm": 2.1150068665353086,
      "language_loss": 0.80418289,
      "learning_rate": 3.2661126196755927e-06,
      "loss": 0.82694364,
      "num_input_tokens_seen": 54283025,
      "step": 2523,
      "time_per_iteration": 2.6276869773864746
    },
    {
      "auxiliary_loss_clip": 0.01100797,
      "auxiliary_loss_mlp": 0.0100363,
      "balance_loss_clip": 1.02777541,
      "balance_loss_mlp": 1.00255764,
      "epoch": 0.3034930559730656,
      "flos": 57824298426240.0,
      "grad_norm": 0.8447652536836038,
      "language_loss": 0.56015849,
      "learning_rate": 3.265509517227248e-06,
      "loss": 0.58120275,
      "num_input_tokens_seen": 54339840,
      "step": 2524,
      "time_per_iteration": 3.137282609939575
    },
    {
      "auxiliary_loss_clip": 0.01296308,
      "auxiliary_loss_mlp": 0.01027188,
      "balance_loss_clip": 1.0512681,
      "balance_loss_mlp": 1.01977885,
      "epoch": 0.3036132988637047,
      "flos": 14755587419520.0,
      "grad_norm": 3.0133564204999392,
      "language_loss": 0.80889857,
      "learning_rate": 3.264906222802115e-06,
      "loss": 0.83213353,
      "num_input_tokens_seen": 54357690,
      "step": 2525,
      "time_per_iteration": 2.704467535018921
    },
    {
      "auxiliary_loss_clip": 0.01202548,
      "auxiliary_loss_mlp": 0.01029548,
      "balance_loss_clip": 1.05699027,
      "balance_loss_mlp": 1.02195466,
      "epoch": 0.30373354175434375,
      "flos": 21033203460480.0,
      "grad_norm": 3.8277236661528726,
      "language_loss": 0.78340721,
      "learning_rate": 3.264302736491715e-06,
      "loss": 0.80572814,
      "num_input_tokens_seen": 54377810,
      "step": 2526,
      "time_per_iteration": 2.6257448196411133
    },
    {
      "auxiliary_loss_clip": 0.01256009,
      "auxiliary_loss_mlp": 0.01034762,
      "balance_loss_clip": 1.06261122,
      "balance_loss_mlp": 1.02698958,
      "epoch": 0.30385378464498286,
      "flos": 21143233797120.0,
      "grad_norm": 3.672758508548966,
      "language_loss": 0.87621808,
      "learning_rate": 3.263699058387594e-06,
      "loss": 0.89912575,
      "num_input_tokens_seen": 54395245,
      "step": 2527,
      "time_per_iteration": 2.624135732650757
    },
    {
      "auxiliary_loss_clip": 0.01348188,
      "auxiliary_loss_mlp": 0.01026366,
      "balance_loss_clip": 1.0503186,
      "balance_loss_mlp": 1.01864171,
      "epoch": 0.30397402753562197,
      "flos": 20629244131200.0,
      "grad_norm": 2.602370140889034,
      "language_loss": 0.90630329,
      "learning_rate": 3.2630951885813315e-06,
      "loss": 0.93004882,
      "num_input_tokens_seen": 54412640,
      "step": 2528,
      "time_per_iteration": 2.7357165813446045
    },
    {
      "auxiliary_loss_clip": 0.01301159,
      "auxiliary_loss_mlp": 0.01023443,
      "balance_loss_clip": 1.05262542,
      "balance_loss_mlp": 1.01564097,
      "epoch": 0.304094270426261,
      "flos": 15085678429440.0,
      "grad_norm": 1.9663557165831282,
      "language_loss": 0.78358603,
      "learning_rate": 3.262491127164533e-06,
      "loss": 0.80683208,
      "num_input_tokens_seen": 54431455,
      "step": 2529,
      "time_per_iteration": 2.65051007270813
    },
    {
      "auxiliary_loss_clip": 0.01216331,
      "auxiliary_loss_mlp": 0.02570241,
      "balance_loss_clip": 1.05569923,
      "balance_loss_mlp": 1.0000906,
      "epoch": 0.30421451331690014,
      "flos": 13845216193920.0,
      "grad_norm": 2.2238041062136458,
      "language_loss": 0.80256325,
      "learning_rate": 3.2618868742288337e-06,
      "loss": 0.84042907,
      "num_input_tokens_seen": 54448380,
      "step": 2530,
      "time_per_iteration": 2.6889004707336426
    },
    {
      "auxiliary_loss_clip": 0.01250853,
      "auxiliary_loss_mlp": 0.01027193,
      "balance_loss_clip": 1.05871177,
      "balance_loss_mlp": 1.01942682,
      "epoch": 0.30433475620753925,
      "flos": 17384212615680.0,
      "grad_norm": 2.0666093877969867,
      "language_loss": 0.7263577,
      "learning_rate": 3.261282429865899e-06,
      "loss": 0.74913824,
      "num_input_tokens_seen": 54466385,
      "step": 2531,
      "time_per_iteration": 2.6811068058013916
    },
    {
      "auxiliary_loss_clip": 0.01308394,
      "auxiliary_loss_mlp": 0.02563249,
      "balance_loss_clip": 1.05719841,
      "balance_loss_mlp": 1.00000238,
      "epoch": 0.3044549990981783,
      "flos": 18916951818240.0,
      "grad_norm": 1.7200598045907363,
      "language_loss": 0.72689867,
      "learning_rate": 3.2606777941674225e-06,
      "loss": 0.76561511,
      "num_input_tokens_seen": 54485040,
      "step": 2532,
      "time_per_iteration": 2.67688250541687
    },
    {
      "auxiliary_loss_clip": 0.01385305,
      "auxiliary_loss_mlp": 0.01023247,
      "balance_loss_clip": 1.0494833,
      "balance_loss_mlp": 1.01508701,
      "epoch": 0.3045752419888174,
      "flos": 21068431724160.0,
      "grad_norm": 2.484047027518094,
      "language_loss": 0.84481812,
      "learning_rate": 3.2600729672251276e-06,
      "loss": 0.8689037,
      "num_input_tokens_seen": 54502755,
      "step": 2533,
      "time_per_iteration": 3.6778342723846436
    },
    {
      "auxiliary_loss_clip": 0.01202035,
      "auxiliary_loss_mlp": 0.02566142,
      "balance_loss_clip": 1.05827415,
      "balance_loss_mlp": 0.9999752,
      "epoch": 0.3046954848794565,
      "flos": 29096405516160.0,
      "grad_norm": 2.152050552038604,
      "language_loss": 0.65264976,
      "learning_rate": 3.259467949130765e-06,
      "loss": 0.69033158,
      "num_input_tokens_seen": 54524165,
      "step": 2534,
      "time_per_iteration": 3.5239288806915283
    },
    {
      "auxiliary_loss_clip": 0.01302494,
      "auxiliary_loss_mlp": 0.01030493,
      "balance_loss_clip": 1.05788243,
      "balance_loss_mlp": 1.02205324,
      "epoch": 0.3048157277700956,
      "flos": 20295346279680.0,
      "grad_norm": 16.648093819325613,
      "language_loss": 0.82615572,
      "learning_rate": 3.2588627399761164e-06,
      "loss": 0.84948552,
      "num_input_tokens_seen": 54540160,
      "step": 2535,
      "time_per_iteration": 2.676400899887085
    },
    {
      "auxiliary_loss_clip": 0.01299615,
      "auxiliary_loss_mlp": 0.01029067,
      "balance_loss_clip": 1.05609655,
      "balance_loss_mlp": 1.02147353,
      "epoch": 0.3049359706607347,
      "flos": 22739929165440.0,
      "grad_norm": 1.8250331740192618,
      "language_loss": 0.70759428,
      "learning_rate": 3.2582573398529903e-06,
      "loss": 0.73088109,
      "num_input_tokens_seen": 54557515,
      "step": 2536,
      "time_per_iteration": 2.6638200283050537
    },
    {
      "auxiliary_loss_clip": 0.01358672,
      "auxiliary_loss_mlp": 0.01026531,
      "balance_loss_clip": 1.05446672,
      "balance_loss_mlp": 1.01828456,
      "epoch": 0.3050562135513738,
      "flos": 18434634969600.0,
      "grad_norm": 2.041445112567677,
      "language_loss": 0.73794508,
      "learning_rate": 3.2576517488532265e-06,
      "loss": 0.76179707,
      "num_input_tokens_seen": 54573865,
      "step": 2537,
      "time_per_iteration": 2.767411708831787
    },
    {
      "auxiliary_loss_clip": 0.01250117,
      "auxiliary_loss_mlp": 0.01024435,
      "balance_loss_clip": 1.0551393,
      "balance_loss_mlp": 1.01688349,
      "epoch": 0.30517645644201286,
      "flos": 20370327920640.0,
      "grad_norm": 1.752339072959316,
      "language_loss": 0.8767013,
      "learning_rate": 3.257045967068692e-06,
      "loss": 0.89944679,
      "num_input_tokens_seen": 54593120,
      "step": 2538,
      "time_per_iteration": 2.638176918029785
    },
    {
      "auxiliary_loss_clip": 0.01203395,
      "auxiliary_loss_mlp": 0.01032292,
      "balance_loss_clip": 1.05896235,
      "balance_loss_mlp": 1.02376306,
      "epoch": 0.30529669933265197,
      "flos": 21945118970880.0,
      "grad_norm": 1.7513904494084631,
      "language_loss": 0.81915009,
      "learning_rate": 3.2564399945912848e-06,
      "loss": 0.84150696,
      "num_input_tokens_seen": 54612910,
      "step": 2539,
      "time_per_iteration": 3.564380645751953
    },
    {
      "auxiliary_loss_clip": 0.01316573,
      "auxiliary_loss_mlp": 0.01028897,
      "balance_loss_clip": 1.05197012,
      "balance_loss_mlp": 1.02145243,
      "epoch": 0.305416942223291,
      "flos": 21835411856640.0,
      "grad_norm": 2.4562166509168666,
      "language_loss": 0.81860638,
      "learning_rate": 3.2558338315129287e-06,
      "loss": 0.84206104,
      "num_input_tokens_seen": 54631055,
      "step": 2540,
      "time_per_iteration": 2.7506237030029297
    },
    {
      "auxiliary_loss_clip": 0.01250637,
      "auxiliary_loss_mlp": 0.01034086,
      "balance_loss_clip": 1.05724478,
      "balance_loss_mlp": 1.02629602,
      "epoch": 0.30553718511393013,
      "flos": 33911810709120.0,
      "grad_norm": 4.197721539307683,
      "language_loss": 0.76268744,
      "learning_rate": 3.2552274779255785e-06,
      "loss": 0.78553474,
      "num_input_tokens_seen": 54651985,
      "step": 2541,
      "time_per_iteration": 2.712716817855835
    },
    {
      "auxiliary_loss_clip": 0.01248028,
      "auxiliary_loss_mlp": 0.01026255,
      "balance_loss_clip": 1.05657816,
      "balance_loss_mlp": 1.01810062,
      "epoch": 0.30565742800456924,
      "flos": 22268530051200.0,
      "grad_norm": 3.2018299524224565,
      "language_loss": 0.77009475,
      "learning_rate": 3.2546209339212184e-06,
      "loss": 0.79283762,
      "num_input_tokens_seen": 54671005,
      "step": 2542,
      "time_per_iteration": 2.600860834121704
    },
    {
      "auxiliary_loss_clip": 0.01301301,
      "auxiliary_loss_mlp": 0.01030602,
      "balance_loss_clip": 1.05394197,
      "balance_loss_mlp": 1.02267504,
      "epoch": 0.3057776708952083,
      "flos": 22565044823040.0,
      "grad_norm": 1.4375091055414382,
      "language_loss": 0.77807581,
      "learning_rate": 3.25401419959186e-06,
      "loss": 0.80139488,
      "num_input_tokens_seen": 54691615,
      "step": 2543,
      "time_per_iteration": 3.639772653579712
    },
    {
      "auxiliary_loss_clip": 0.01314342,
      "auxiliary_loss_mlp": 0.01029523,
      "balance_loss_clip": 1.06044197,
      "balance_loss_mlp": 1.02172685,
      "epoch": 0.3058979137858474,
      "flos": 21799213925760.0,
      "grad_norm": 1.789122574919612,
      "language_loss": 0.76102608,
      "learning_rate": 3.253407275029545e-06,
      "loss": 0.78446478,
      "num_input_tokens_seen": 54710520,
      "step": 2544,
      "time_per_iteration": 2.664422035217285
    },
    {
      "auxiliary_loss_clip": 0.01351116,
      "auxiliary_loss_mlp": 0.01029434,
      "balance_loss_clip": 1.05606711,
      "balance_loss_mlp": 1.02111363,
      "epoch": 0.3060181566764865,
      "flos": 26979435601920.0,
      "grad_norm": 1.8486770256519318,
      "language_loss": 0.80281311,
      "learning_rate": 3.2528001603263425e-06,
      "loss": 0.82661855,
      "num_input_tokens_seen": 54732590,
      "step": 2545,
      "time_per_iteration": 2.7870750427246094
    },
    {
      "auxiliary_loss_clip": 0.01259666,
      "auxiliary_loss_mlp": 0.01034686,
      "balance_loss_clip": 1.06380367,
      "balance_loss_mlp": 1.02705646,
      "epoch": 0.3061383995671256,
      "flos": 19865101173120.0,
      "grad_norm": 2.1314360701016057,
      "language_loss": 0.81658125,
      "learning_rate": 3.2521928555743514e-06,
      "loss": 0.83952475,
      "num_input_tokens_seen": 54749935,
      "step": 2546,
      "time_per_iteration": 2.590965747833252
    },
    {
      "auxiliary_loss_clip": 0.01295212,
      "auxiliary_loss_mlp": 0.02567539,
      "balance_loss_clip": 1.05312061,
      "balance_loss_mlp": 1.000072,
      "epoch": 0.3062586424577647,
      "flos": 22127509255680.0,
      "grad_norm": 2.0753982220814646,
      "language_loss": 0.67206252,
      "learning_rate": 3.2515853608657e-06,
      "loss": 0.71069008,
      "num_input_tokens_seen": 54767935,
      "step": 2547,
      "time_per_iteration": 2.8088858127593994
    },
    {
      "auxiliary_loss_clip": 0.01252896,
      "auxiliary_loss_mlp": 0.01032458,
      "balance_loss_clip": 1.05722165,
      "balance_loss_mlp": 1.02410746,
      "epoch": 0.3063788853484038,
      "flos": 20845497962880.0,
      "grad_norm": 3.703993399830302,
      "language_loss": 0.75077653,
      "learning_rate": 3.250977676292545e-06,
      "loss": 0.77363002,
      "num_input_tokens_seen": 54786175,
      "step": 2548,
      "time_per_iteration": 2.616637706756592
    },
    {
      "auxiliary_loss_clip": 0.01301348,
      "auxiliary_loss_mlp": 0.01031788,
      "balance_loss_clip": 1.05409527,
      "balance_loss_mlp": 1.02348495,
      "epoch": 0.30649912823904285,
      "flos": 16209717707520.0,
      "grad_norm": 2.675162577092554,
      "language_loss": 0.79185051,
      "learning_rate": 3.2503698019470712e-06,
      "loss": 0.81518185,
      "num_input_tokens_seen": 54801945,
      "step": 2549,
      "time_per_iteration": 2.638261318206787
    },
    {
      "auxiliary_loss_clip": 0.01251484,
      "auxiliary_loss_mlp": 0.01027868,
      "balance_loss_clip": 1.05490565,
      "balance_loss_mlp": 1.01926076,
      "epoch": 0.30661937112968196,
      "flos": 18617815353600.0,
      "grad_norm": 2.34030730150707,
      "language_loss": 0.77909076,
      "learning_rate": 3.249761737921492e-06,
      "loss": 0.80188429,
      "num_input_tokens_seen": 54818475,
      "step": 2550,
      "time_per_iteration": 2.5986745357513428
    },
    {
      "auxiliary_loss_clip": 0.01294499,
      "auxiliary_loss_mlp": 0.01031192,
      "balance_loss_clip": 1.05551744,
      "balance_loss_mlp": 1.02324939,
      "epoch": 0.30673961402032107,
      "flos": 31390809638400.0,
      "grad_norm": 2.0790126382708047,
      "language_loss": 0.74316883,
      "learning_rate": 3.249153484308051e-06,
      "loss": 0.76642579,
      "num_input_tokens_seen": 54837090,
      "step": 2551,
      "time_per_iteration": 2.6573047637939453
    },
    {
      "auxiliary_loss_clip": 0.01390645,
      "auxiliary_loss_mlp": 0.01034372,
      "balance_loss_clip": 1.05044532,
      "balance_loss_mlp": 1.02584219,
      "epoch": 0.3068598569109601,
      "flos": 20229809915520.0,
      "grad_norm": 9.392472395951206,
      "language_loss": 0.78135347,
      "learning_rate": 3.2485450411990194e-06,
      "loss": 0.80560368,
      "num_input_tokens_seen": 54856445,
      "step": 2552,
      "time_per_iteration": 2.6613078117370605
    },
    {
      "auxiliary_loss_clip": 0.01202853,
      "auxiliary_loss_mlp": 0.01032689,
      "balance_loss_clip": 1.05706644,
      "balance_loss_mlp": 1.02401066,
      "epoch": 0.30698009980159924,
      "flos": 29601991399680.0,
      "grad_norm": 1.8091694738391721,
      "language_loss": 0.82667589,
      "learning_rate": 3.2479364086866983e-06,
      "loss": 0.84903133,
      "num_input_tokens_seen": 54876700,
      "step": 2553,
      "time_per_iteration": 2.672306776046753
    },
    {
      "auxiliary_loss_clip": 0.01308885,
      "auxiliary_loss_mlp": 0.02571101,
      "balance_loss_clip": 1.06052113,
      "balance_loss_mlp": 1.00004208,
      "epoch": 0.30710034269223835,
      "flos": 23842423261440.0,
      "grad_norm": 1.743075755029424,
      "language_loss": 0.81106478,
      "learning_rate": 3.247327586863416e-06,
      "loss": 0.84986466,
      "num_input_tokens_seen": 54897580,
      "step": 2554,
      "time_per_iteration": 2.717473268508911
    },
    {
      "auxiliary_loss_clip": 0.01355851,
      "auxiliary_loss_mlp": 0.01024643,
      "balance_loss_clip": 1.05311322,
      "balance_loss_mlp": 1.01635754,
      "epoch": 0.3072205855828774,
      "flos": 25884986152320.0,
      "grad_norm": 2.633923560465508,
      "language_loss": 0.7726649,
      "learning_rate": 3.2467185758215304e-06,
      "loss": 0.79646981,
      "num_input_tokens_seen": 54917320,
      "step": 2555,
      "time_per_iteration": 2.832066297531128
    },
    {
      "auxiliary_loss_clip": 0.01270114,
      "auxiliary_loss_mlp": 0.02568833,
      "balance_loss_clip": 1.05841279,
      "balance_loss_mlp": 1.00013447,
      "epoch": 0.3073408284735165,
      "flos": 22236390357120.0,
      "grad_norm": 9.682977915624967,
      "language_loss": 0.85518312,
      "learning_rate": 3.246109375653428e-06,
      "loss": 0.89357257,
      "num_input_tokens_seen": 54934085,
      "step": 2556,
      "time_per_iteration": 2.7564468383789062
    },
    {
      "auxiliary_loss_clip": 0.01201086,
      "auxiliary_loss_mlp": 0.01028759,
      "balance_loss_clip": 1.05741608,
      "balance_loss_mlp": 1.02064717,
      "epoch": 0.30746107136415557,
      "flos": 19500284689920.0,
      "grad_norm": 1.7988040101209102,
      "language_loss": 0.78632057,
      "learning_rate": 3.2454999864515243e-06,
      "loss": 0.80861896,
      "num_input_tokens_seen": 54953460,
      "step": 2557,
      "time_per_iteration": 2.569892406463623
    },
    {
      "auxiliary_loss_clip": 0.01295941,
      "auxiliary_loss_mlp": 0.02569396,
      "balance_loss_clip": 1.05587137,
      "balance_loss_mlp": 1.00008631,
      "epoch": 0.3075813142547947,
      "flos": 21724806902400.0,
      "grad_norm": 2.4858783993755433,
      "language_loss": 0.69388038,
      "learning_rate": 3.244890408308263e-06,
      "loss": 0.73253375,
      "num_input_tokens_seen": 54974165,
      "step": 2558,
      "time_per_iteration": 2.7188351154327393
    },
    {
      "auxiliary_loss_clip": 0.01396855,
      "auxiliary_loss_mlp": 0.01031517,
      "balance_loss_clip": 1.05011022,
      "balance_loss_mlp": 1.02358925,
      "epoch": 0.3077015571454338,
      "flos": 24097963593600.0,
      "grad_norm": 2.917766641519629,
      "language_loss": 0.61270016,
      "learning_rate": 3.2442806413161165e-06,
      "loss": 0.63698387,
      "num_input_tokens_seen": 54993810,
      "step": 2559,
      "time_per_iteration": 3.6266186237335205
    },
    {
      "auxiliary_loss_clip": 0.01315991,
      "auxiliary_loss_mlp": 0.01031084,
      "balance_loss_clip": 1.05416286,
      "balance_loss_mlp": 1.02263236,
      "epoch": 0.30782180003607285,
      "flos": 18405476104320.0,
      "grad_norm": 2.1564983286728068,
      "language_loss": 0.75827092,
      "learning_rate": 3.243670685567586e-06,
      "loss": 0.78174162,
      "num_input_tokens_seen": 55011210,
      "step": 2560,
      "time_per_iteration": 3.6124324798583984
    },
    {
      "auxiliary_loss_clip": 0.0129947,
      "auxiliary_loss_mlp": 0.02568029,
      "balance_loss_clip": 1.05536747,
      "balance_loss_mlp": 1.00007463,
      "epoch": 0.30794204292671196,
      "flos": 23878549365120.0,
      "grad_norm": 2.4450806792421003,
      "language_loss": 0.80385941,
      "learning_rate": 3.2430605411552012e-06,
      "loss": 0.84253442,
      "num_input_tokens_seen": 55031325,
      "step": 2561,
      "time_per_iteration": 2.7021894454956055
    },
    {
      "auxiliary_loss_clip": 0.01270938,
      "auxiliary_loss_mlp": 0.01001302,
      "balance_loss_clip": 1.03291464,
      "balance_loss_mlp": 1.00005674,
      "epoch": 0.30806228581735107,
      "flos": 67927800816000.0,
      "grad_norm": 0.8965452254521274,
      "language_loss": 0.705531,
      "learning_rate": 3.2424502081715205e-06,
      "loss": 0.72825348,
      "num_input_tokens_seen": 55094440,
      "step": 2562,
      "time_per_iteration": 3.329286813735962
    },
    {
      "auxiliary_loss_clip": 0.01214356,
      "auxiliary_loss_mlp": 0.01033463,
      "balance_loss_clip": 1.05652273,
      "balance_loss_mlp": 1.02510071,
      "epoch": 0.3081825287079901,
      "flos": 23843213360640.0,
      "grad_norm": 1.983980680246325,
      "language_loss": 0.78371704,
      "learning_rate": 3.241839686709132e-06,
      "loss": 0.80619526,
      "num_input_tokens_seen": 55115375,
      "step": 2563,
      "time_per_iteration": 2.752105712890625
    },
    {
      "auxiliary_loss_clip": 0.01252223,
      "auxiliary_loss_mlp": 0.01026791,
      "balance_loss_clip": 1.05455589,
      "balance_loss_mlp": 1.01837468,
      "epoch": 0.30830277159862923,
      "flos": 16209969102720.0,
      "grad_norm": 4.015592856918152,
      "language_loss": 0.82638413,
      "learning_rate": 3.2412289768606495e-06,
      "loss": 0.84917426,
      "num_input_tokens_seen": 55131945,
      "step": 2564,
      "time_per_iteration": 2.6094212532043457
    },
    {
      "auxiliary_loss_clip": 0.01254934,
      "auxiliary_loss_mlp": 0.01035525,
      "balance_loss_clip": 1.05752993,
      "balance_loss_mlp": 1.02763319,
      "epoch": 0.30842301448926834,
      "flos": 29349503723520.0,
      "grad_norm": 1.9822157545160815,
      "language_loss": 0.82378417,
      "learning_rate": 3.240618078718718e-06,
      "loss": 0.84668875,
      "num_input_tokens_seen": 55153405,
      "step": 2565,
      "time_per_iteration": 3.6713907718658447
    },
    {
      "auxiliary_loss_clip": 0.01354066,
      "auxiliary_loss_mlp": 0.01039894,
      "balance_loss_clip": 1.05279922,
      "balance_loss_mlp": 1.03078032,
      "epoch": 0.3085432573799074,
      "flos": 21945190798080.0,
      "grad_norm": 1.9788295634654907,
      "language_loss": 0.73930508,
      "learning_rate": 3.240006992376011e-06,
      "loss": 0.76324475,
      "num_input_tokens_seen": 55173030,
      "step": 2566,
      "time_per_iteration": 2.740485906600952
    },
    {
      "auxiliary_loss_clip": 0.01309011,
      "auxiliary_loss_mlp": 0.01035593,
      "balance_loss_clip": 1.05854392,
      "balance_loss_mlp": 1.0272243,
      "epoch": 0.3086635002705465,
      "flos": 22054718344320.0,
      "grad_norm": 3.353193239022271,
      "language_loss": 0.76071918,
      "learning_rate": 3.2393957179252284e-06,
      "loss": 0.78416514,
      "num_input_tokens_seen": 55189565,
      "step": 2567,
      "time_per_iteration": 2.693664312362671
    },
    {
      "auxiliary_loss_clip": 0.01204481,
      "auxiliary_loss_mlp": 0.0103002,
      "balance_loss_clip": 1.05976403,
      "balance_loss_mlp": 1.02151406,
      "epoch": 0.3087837431611856,
      "flos": 32665925520000.0,
      "grad_norm": 2.224481049892084,
      "language_loss": 0.80255616,
      "learning_rate": 3.2387842554591016e-06,
      "loss": 0.82490122,
      "num_input_tokens_seen": 55210380,
      "step": 2568,
      "time_per_iteration": 3.6758198738098145
    },
    {
      "auxiliary_loss_clip": 0.01206593,
      "auxiliary_loss_mlp": 0.01035904,
      "balance_loss_clip": 1.06116414,
      "balance_loss_mlp": 1.02757168,
      "epoch": 0.3089039860518247,
      "flos": 17599245384960.0,
      "grad_norm": 2.7677424475190295,
      "language_loss": 0.88152587,
      "learning_rate": 3.238172605070388e-06,
      "loss": 0.90395081,
      "num_input_tokens_seen": 55225795,
      "step": 2569,
      "time_per_iteration": 2.585561990737915
    },
    {
      "auxiliary_loss_clip": 0.01252035,
      "auxiliary_loss_mlp": 0.0257224,
      "balance_loss_clip": 1.0574913,
      "balance_loss_mlp": 1.00007486,
      "epoch": 0.3090242289424638,
      "flos": 14383839611520.0,
      "grad_norm": 3.0914221368211208,
      "language_loss": 0.78696358,
      "learning_rate": 3.2375607668518745e-06,
      "loss": 0.82520628,
      "num_input_tokens_seen": 55238830,
      "step": 2570,
      "time_per_iteration": 2.6023786067962646
    },
    {
      "auxiliary_loss_clip": 0.01297431,
      "auxiliary_loss_mlp": 0.01032491,
      "balance_loss_clip": 1.0551827,
      "balance_loss_mlp": 1.02382493,
      "epoch": 0.30914447183310284,
      "flos": 16068625084800.0,
      "grad_norm": 2.4990489493216352,
      "language_loss": 0.90136087,
      "learning_rate": 3.236948740896377e-06,
      "loss": 0.92466009,
      "num_input_tokens_seen": 55253630,
      "step": 2571,
      "time_per_iteration": 2.7142250537872314
    },
    {
      "auxiliary_loss_clip": 0.01257607,
      "auxiliary_loss_mlp": 0.01033299,
      "balance_loss_clip": 1.06099617,
      "balance_loss_mlp": 1.02485895,
      "epoch": 0.30926471472374195,
      "flos": 32230221546240.0,
      "grad_norm": 1.567303887259897,
      "language_loss": 0.84342611,
      "learning_rate": 3.2363365272967384e-06,
      "loss": 0.86633515,
      "num_input_tokens_seen": 55276200,
      "step": 2572,
      "time_per_iteration": 2.727769374847412
    },
    {
      "auxiliary_loss_clip": 0.01262043,
      "auxiliary_loss_mlp": 0.01033711,
      "balance_loss_clip": 1.06499553,
      "balance_loss_mlp": 1.02450168,
      "epoch": 0.30938495761438106,
      "flos": 20370722970240.0,
      "grad_norm": 2.0399627037753842,
      "language_loss": 0.81581736,
      "learning_rate": 3.235724126145832e-06,
      "loss": 0.83877492,
      "num_input_tokens_seen": 55292235,
      "step": 2573,
      "time_per_iteration": 2.6365392208099365
    },
    {
      "auxiliary_loss_clip": 0.01245663,
      "auxiliary_loss_mlp": 0.0103035,
      "balance_loss_clip": 1.05361378,
      "balance_loss_mlp": 1.02160573,
      "epoch": 0.3095052005050201,
      "flos": 24061155131520.0,
      "grad_norm": 1.6509954668390137,
      "language_loss": 0.77672851,
      "learning_rate": 3.235111537536558e-06,
      "loss": 0.7994886,
      "num_input_tokens_seen": 55313050,
      "step": 2574,
      "time_per_iteration": 2.6493844985961914
    },
    {
      "auxiliary_loss_clip": 0.01254801,
      "auxiliary_loss_mlp": 0.01031738,
      "balance_loss_clip": 1.05844808,
      "balance_loss_mlp": 1.02398968,
      "epoch": 0.30962544339565923,
      "flos": 23401547729280.0,
      "grad_norm": 2.560953187789091,
      "language_loss": 0.83241045,
      "learning_rate": 3.2344987615618456e-06,
      "loss": 0.85527581,
      "num_input_tokens_seen": 55332885,
      "step": 2575,
      "time_per_iteration": 2.734856128692627
    },
    {
      "auxiliary_loss_clip": 0.01358466,
      "auxiliary_loss_mlp": 0.01032452,
      "balance_loss_clip": 1.05991209,
      "balance_loss_mlp": 1.02445948,
      "epoch": 0.30974568628629834,
      "flos": 33799984692480.0,
      "grad_norm": 1.8082393891677595,
      "language_loss": 0.7883271,
      "learning_rate": 3.2338857983146533e-06,
      "loss": 0.81223631,
      "num_input_tokens_seen": 55354385,
      "step": 2576,
      "time_per_iteration": 2.8065104484558105
    },
    {
      "auxiliary_loss_clip": 0.01296711,
      "auxiliary_loss_mlp": 0.01034541,
      "balance_loss_clip": 1.05661082,
      "balance_loss_mlp": 1.02561784,
      "epoch": 0.3098659291769374,
      "flos": 20229594433920.0,
      "grad_norm": 1.9312607628618408,
      "language_loss": 0.76604176,
      "learning_rate": 3.233272647887966e-06,
      "loss": 0.78935432,
      "num_input_tokens_seen": 55373275,
      "step": 2577,
      "time_per_iteration": 2.7169435024261475
    },
    {
      "auxiliary_loss_clip": 0.01207463,
      "auxiliary_loss_mlp": 0.01036724,
      "balance_loss_clip": 1.06110811,
      "balance_loss_mlp": 1.0279386,
      "epoch": 0.3099861720675765,
      "flos": 24748556682240.0,
      "grad_norm": 5.819643060748056,
      "language_loss": 0.90072691,
      "learning_rate": 3.2326593103747985e-06,
      "loss": 0.92316878,
      "num_input_tokens_seen": 55392290,
      "step": 2578,
      "time_per_iteration": 2.6125731468200684
    },
    {
      "auxiliary_loss_clip": 0.01254327,
      "auxiliary_loss_mlp": 0.01032307,
      "balance_loss_clip": 1.06139898,
      "balance_loss_mlp": 1.02413511,
      "epoch": 0.3101064149582156,
      "flos": 11765485704960.0,
      "grad_norm": 2.2000721076735457,
      "language_loss": 0.85142398,
      "learning_rate": 3.2320457858681936e-06,
      "loss": 0.87429029,
      "num_input_tokens_seen": 55410680,
      "step": 2579,
      "time_per_iteration": 2.653794050216675
    },
    {
      "auxiliary_loss_clip": 0.01300898,
      "auxiliary_loss_mlp": 0.0103566,
      "balance_loss_clip": 1.05523396,
      "balance_loss_mlp": 1.02783382,
      "epoch": 0.31022665784885467,
      "flos": 23033247626880.0,
      "grad_norm": 2.6754616918353036,
      "language_loss": 0.85365307,
      "learning_rate": 3.2314320744612228e-06,
      "loss": 0.87701869,
      "num_input_tokens_seen": 55425980,
      "step": 2580,
      "time_per_iteration": 2.685758113861084
    },
    {
      "auxiliary_loss_clip": 0.01248189,
      "auxiliary_loss_mlp": 0.01036298,
      "balance_loss_clip": 1.05706644,
      "balance_loss_mlp": 1.02825129,
      "epoch": 0.3103469007394938,
      "flos": 16289188548480.0,
      "grad_norm": 3.136061493129395,
      "language_loss": 0.76703787,
      "learning_rate": 3.2308181762469854e-06,
      "loss": 0.78988278,
      "num_input_tokens_seen": 55443925,
      "step": 2581,
      "time_per_iteration": 2.6801700592041016
    },
    {
      "auxiliary_loss_clip": 0.01209423,
      "auxiliary_loss_mlp": 0.01029757,
      "balance_loss_clip": 1.06164432,
      "balance_loss_mlp": 1.02154326,
      "epoch": 0.3104671436301329,
      "flos": 30515271626880.0,
      "grad_norm": 2.272375049921895,
      "language_loss": 0.78607845,
      "learning_rate": 3.230204091318609e-06,
      "loss": 0.80847025,
      "num_input_tokens_seen": 55464465,
      "step": 2582,
      "time_per_iteration": 2.690950393676758
    },
    {
      "auxiliary_loss_clip": 0.01203913,
      "auxiliary_loss_mlp": 0.02566156,
      "balance_loss_clip": 1.05988336,
      "balance_loss_mlp": 1.00005782,
      "epoch": 0.31058738652077195,
      "flos": 20047240062720.0,
      "grad_norm": 1.8734187265432434,
      "language_loss": 0.84478903,
      "learning_rate": 3.2295898197692503e-06,
      "loss": 0.88248968,
      "num_input_tokens_seen": 55483425,
      "step": 2583,
      "time_per_iteration": 2.6488585472106934
    },
    {
      "auxiliary_loss_clip": 0.01203201,
      "auxiliary_loss_mlp": 0.01029858,
      "balance_loss_clip": 1.0593245,
      "balance_loss_mlp": 1.02148962,
      "epoch": 0.31070762941141106,
      "flos": 28074639237120.0,
      "grad_norm": 3.0773850247500896,
      "language_loss": 0.79427254,
      "learning_rate": 3.228975361692094e-06,
      "loss": 0.81660312,
      "num_input_tokens_seen": 55504445,
      "step": 2584,
      "time_per_iteration": 3.688056230545044
    },
    {
      "auxiliary_loss_clip": 0.01257344,
      "auxiliary_loss_mlp": 0.02572953,
      "balance_loss_clip": 1.05572379,
      "balance_loss_mlp": 1.00009215,
      "epoch": 0.31082787230205017,
      "flos": 20521907314560.0,
      "grad_norm": 2.280309071984574,
      "language_loss": 0.80451769,
      "learning_rate": 3.228360717180352e-06,
      "loss": 0.84282064,
      "num_input_tokens_seen": 55521970,
      "step": 2585,
      "time_per_iteration": 2.6695268154144287
    },
    {
      "auxiliary_loss_clip": 0.01107311,
      "auxiliary_loss_mlp": 0.02517025,
      "balance_loss_clip": 1.0330776,
      "balance_loss_mlp": 1.00003421,
      "epoch": 0.3109481151926892,
      "flos": 62445928723200.0,
      "grad_norm": 0.8104947979672548,
      "language_loss": 0.59378445,
      "learning_rate": 3.227745886327266e-06,
      "loss": 0.63002783,
      "num_input_tokens_seen": 55580665,
      "step": 2586,
      "time_per_iteration": 3.9291703701019287
    },
    {
      "auxiliary_loss_clip": 0.01107244,
      "auxiliary_loss_mlp": 0.01004126,
      "balance_loss_clip": 1.03292561,
      "balance_loss_mlp": 1.00304151,
      "epoch": 0.31106835808332833,
      "flos": 44746744723200.0,
      "grad_norm": 0.8323980271178039,
      "language_loss": 0.55842137,
      "learning_rate": 3.227130869226105e-06,
      "loss": 0.57953513,
      "num_input_tokens_seen": 55637825,
      "step": 2587,
      "time_per_iteration": 3.197234869003296
    },
    {
      "auxiliary_loss_clip": 0.01256879,
      "auxiliary_loss_mlp": 0.01029961,
      "balance_loss_clip": 1.05958319,
      "balance_loss_mlp": 1.02170539,
      "epoch": 0.3111886009739674,
      "flos": 23403056100480.0,
      "grad_norm": 2.6996845583101314,
      "language_loss": 0.82914019,
      "learning_rate": 3.226515665970167e-06,
      "loss": 0.85200846,
      "num_input_tokens_seen": 55655365,
      "step": 2588,
      "time_per_iteration": 2.6235191822052
    },
    {
      "auxiliary_loss_clip": 0.01253638,
      "auxiliary_loss_mlp": 0.01028366,
      "balance_loss_clip": 1.05981708,
      "balance_loss_mlp": 1.01983023,
      "epoch": 0.3113088438646065,
      "flos": 17530728192000.0,
      "grad_norm": 2.3039205507084506,
      "language_loss": 0.86523569,
      "learning_rate": 3.225900276652777e-06,
      "loss": 0.88805568,
      "num_input_tokens_seen": 55672140,
      "step": 2589,
      "time_per_iteration": 2.6641392707824707
    },
    {
      "auxiliary_loss_clip": 0.01309584,
      "auxiliary_loss_mlp": 0.01035138,
      "balance_loss_clip": 1.05664945,
      "balance_loss_mlp": 1.02785993,
      "epoch": 0.3114290867552456,
      "flos": 28365802882560.0,
      "grad_norm": 1.6719137740631647,
      "language_loss": 0.75417501,
      "learning_rate": 3.2252847013672906e-06,
      "loss": 0.77762222,
      "num_input_tokens_seen": 55694800,
      "step": 2590,
      "time_per_iteration": 2.733449935913086
    },
    {
      "auxiliary_loss_clip": 0.01349142,
      "auxiliary_loss_mlp": 0.01038151,
      "balance_loss_clip": 1.0538888,
      "balance_loss_mlp": 1.02962804,
      "epoch": 0.31154932964588467,
      "flos": 27379157126400.0,
      "grad_norm": 2.0349243057469755,
      "language_loss": 0.75895315,
      "learning_rate": 3.224668940207089e-06,
      "loss": 0.78282607,
      "num_input_tokens_seen": 55713785,
      "step": 2591,
      "time_per_iteration": 3.760448455810547
    },
    {
      "auxiliary_loss_clip": 0.01398454,
      "auxiliary_loss_mlp": 0.01030171,
      "balance_loss_clip": 1.04899609,
      "balance_loss_mlp": 1.02186847,
      "epoch": 0.3116695725365238,
      "flos": 26541864120960.0,
      "grad_norm": 2.212640223224331,
      "language_loss": 0.87108296,
      "learning_rate": 3.2240529932655828e-06,
      "loss": 0.89536929,
      "num_input_tokens_seen": 55733050,
      "step": 2592,
      "time_per_iteration": 2.8786935806274414
    },
    {
      "auxiliary_loss_clip": 0.0130561,
      "auxiliary_loss_mlp": 0.01029506,
      "balance_loss_clip": 1.06143069,
      "balance_loss_mlp": 1.02124488,
      "epoch": 0.3117898154271629,
      "flos": 21177600134400.0,
      "grad_norm": 2.6297873591271523,
      "language_loss": 0.88620102,
      "learning_rate": 3.223436860636211e-06,
      "loss": 0.90955216,
      "num_input_tokens_seen": 55748685,
      "step": 2593,
      "time_per_iteration": 2.753444194793701
    },
    {
      "auxiliary_loss_clip": 0.01205636,
      "auxiliary_loss_mlp": 0.01028164,
      "balance_loss_clip": 1.06051803,
      "balance_loss_mlp": 1.02057612,
      "epoch": 0.31191005831780194,
      "flos": 27272430840960.0,
      "grad_norm": 1.7254291564189985,
      "language_loss": 0.73916161,
      "learning_rate": 3.2228205424124403e-06,
      "loss": 0.76149964,
      "num_input_tokens_seen": 55771840,
      "step": 2594,
      "time_per_iteration": 3.726764440536499
    },
    {
      "auxiliary_loss_clip": 0.01293893,
      "auxiliary_loss_mlp": 0.01035372,
      "balance_loss_clip": 1.05482888,
      "balance_loss_mlp": 1.02752244,
      "epoch": 0.31203030120844105,
      "flos": 12963501043200.0,
      "grad_norm": 2.4478722514610975,
      "language_loss": 0.74964136,
      "learning_rate": 3.222204038687765e-06,
      "loss": 0.77293396,
      "num_input_tokens_seen": 55784975,
      "step": 2595,
      "time_per_iteration": 2.620241641998291
    },
    {
      "auxiliary_loss_clip": 0.01254651,
      "auxiliary_loss_mlp": 0.0102947,
      "balance_loss_clip": 1.06047249,
      "balance_loss_mlp": 1.02171564,
      "epoch": 0.31215054409908016,
      "flos": 27562014288000.0,
      "grad_norm": 1.8520011119193083,
      "language_loss": 0.88272351,
      "learning_rate": 3.221587349555709e-06,
      "loss": 0.90556479,
      "num_input_tokens_seen": 55805235,
      "step": 2596,
      "time_per_iteration": 2.724022388458252
    },
    {
      "auxiliary_loss_clip": 0.01310072,
      "auxiliary_loss_mlp": 0.01028608,
      "balance_loss_clip": 1.05961442,
      "balance_loss_mlp": 1.0208478,
      "epoch": 0.3122707869897192,
      "flos": 21506326427520.0,
      "grad_norm": 1.667087980491596,
      "language_loss": 0.69487178,
      "learning_rate": 3.2209704751098236e-06,
      "loss": 0.7182585,
      "num_input_tokens_seen": 55824265,
      "step": 2597,
      "time_per_iteration": 2.6754043102264404
    },
    {
      "auxiliary_loss_clip": 0.01308717,
      "auxiliary_loss_mlp": 0.0103597,
      "balance_loss_clip": 1.06111884,
      "balance_loss_mlp": 1.02802432,
      "epoch": 0.31239102988035833,
      "flos": 15187017674880.0,
      "grad_norm": 2.1665709191409817,
      "language_loss": 0.8284201,
      "learning_rate": 3.2203534154436875e-06,
      "loss": 0.85186696,
      "num_input_tokens_seen": 55838620,
      "step": 2598,
      "time_per_iteration": 2.647808313369751
    },
    {
      "auxiliary_loss_clip": 0.01456195,
      "auxiliary_loss_mlp": 0.01036203,
      "balance_loss_clip": 1.05263519,
      "balance_loss_mlp": 1.02793562,
      "epoch": 0.31251127277099744,
      "flos": 22053712763520.0,
      "grad_norm": 3.228391160571132,
      "language_loss": 0.7574358,
      "learning_rate": 3.2197361706509084e-06,
      "loss": 0.78235978,
      "num_input_tokens_seen": 55859375,
      "step": 2599,
      "time_per_iteration": 2.9194610118865967
    },
    {
      "auxiliary_loss_clip": 0.01208846,
      "auxiliary_loss_mlp": 0.01032172,
      "balance_loss_clip": 1.06102347,
      "balance_loss_mlp": 1.02255738,
      "epoch": 0.3126315156616365,
      "flos": 15193984913280.0,
      "grad_norm": 2.568158748483563,
      "language_loss": 0.83345377,
      "learning_rate": 3.2191187408251228e-06,
      "loss": 0.85586399,
      "num_input_tokens_seen": 55876535,
      "step": 2600,
      "time_per_iteration": 2.7066690921783447
    },
    {
      "auxiliary_loss_clip": 0.01172101,
      "auxiliary_loss_mlp": 0.01033967,
      "balance_loss_clip": 1.05897069,
      "balance_loss_mlp": 1.02449012,
      "epoch": 0.3127517585522756,
      "flos": 18145338831360.0,
      "grad_norm": 2.2744671706200794,
      "language_loss": 0.79495949,
      "learning_rate": 3.218501126059993e-06,
      "loss": 0.81702012,
      "num_input_tokens_seen": 55891930,
      "step": 2601,
      "time_per_iteration": 2.6325268745422363
    },
    {
      "auxiliary_loss_clip": 0.01256971,
      "auxiliary_loss_mlp": 0.01030007,
      "balance_loss_clip": 1.05627525,
      "balance_loss_mlp": 1.02150726,
      "epoch": 0.31287200144291466,
      "flos": 21908633731200.0,
      "grad_norm": 2.3161342847400683,
      "language_loss": 0.8171888,
      "learning_rate": 3.2178833264492116e-06,
      "loss": 0.84005857,
      "num_input_tokens_seen": 55910635,
      "step": 2602,
      "time_per_iteration": 2.641519784927368
    },
    {
      "auxiliary_loss_clip": 0.01263344,
      "auxiliary_loss_mlp": 0.01031188,
      "balance_loss_clip": 1.06111741,
      "balance_loss_mlp": 1.02302837,
      "epoch": 0.31299224433355377,
      "flos": 29896997800320.0,
      "grad_norm": 2.189328587643372,
      "language_loss": 0.76024026,
      "learning_rate": 3.217265342086498e-06,
      "loss": 0.7831856,
      "num_input_tokens_seen": 55931125,
      "step": 2603,
      "time_per_iteration": 2.8644754886627197
    },
    {
      "auxiliary_loss_clip": 0.01367165,
      "auxiliary_loss_mlp": 0.02576441,
      "balance_loss_clip": 1.06262064,
      "balance_loss_mlp": 1.00011706,
      "epoch": 0.3131124872241929,
      "flos": 11655886331520.0,
      "grad_norm": 2.3854601730618055,
      "language_loss": 0.73357391,
      "learning_rate": 3.216647173065599e-06,
      "loss": 0.77301002,
      "num_input_tokens_seen": 55946590,
      "step": 2604,
      "time_per_iteration": 2.7110767364501953
    },
    {
      "auxiliary_loss_clip": 0.0130672,
      "auxiliary_loss_mlp": 0.01033072,
      "balance_loss_clip": 1.06281352,
      "balance_loss_mlp": 1.02442908,
      "epoch": 0.31323273011483194,
      "flos": 49848785470080.0,
      "grad_norm": 2.0629840639839996,
      "language_loss": 0.7400552,
      "learning_rate": 3.216028819480292e-06,
      "loss": 0.76345313,
      "num_input_tokens_seen": 55967930,
      "step": 2605,
      "time_per_iteration": 2.97039532661438
    },
    {
      "auxiliary_loss_clip": 0.01300112,
      "auxiliary_loss_mlp": 0.01025335,
      "balance_loss_clip": 1.05938387,
      "balance_loss_mlp": 1.01755977,
      "epoch": 0.31335297300547105,
      "flos": 22601278667520.0,
      "grad_norm": 1.8865230500203014,
      "language_loss": 0.75654948,
      "learning_rate": 3.2154102814243793e-06,
      "loss": 0.77980399,
      "num_input_tokens_seen": 55987070,
      "step": 2606,
      "time_per_iteration": 2.712132453918457
    },
    {
      "auxiliary_loss_clip": 0.01359426,
      "auxiliary_loss_mlp": 0.01032345,
      "balance_loss_clip": 1.05772161,
      "balance_loss_mlp": 1.02432859,
      "epoch": 0.31347321589611016,
      "flos": 34710858708480.0,
      "grad_norm": 1.936714142345138,
      "language_loss": 0.67032695,
      "learning_rate": 3.2147915589916937e-06,
      "loss": 0.69424462,
      "num_input_tokens_seen": 56008630,
      "step": 2607,
      "time_per_iteration": 2.7964301109313965
    },
    {
      "auxiliary_loss_clip": 0.01301666,
      "auxiliary_loss_mlp": 0.01033256,
      "balance_loss_clip": 1.05456531,
      "balance_loss_mlp": 1.02519774,
      "epoch": 0.3135934587867492,
      "flos": 19755789108480.0,
      "grad_norm": 1.977524694433244,
      "language_loss": 0.83059335,
      "learning_rate": 3.2141726522760938e-06,
      "loss": 0.85394251,
      "num_input_tokens_seen": 56026690,
      "step": 2608,
      "time_per_iteration": 2.657794237136841
    },
    {
      "auxiliary_loss_clip": 0.01153889,
      "auxiliary_loss_mlp": 0.01006151,
      "balance_loss_clip": 1.03073955,
      "balance_loss_mlp": 1.00495899,
      "epoch": 0.3137137016773883,
      "flos": 65815535583360.0,
      "grad_norm": 0.7050518329998078,
      "language_loss": 0.52671015,
      "learning_rate": 3.213553561371469e-06,
      "loss": 0.54831058,
      "num_input_tokens_seen": 56090425,
      "step": 2609,
      "time_per_iteration": 3.2950243949890137
    },
    {
      "auxiliary_loss_clip": 0.0140312,
      "auxiliary_loss_mlp": 0.01031283,
      "balance_loss_clip": 1.05880332,
      "balance_loss_mlp": 1.02329648,
      "epoch": 0.31383394456802743,
      "flos": 16252739222400.0,
      "grad_norm": 2.4308381432049107,
      "language_loss": 0.95758235,
      "learning_rate": 3.212934286371733e-06,
      "loss": 0.98192644,
      "num_input_tokens_seen": 56107135,
      "step": 2610,
      "time_per_iteration": 3.6874406337738037
    },
    {
      "auxiliary_loss_clip": 0.01265827,
      "auxiliary_loss_mlp": 0.01040609,
      "balance_loss_clip": 1.06736541,
      "balance_loss_mlp": 1.03182912,
      "epoch": 0.3139541874586665,
      "flos": 38795517613440.0,
      "grad_norm": 2.2437586119147745,
      "language_loss": 0.83287954,
      "learning_rate": 3.2123148273708304e-06,
      "loss": 0.85594392,
      "num_input_tokens_seen": 56127325,
      "step": 2611,
      "time_per_iteration": 2.901592493057251
    },
    {
      "auxiliary_loss_clip": 0.01206533,
      "auxiliary_loss_mlp": 0.01029842,
      "balance_loss_clip": 1.06112742,
      "balance_loss_mlp": 1.02199793,
      "epoch": 0.3140744303493056,
      "flos": 25046328430080.0,
      "grad_norm": 1.9885285550459908,
      "language_loss": 0.76903784,
      "learning_rate": 3.211695184462733e-06,
      "loss": 0.79140151,
      "num_input_tokens_seen": 56148500,
      "step": 2612,
      "time_per_iteration": 3.5375256538391113
    },
    {
      "auxiliary_loss_clip": 0.0116732,
      "auxiliary_loss_mlp": 0.01002378,
      "balance_loss_clip": 1.02844024,
      "balance_loss_mlp": 1.0011797,
      "epoch": 0.3141946732399447,
      "flos": 72504254782080.0,
      "grad_norm": 0.8951932624017819,
      "language_loss": 0.60478055,
      "learning_rate": 3.2110753577414383e-06,
      "loss": 0.6264776,
      "num_input_tokens_seen": 56210080,
      "step": 2613,
      "time_per_iteration": 3.2703185081481934
    },
    {
      "auxiliary_loss_clip": 0.01304312,
      "auxiliary_loss_mlp": 0.01034386,
      "balance_loss_clip": 1.05470657,
      "balance_loss_mlp": 1.02642918,
      "epoch": 0.31431491613058377,
      "flos": 19239788280960.0,
      "grad_norm": 1.8547883805677348,
      "language_loss": 0.79282504,
      "learning_rate": 3.2104553473009757e-06,
      "loss": 0.81621206,
      "num_input_tokens_seen": 56228200,
      "step": 2614,
      "time_per_iteration": 2.6808419227600098
    },
    {
      "auxiliary_loss_clip": 0.01397867,
      "auxiliary_loss_mlp": 0.01030678,
      "balance_loss_clip": 1.05474305,
      "balance_loss_mlp": 1.02238166,
      "epoch": 0.3144351590212229,
      "flos": 36210596290560.0,
      "grad_norm": 3.2184628539957,
      "language_loss": 0.67738903,
      "learning_rate": 3.209835153235399e-06,
      "loss": 0.70167446,
      "num_input_tokens_seen": 56249755,
      "step": 2615,
      "time_per_iteration": 2.8863797187805176
    },
    {
      "auxiliary_loss_clip": 0.01258476,
      "auxiliary_loss_mlp": 0.01035241,
      "balance_loss_clip": 1.05202198,
      "balance_loss_mlp": 1.02742672,
      "epoch": 0.314555401911862,
      "flos": 18551740285440.0,
      "grad_norm": 1.7007504015998862,
      "language_loss": 0.67697024,
      "learning_rate": 3.2092147756387916e-06,
      "loss": 0.69990736,
      "num_input_tokens_seen": 56270080,
      "step": 2616,
      "time_per_iteration": 2.790973424911499
    },
    {
      "auxiliary_loss_clip": 0.01299107,
      "auxiliary_loss_mlp": 0.01028711,
      "balance_loss_clip": 1.05315101,
      "balance_loss_mlp": 1.0202111,
      "epoch": 0.31467564480250104,
      "flos": 16362877299840.0,
      "grad_norm": 1.8453704888393034,
      "language_loss": 0.83759099,
      "learning_rate": 3.208594214605264e-06,
      "loss": 0.86086917,
      "num_input_tokens_seen": 56288625,
      "step": 2617,
      "time_per_iteration": 3.732304334640503
    },
    {
      "auxiliary_loss_clip": 0.01294631,
      "auxiliary_loss_mlp": 0.0102312,
      "balance_loss_clip": 1.05522943,
      "balance_loss_mlp": 1.01566029,
      "epoch": 0.31479588769314015,
      "flos": 21652375127040.0,
      "grad_norm": 2.1848674092657827,
      "language_loss": 0.77311689,
      "learning_rate": 3.2079734702289553e-06,
      "loss": 0.79629439,
      "num_input_tokens_seen": 56307520,
      "step": 2618,
      "time_per_iteration": 2.718733072280884
    },
    {
      "auxiliary_loss_clip": 0.01146531,
      "auxiliary_loss_mlp": 0.02516265,
      "balance_loss_clip": 1.02582407,
      "balance_loss_mlp": 0.99998778,
      "epoch": 0.3149161305837792,
      "flos": 66051072040320.0,
      "grad_norm": 0.8017900479752503,
      "language_loss": 0.60366929,
      "learning_rate": 3.207352542604031e-06,
      "loss": 0.64029723,
      "num_input_tokens_seen": 56369855,
      "step": 2619,
      "time_per_iteration": 3.3148999214172363
    },
    {
      "auxiliary_loss_clip": 0.01342324,
      "auxiliary_loss_mlp": 0.01032421,
      "balance_loss_clip": 1.05376124,
      "balance_loss_mlp": 1.02497911,
      "epoch": 0.3150363734744183,
      "flos": 28987201192320.0,
      "grad_norm": 2.390757715221807,
      "language_loss": 0.78314531,
      "learning_rate": 3.2067314318246864e-06,
      "loss": 0.80689275,
      "num_input_tokens_seen": 56390570,
      "step": 2620,
      "time_per_iteration": 3.7029073238372803
    },
    {
      "auxiliary_loss_clip": 0.0135603,
      "auxiliary_loss_mlp": 0.0102822,
      "balance_loss_clip": 1.05998504,
      "balance_loss_mlp": 1.01958954,
      "epoch": 0.31515661636505743,
      "flos": 27636600879360.0,
      "grad_norm": 4.402638896751761,
      "language_loss": 0.77667427,
      "learning_rate": 3.206110137985143e-06,
      "loss": 0.80051684,
      "num_input_tokens_seen": 56410775,
      "step": 2621,
      "time_per_iteration": 2.825878620147705
    },
    {
      "auxiliary_loss_clip": 0.01345006,
      "auxiliary_loss_mlp": 0.01028894,
      "balance_loss_clip": 1.05534995,
      "balance_loss_mlp": 1.02106762,
      "epoch": 0.3152768592556965,
      "flos": 24605632465920.0,
      "grad_norm": 2.0470862330913158,
      "language_loss": 0.92449665,
      "learning_rate": 3.2054886611796505e-06,
      "loss": 0.94823557,
      "num_input_tokens_seen": 56429770,
      "step": 2622,
      "time_per_iteration": 2.74845552444458
    },
    {
      "auxiliary_loss_clip": 0.01100579,
      "auxiliary_loss_mlp": 0.01000162,
      "balance_loss_clip": 1.02807856,
      "balance_loss_mlp": 0.99902326,
      "epoch": 0.3153971021463356,
      "flos": 68476908026880.0,
      "grad_norm": 0.8852333712054345,
      "language_loss": 0.63478971,
      "learning_rate": 3.204867001502487e-06,
      "loss": 0.65579712,
      "num_input_tokens_seen": 56488425,
      "step": 2623,
      "time_per_iteration": 3.215188980102539
    },
    {
      "auxiliary_loss_clip": 0.0120637,
      "auxiliary_loss_mlp": 0.0103167,
      "balance_loss_clip": 1.06150568,
      "balance_loss_mlp": 1.02317023,
      "epoch": 0.3155173450369747,
      "flos": 25593714766080.0,
      "grad_norm": 3.127188905721485,
      "language_loss": 0.81127769,
      "learning_rate": 3.2042451590479567e-06,
      "loss": 0.8336581,
      "num_input_tokens_seen": 56508940,
      "step": 2624,
      "time_per_iteration": 2.687311887741089
    },
    {
      "auxiliary_loss_clip": 0.0120062,
      "auxiliary_loss_mlp": 0.01028356,
      "balance_loss_clip": 1.05941081,
      "balance_loss_mlp": 1.02081609,
      "epoch": 0.31563758792761376,
      "flos": 24309333175680.0,
      "grad_norm": 2.0302451979005802,
      "language_loss": 0.87010062,
      "learning_rate": 3.203623133910394e-06,
      "loss": 0.89239037,
      "num_input_tokens_seen": 56527245,
      "step": 2625,
      "time_per_iteration": 2.6549482345581055
    },
    {
      "auxiliary_loss_clip": 0.01358466,
      "auxiliary_loss_mlp": 0.01029363,
      "balance_loss_clip": 1.0503552,
      "balance_loss_mlp": 1.02105403,
      "epoch": 0.31575783081825287,
      "flos": 31903865550720.0,
      "grad_norm": 2.3108245918482395,
      "language_loss": 0.77736545,
      "learning_rate": 3.203000926184158e-06,
      "loss": 0.80124372,
      "num_input_tokens_seen": 56546170,
      "step": 2626,
      "time_per_iteration": 2.8824033737182617
    },
    {
      "auxiliary_loss_clip": 0.0120297,
      "auxiliary_loss_mlp": 0.0102895,
      "balance_loss_clip": 1.06040978,
      "balance_loss_mlp": 1.02139211,
      "epoch": 0.315878073708892,
      "flos": 30810960385920.0,
      "grad_norm": 1.9777753666211824,
      "language_loss": 0.77584028,
      "learning_rate": 3.202378535963639e-06,
      "loss": 0.79815948,
      "num_input_tokens_seen": 56567085,
      "step": 2627,
      "time_per_iteration": 2.6417059898376465
    },
    {
      "auxiliary_loss_clip": 0.01294119,
      "auxiliary_loss_mlp": 0.02574313,
      "balance_loss_clip": 1.05333233,
      "balance_loss_mlp": 1.00019026,
      "epoch": 0.31599831659953104,
      "flos": 22200264253440.0,
      "grad_norm": 1.7332815697587876,
      "language_loss": 0.83744431,
      "learning_rate": 3.2017559633432516e-06,
      "loss": 0.87612861,
      "num_input_tokens_seen": 56586715,
      "step": 2628,
      "time_per_iteration": 2.7136871814727783
    },
    {
      "auxiliary_loss_clip": 0.01220465,
      "auxiliary_loss_mlp": 0.01031985,
      "balance_loss_clip": 1.05743194,
      "balance_loss_mlp": 1.02423668,
      "epoch": 0.31611855949017015,
      "flos": 25593463370880.0,
      "grad_norm": 2.1370897876276964,
      "language_loss": 0.66224897,
      "learning_rate": 3.2011332084174398e-06,
      "loss": 0.68477345,
      "num_input_tokens_seen": 56607585,
      "step": 2629,
      "time_per_iteration": 2.7201805114746094
    },
    {
      "auxiliary_loss_clip": 0.01251433,
      "auxiliary_loss_mlp": 0.0103244,
      "balance_loss_clip": 1.05958056,
      "balance_loss_mlp": 1.02392292,
      "epoch": 0.31623880238080926,
      "flos": 20594087694720.0,
      "grad_norm": 1.8572268910444019,
      "language_loss": 0.89153993,
      "learning_rate": 3.2005102712806756e-06,
      "loss": 0.91437864,
      "num_input_tokens_seen": 56626415,
      "step": 2630,
      "time_per_iteration": 2.699655771255493
    },
    {
      "auxiliary_loss_clip": 0.01258004,
      "auxiliary_loss_mlp": 0.01040211,
      "balance_loss_clip": 1.05971277,
      "balance_loss_mlp": 1.03190827,
      "epoch": 0.3163590452714483,
      "flos": 12784917600000.0,
      "grad_norm": 2.371543973308338,
      "language_loss": 0.73298109,
      "learning_rate": 3.1998871520274575e-06,
      "loss": 0.75596333,
      "num_input_tokens_seen": 56641750,
      "step": 2631,
      "time_per_iteration": 2.5854902267456055
    },
    {
      "auxiliary_loss_clip": 0.01305815,
      "auxiliary_loss_mlp": 0.01034943,
      "balance_loss_clip": 1.05491173,
      "balance_loss_mlp": 1.02657402,
      "epoch": 0.3164792881620874,
      "flos": 23041292273280.0,
      "grad_norm": 1.978370651424368,
      "language_loss": 0.85087037,
      "learning_rate": 3.199263850752312e-06,
      "loss": 0.87427801,
      "num_input_tokens_seen": 56662585,
      "step": 2632,
      "time_per_iteration": 2.6568939685821533
    },
    {
      "auxiliary_loss_clip": 0.01254642,
      "auxiliary_loss_mlp": 0.01028002,
      "balance_loss_clip": 1.05916095,
      "balance_loss_mlp": 1.01972342,
      "epoch": 0.31659953105272653,
      "flos": 18296271780480.0,
      "grad_norm": 2.193516818986262,
      "language_loss": 0.86128891,
      "learning_rate": 3.198640367549795e-06,
      "loss": 0.8841154,
      "num_input_tokens_seen": 56681480,
      "step": 2633,
      "time_per_iteration": 2.5120790004730225
    },
    {
      "auxiliary_loss_clip": 0.01251963,
      "auxiliary_loss_mlp": 0.02566585,
      "balance_loss_clip": 1.05615687,
      "balance_loss_mlp": 1.00010931,
      "epoch": 0.3167197739433656,
      "flos": 25703421880320.0,
      "grad_norm": 1.8682370984114252,
      "language_loss": 0.85918224,
      "learning_rate": 3.198016702514487e-06,
      "loss": 0.89736766,
      "num_input_tokens_seen": 56701760,
      "step": 2634,
      "time_per_iteration": 2.58004093170166
    },
    {
      "auxiliary_loss_clip": 0.01202193,
      "auxiliary_loss_mlp": 0.01033024,
      "balance_loss_clip": 1.05894196,
      "balance_loss_mlp": 1.02525783,
      "epoch": 0.3168400168340047,
      "flos": 23546016230400.0,
      "grad_norm": 1.8367441601476888,
      "language_loss": 0.84710979,
      "learning_rate": 3.1973928557409972e-06,
      "loss": 0.86946195,
      "num_input_tokens_seen": 56719800,
      "step": 2635,
      "time_per_iteration": 2.5049662590026855
    },
    {
      "auxiliary_loss_clip": 0.01199098,
      "auxiliary_loss_mlp": 0.01030581,
      "balance_loss_clip": 1.05780101,
      "balance_loss_mlp": 1.02304077,
      "epoch": 0.31696025972464376,
      "flos": 28366449327360.0,
      "grad_norm": 1.811254197334729,
      "language_loss": 0.71188992,
      "learning_rate": 3.1967688273239636e-06,
      "loss": 0.73418677,
      "num_input_tokens_seen": 56739605,
      "step": 2636,
      "time_per_iteration": 3.4778122901916504
    },
    {
      "auxiliary_loss_clip": 0.01256921,
      "auxiliary_loss_mlp": 0.01031362,
      "balance_loss_clip": 1.0547595,
      "balance_loss_mlp": 1.02283263,
      "epoch": 0.31708050261528287,
      "flos": 16399111144320.0,
      "grad_norm": 1.8445643560214395,
      "language_loss": 0.82240534,
      "learning_rate": 3.1961446173580503e-06,
      "loss": 0.84528816,
      "num_input_tokens_seen": 56756545,
      "step": 2637,
      "time_per_iteration": 2.648057222366333
    },
    {
      "auxiliary_loss_clip": 0.01298445,
      "auxiliary_loss_mlp": 0.01040947,
      "balance_loss_clip": 1.05530131,
      "balance_loss_mlp": 1.0331502,
      "epoch": 0.317200745505922,
      "flos": 26212347728640.0,
      "grad_norm": 1.9843795439982141,
      "language_loss": 0.7735458,
      "learning_rate": 3.1955202259379502e-06,
      "loss": 0.79693967,
      "num_input_tokens_seen": 56778275,
      "step": 2638,
      "time_per_iteration": 3.7504994869232178
    },
    {
      "auxiliary_loss_clip": 0.01248224,
      "auxiliary_loss_mlp": 0.01032965,
      "balance_loss_clip": 1.05383098,
      "balance_loss_mlp": 1.02465045,
      "epoch": 0.31732098839656103,
      "flos": 31350876693120.0,
      "grad_norm": 1.8478426675070594,
      "language_loss": 0.82787496,
      "learning_rate": 3.194895653158381e-06,
      "loss": 0.85068685,
      "num_input_tokens_seen": 56797215,
      "step": 2639,
      "time_per_iteration": 2.719618558883667
    },
    {
      "auxiliary_loss_clip": 0.01099874,
      "auxiliary_loss_mlp": 0.01000709,
      "balance_loss_clip": 1.02831197,
      "balance_loss_mlp": 0.99962467,
      "epoch": 0.31744123128720014,
      "flos": 58989024835200.0,
      "grad_norm": 0.7699304627334224,
      "language_loss": 0.55478692,
      "learning_rate": 3.194270899114093e-06,
      "loss": 0.57579279,
      "num_input_tokens_seen": 56863010,
      "step": 2640,
      "time_per_iteration": 3.2789809703826904
    },
    {
      "auxiliary_loss_clip": 0.01261305,
      "auxiliary_loss_mlp": 0.01031117,
      "balance_loss_clip": 1.05980361,
      "balance_loss_mlp": 1.02248597,
      "epoch": 0.31756147417783925,
      "flos": 17417573372160.0,
      "grad_norm": 1.6886904271210978,
      "language_loss": 0.82046837,
      "learning_rate": 3.193645963899858e-06,
      "loss": 0.84339261,
      "num_input_tokens_seen": 56880625,
      "step": 2641,
      "time_per_iteration": 2.581281900405884
    },
    {
      "auxiliary_loss_clip": 0.0129699,
      "auxiliary_loss_mlp": 0.01025089,
      "balance_loss_clip": 1.05566645,
      "balance_loss_mlp": 1.01725078,
      "epoch": 0.3176817170684783,
      "flos": 25481673267840.0,
      "grad_norm": 1.8294497974816792,
      "language_loss": 0.83934969,
      "learning_rate": 3.193020847610479e-06,
      "loss": 0.86257052,
      "num_input_tokens_seen": 56900945,
      "step": 2642,
      "time_per_iteration": 2.7501516342163086
    },
    {
      "auxiliary_loss_clip": 0.01299596,
      "auxiliary_loss_mlp": 0.01029933,
      "balance_loss_clip": 1.05723834,
      "balance_loss_mlp": 1.02171373,
      "epoch": 0.3178019599591174,
      "flos": 24972603765120.0,
      "grad_norm": 2.51364756056584,
      "language_loss": 0.71619183,
      "learning_rate": 3.192395550340787e-06,
      "loss": 0.73948711,
      "num_input_tokens_seen": 56918895,
      "step": 2643,
      "time_per_iteration": 2.7007617950439453
    },
    {
      "auxiliary_loss_clip": 0.0125213,
      "auxiliary_loss_mlp": 0.01027827,
      "balance_loss_clip": 1.06042695,
      "balance_loss_mlp": 1.01972675,
      "epoch": 0.31792220284975653,
      "flos": 12422220019200.0,
      "grad_norm": 2.612765236694129,
      "language_loss": 0.76524568,
      "learning_rate": 3.191770072185638e-06,
      "loss": 0.78804529,
      "num_input_tokens_seen": 56935890,
      "step": 2644,
      "time_per_iteration": 3.5311789512634277
    },
    {
      "auxiliary_loss_clip": 0.01249309,
      "auxiliary_loss_mlp": 0.01034126,
      "balance_loss_clip": 1.05841804,
      "balance_loss_mlp": 1.02616262,
      "epoch": 0.3180424457403956,
      "flos": 15485759089920.0,
      "grad_norm": 3.0292223950007764,
      "language_loss": 0.7276386,
      "learning_rate": 3.191144413239916e-06,
      "loss": 0.7504729,
      "num_input_tokens_seen": 56952460,
      "step": 2645,
      "time_per_iteration": 2.6155741214752197
    },
    {
      "auxiliary_loss_clip": 0.01303033,
      "auxiliary_loss_mlp": 0.01035237,
      "balance_loss_clip": 1.05682254,
      "balance_loss_mlp": 1.0263617,
      "epoch": 0.3181626886310347,
      "flos": 26174964648960.0,
      "grad_norm": 2.291325710728498,
      "language_loss": 0.88244081,
      "learning_rate": 3.190518573598534e-06,
      "loss": 0.90582347,
      "num_input_tokens_seen": 56969065,
      "step": 2646,
      "time_per_iteration": 3.6767804622650146
    },
    {
      "auxiliary_loss_clip": 0.01268043,
      "auxiliary_loss_mlp": 0.01033095,
      "balance_loss_clip": 1.05399013,
      "balance_loss_mlp": 1.0249238,
      "epoch": 0.3182829315216738,
      "flos": 25483109811840.0,
      "grad_norm": 2.0747474300194906,
      "language_loss": 0.77235401,
      "learning_rate": 3.1898925533564308e-06,
      "loss": 0.79536545,
      "num_input_tokens_seen": 56990535,
      "step": 2647,
      "time_per_iteration": 2.7842373847961426
    },
    {
      "auxiliary_loss_clip": 0.01348471,
      "auxiliary_loss_mlp": 0.01029112,
      "balance_loss_clip": 1.05457294,
      "balance_loss_mlp": 1.02098155,
      "epoch": 0.31840317441231286,
      "flos": 18113701927680.0,
      "grad_norm": 3.6533042660697745,
      "language_loss": 0.64172709,
      "learning_rate": 3.1892663526085733e-06,
      "loss": 0.66550291,
      "num_input_tokens_seen": 57008910,
      "step": 2648,
      "time_per_iteration": 2.6983659267425537
    },
    {
      "auxiliary_loss_clip": 0.01097223,
      "auxiliary_loss_mlp": 0.0100254,
      "balance_loss_clip": 1.02583635,
      "balance_loss_mlp": 1.00140762,
      "epoch": 0.31852341730295197,
      "flos": 64741948957440.0,
      "grad_norm": 0.7556654036943657,
      "language_loss": 0.56970489,
      "learning_rate": 3.188639971449956e-06,
      "loss": 0.59070253,
      "num_input_tokens_seen": 57074960,
      "step": 2649,
      "time_per_iteration": 3.1917014122009277
    },
    {
      "auxiliary_loss_clip": 0.01208223,
      "auxiliary_loss_mlp": 0.01035754,
      "balance_loss_clip": 1.0625751,
      "balance_loss_mlp": 1.02716517,
      "epoch": 0.318643660193591,
      "flos": 20668135582080.0,
      "grad_norm": 2.0433909391935483,
      "language_loss": 0.72406197,
      "learning_rate": 3.1880134099756e-06,
      "loss": 0.74650168,
      "num_input_tokens_seen": 57094595,
      "step": 2650,
      "time_per_iteration": 2.7830114364624023
    },
    {
      "auxiliary_loss_clip": 0.01251075,
      "auxiliary_loss_mlp": 0.01029821,
      "balance_loss_clip": 1.05611086,
      "balance_loss_mlp": 1.02210212,
      "epoch": 0.31876390308423014,
      "flos": 26943345411840.0,
      "grad_norm": 2.676414912668214,
      "language_loss": 0.69560033,
      "learning_rate": 3.1873866682805535e-06,
      "loss": 0.7184093,
      "num_input_tokens_seen": 57115290,
      "step": 2651,
      "time_per_iteration": 2.7680039405822754
    },
    {
      "auxiliary_loss_clip": 0.01212968,
      "auxiliary_loss_mlp": 0.01032902,
      "balance_loss_clip": 1.05739355,
      "balance_loss_mlp": 1.02449739,
      "epoch": 0.31888414597486925,
      "flos": 18041916597120.0,
      "grad_norm": 2.5013457451435017,
      "language_loss": 0.88686419,
      "learning_rate": 3.186759746459894e-06,
      "loss": 0.90932286,
      "num_input_tokens_seen": 57134400,
      "step": 2652,
      "time_per_iteration": 2.674445867538452
    },
    {
      "auxiliary_loss_clip": 0.01306501,
      "auxiliary_loss_mlp": 0.0103367,
      "balance_loss_clip": 1.05873799,
      "balance_loss_mlp": 1.02599287,
      "epoch": 0.3190043888655083,
      "flos": 25149319701120.0,
      "grad_norm": 2.2054625124492713,
      "language_loss": 0.79478741,
      "learning_rate": 3.1861326446087246e-06,
      "loss": 0.81818908,
      "num_input_tokens_seen": 57153140,
      "step": 2653,
      "time_per_iteration": 2.6881103515625
    },
    {
      "auxiliary_loss_clip": 0.01257402,
      "auxiliary_loss_mlp": 0.01031666,
      "balance_loss_clip": 1.05914283,
      "balance_loss_mlp": 1.0237149,
      "epoch": 0.3191246317561474,
      "flos": 22053892331520.0,
      "grad_norm": 2.0193220124165023,
      "language_loss": 0.71967405,
      "learning_rate": 3.1855053628221763e-06,
      "loss": 0.74256474,
      "num_input_tokens_seen": 57172395,
      "step": 2654,
      "time_per_iteration": 2.6911089420318604
    },
    {
      "auxiliary_loss_clip": 0.01345185,
      "auxiliary_loss_mlp": 0.0103353,
      "balance_loss_clip": 1.05012465,
      "balance_loss_mlp": 1.02523935,
      "epoch": 0.3192448746467865,
      "flos": 14901815687040.0,
      "grad_norm": 3.6443780678865227,
      "language_loss": 0.90298426,
      "learning_rate": 3.184877901195407e-06,
      "loss": 0.92677146,
      "num_input_tokens_seen": 57189090,
      "step": 2655,
      "time_per_iteration": 2.7021780014038086
    },
    {
      "auxiliary_loss_clip": 0.01223602,
      "auxiliary_loss_mlp": 0.01021292,
      "balance_loss_clip": 1.0370307,
      "balance_loss_mlp": 1.01974225,
      "epoch": 0.3193651175374256,
      "flos": 67234832657280.0,
      "grad_norm": 0.807462854414125,
      "language_loss": 0.62830091,
      "learning_rate": 3.184250259823602e-06,
      "loss": 0.6507498,
      "num_input_tokens_seen": 57251620,
      "step": 2656,
      "time_per_iteration": 3.3388729095458984
    },
    {
      "auxiliary_loss_clip": 0.01351347,
      "auxiliary_loss_mlp": 0.01036257,
      "balance_loss_clip": 1.05535078,
      "balance_loss_mlp": 1.0276804,
      "epoch": 0.3194853604280647,
      "flos": 12233077977600.0,
      "grad_norm": 2.2257390597809823,
      "language_loss": 0.81077009,
      "learning_rate": 3.183622438801974e-06,
      "loss": 0.83464617,
      "num_input_tokens_seen": 57266910,
      "step": 2657,
      "time_per_iteration": 2.7454466819763184
    },
    {
      "auxiliary_loss_clip": 0.01205387,
      "auxiliary_loss_mlp": 0.01030154,
      "balance_loss_clip": 1.06143773,
      "balance_loss_mlp": 1.02164245,
      "epoch": 0.3196056033187038,
      "flos": 14939917038720.0,
      "grad_norm": 2.0960073255799125,
      "language_loss": 0.75652945,
      "learning_rate": 3.1829944382257637e-06,
      "loss": 0.77888483,
      "num_input_tokens_seen": 57285040,
      "step": 2658,
      "time_per_iteration": 2.6610891819000244
    },
    {
      "auxiliary_loss_clip": 0.0124764,
      "auxiliary_loss_mlp": 0.01027519,
      "balance_loss_clip": 1.05850732,
      "balance_loss_mlp": 1.01935923,
      "epoch": 0.31972584620934286,
      "flos": 23768878164480.0,
      "grad_norm": 2.3456357389572813,
      "language_loss": 0.81617439,
      "learning_rate": 3.1823662581902373e-06,
      "loss": 0.83892596,
      "num_input_tokens_seen": 57302725,
      "step": 2659,
      "time_per_iteration": 2.6747899055480957
    },
    {
      "auxiliary_loss_clip": 0.0134123,
      "auxiliary_loss_mlp": 0.01033146,
      "balance_loss_clip": 1.04483783,
      "balance_loss_mlp": 1.02509332,
      "epoch": 0.31984608909998197,
      "flos": 21251540280960.0,
      "grad_norm": 2.533359844005296,
      "language_loss": 0.74684155,
      "learning_rate": 3.1817378987906896e-06,
      "loss": 0.7705853,
      "num_input_tokens_seen": 57322230,
      "step": 2660,
      "time_per_iteration": 2.7658181190490723
    },
    {
      "auxiliary_loss_clip": 0.0139937,
      "auxiliary_loss_mlp": 0.01038627,
      "balance_loss_clip": 1.05697954,
      "balance_loss_mlp": 1.0303185,
      "epoch": 0.3199663319906211,
      "flos": 18296235866880.0,
      "grad_norm": 2.197911236475607,
      "language_loss": 0.79761493,
      "learning_rate": 3.181109360122442e-06,
      "loss": 0.8219949,
      "num_input_tokens_seen": 57339820,
      "step": 2661,
      "time_per_iteration": 2.746469259262085
    },
    {
      "auxiliary_loss_clip": 0.01355573,
      "auxiliary_loss_mlp": 0.01034607,
      "balance_loss_clip": 1.05725932,
      "balance_loss_mlp": 1.02682257,
      "epoch": 0.32008657488126013,
      "flos": 18733627779840.0,
      "grad_norm": 2.4000800667214053,
      "language_loss": 0.78407705,
      "learning_rate": 3.1804806422808445e-06,
      "loss": 0.80797887,
      "num_input_tokens_seen": 57356955,
      "step": 2662,
      "time_per_iteration": 3.675178289413452
    },
    {
      "auxiliary_loss_clip": 0.01294723,
      "auxiliary_loss_mlp": 0.01029817,
      "balance_loss_clip": 1.05417264,
      "balance_loss_mlp": 1.02197611,
      "epoch": 0.32020681777189924,
      "flos": 20595344670720.0,
      "grad_norm": 1.6917754394645284,
      "language_loss": 0.72940731,
      "learning_rate": 3.1798517453612714e-06,
      "loss": 0.75265265,
      "num_input_tokens_seen": 57376760,
      "step": 2663,
      "time_per_iteration": 2.781967878341675
    },
    {
      "auxiliary_loss_clip": 0.0125556,
      "auxiliary_loss_mlp": 0.01031481,
      "balance_loss_clip": 1.06351423,
      "balance_loss_mlp": 1.02320182,
      "epoch": 0.32032706066253835,
      "flos": 35261692750080.0,
      "grad_norm": 1.888160123523702,
      "language_loss": 0.75253433,
      "learning_rate": 3.1792226694591265e-06,
      "loss": 0.77540469,
      "num_input_tokens_seen": 57398145,
      "step": 2664,
      "time_per_iteration": 3.6914520263671875
    },
    {
      "auxiliary_loss_clip": 0.01349098,
      "auxiliary_loss_mlp": 0.01032489,
      "balance_loss_clip": 1.05612767,
      "balance_loss_mlp": 1.02452612,
      "epoch": 0.3204473035531774,
      "flos": 15304230731520.0,
      "grad_norm": 2.320748410826593,
      "language_loss": 0.80468017,
      "learning_rate": 3.178593414669841e-06,
      "loss": 0.82849604,
      "num_input_tokens_seen": 57416730,
      "step": 2665,
      "time_per_iteration": 2.7475204467773438
    },
    {
      "auxiliary_loss_clip": 0.0126098,
      "auxiliary_loss_mlp": 0.01030925,
      "balance_loss_clip": 1.06090569,
      "balance_loss_mlp": 1.02199054,
      "epoch": 0.3205675464438165,
      "flos": 24462564595200.0,
      "grad_norm": 2.210410556760773,
      "language_loss": 0.70701647,
      "learning_rate": 3.1779639810888707e-06,
      "loss": 0.72993553,
      "num_input_tokens_seen": 57436325,
      "step": 2666,
      "time_per_iteration": 2.7112438678741455
    },
    {
      "auxiliary_loss_clip": 0.01249318,
      "auxiliary_loss_mlp": 0.01029794,
      "balance_loss_clip": 1.05921412,
      "balance_loss_mlp": 1.0215385,
      "epoch": 0.3206877893344556,
      "flos": 22456235548800.0,
      "grad_norm": 1.807298733297278,
      "language_loss": 0.761729,
      "learning_rate": 3.1773343688117013e-06,
      "loss": 0.78452009,
      "num_input_tokens_seen": 57457235,
      "step": 2667,
      "time_per_iteration": 2.6557557582855225
    },
    {
      "auxiliary_loss_clip": 0.01303419,
      "auxiliary_loss_mlp": 0.02569291,
      "balance_loss_clip": 1.05209088,
      "balance_loss_mlp": 1.00013876,
      "epoch": 0.3208080322250947,
      "flos": 20412236113920.0,
      "grad_norm": 2.169867215241575,
      "language_loss": 0.84351921,
      "learning_rate": 3.1767045779338445e-06,
      "loss": 0.88224632,
      "num_input_tokens_seen": 57474895,
      "step": 2668,
      "time_per_iteration": 2.7146055698394775
    },
    {
      "auxiliary_loss_clip": 0.01248969,
      "auxiliary_loss_mlp": 0.01029789,
      "balance_loss_clip": 1.05317998,
      "balance_loss_mlp": 1.02205276,
      "epoch": 0.3209282751157338,
      "flos": 21762118154880.0,
      "grad_norm": 2.02096144481064,
      "language_loss": 0.91592193,
      "learning_rate": 3.176074608550839e-06,
      "loss": 0.9387095,
      "num_input_tokens_seen": 57490715,
      "step": 2669,
      "time_per_iteration": 2.6530983448028564
    },
    {
      "auxiliary_loss_clip": 0.01447817,
      "auxiliary_loss_mlp": 0.01032737,
      "balance_loss_clip": 1.05010104,
      "balance_loss_mlp": 1.02421355,
      "epoch": 0.32104851800637285,
      "flos": 22055041566720.0,
      "grad_norm": 3.041283556750985,
      "language_loss": 0.82518566,
      "learning_rate": 3.17544446075825e-06,
      "loss": 0.8499912,
      "num_input_tokens_seen": 57509880,
      "step": 2670,
      "time_per_iteration": 3.676870822906494
    },
    {
      "auxiliary_loss_clip": 0.01307251,
      "auxiliary_loss_mlp": 0.01027586,
      "balance_loss_clip": 1.0570029,
      "balance_loss_mlp": 1.0198853,
      "epoch": 0.32116876089701196,
      "flos": 37012301896320.0,
      "grad_norm": 1.6975284600865148,
      "language_loss": 0.71061444,
      "learning_rate": 3.174814134651671e-06,
      "loss": 0.73396277,
      "num_input_tokens_seen": 57532430,
      "step": 2671,
      "time_per_iteration": 2.827807664871216
    },
    {
      "auxiliary_loss_clip": 0.01199361,
      "auxiliary_loss_mlp": 0.01032235,
      "balance_loss_clip": 1.05888462,
      "balance_loss_mlp": 1.02488589,
      "epoch": 0.3212890037876511,
      "flos": 21979233912960.0,
      "grad_norm": 2.055195090173691,
      "language_loss": 0.80585253,
      "learning_rate": 3.1741836303267215e-06,
      "loss": 0.82816851,
      "num_input_tokens_seen": 57551965,
      "step": 2672,
      "time_per_iteration": 3.5749707221984863
    },
    {
      "auxiliary_loss_clip": 0.01197167,
      "auxiliary_loss_mlp": 0.01031301,
      "balance_loss_clip": 1.05715752,
      "balance_loss_mlp": 1.02288449,
      "epoch": 0.32140924667829013,
      "flos": 10342345875840.0,
      "grad_norm": 2.0947846991571515,
      "language_loss": 0.75601339,
      "learning_rate": 3.1735529478790496e-06,
      "loss": 0.77829814,
      "num_input_tokens_seen": 57569955,
      "step": 2673,
      "time_per_iteration": 2.7049672603607178
    },
    {
      "auxiliary_loss_clip": 0.01249699,
      "auxiliary_loss_mlp": 0.01027326,
      "balance_loss_clip": 1.05672681,
      "balance_loss_mlp": 1.01767027,
      "epoch": 0.32152948956892924,
      "flos": 50798910072960.0,
      "grad_norm": 1.8860422552764113,
      "language_loss": 0.79278916,
      "learning_rate": 3.172922087404328e-06,
      "loss": 0.81555939,
      "num_input_tokens_seen": 57592215,
      "step": 2674,
      "time_per_iteration": 2.9103660583496094
    },
    {
      "auxiliary_loss_clip": 0.01096692,
      "auxiliary_loss_mlp": 0.01006167,
      "balance_loss_clip": 1.02585328,
      "balance_loss_mlp": 1.00502861,
      "epoch": 0.32164973245956835,
      "flos": 63863250549120.0,
      "grad_norm": 0.9004829041994183,
      "language_loss": 0.5522089,
      "learning_rate": 3.1722910489982586e-06,
      "loss": 0.57323754,
      "num_input_tokens_seen": 57652575,
      "step": 2675,
      "time_per_iteration": 3.2663371562957764
    },
    {
      "auxiliary_loss_clip": 0.01300003,
      "auxiliary_loss_mlp": 0.01031482,
      "balance_loss_clip": 1.0542351,
      "balance_loss_mlp": 1.02257133,
      "epoch": 0.3217699753502074,
      "flos": 23513948363520.0,
      "grad_norm": 1.6462135902519155,
      "language_loss": 0.79856074,
      "learning_rate": 3.1716598327565694e-06,
      "loss": 0.82187551,
      "num_input_tokens_seen": 57672215,
      "step": 2676,
      "time_per_iteration": 2.708505630493164
    },
    {
      "auxiliary_loss_clip": 0.01200327,
      "auxiliary_loss_mlp": 0.01030654,
      "balance_loss_clip": 1.05839396,
      "balance_loss_mlp": 1.02274477,
      "epoch": 0.3218902182408465,
      "flos": 19062533640960.0,
      "grad_norm": 1.6521142600906313,
      "language_loss": 0.84207749,
      "learning_rate": 3.171028438775015e-06,
      "loss": 0.86438733,
      "num_input_tokens_seen": 57691410,
      "step": 2677,
      "time_per_iteration": 2.6117968559265137
    },
    {
      "auxiliary_loss_clip": 0.01199121,
      "auxiliary_loss_mlp": 0.01035322,
      "balance_loss_clip": 1.05804706,
      "balance_loss_mlp": 1.02738857,
      "epoch": 0.3220104611314856,
      "flos": 20375571306240.0,
      "grad_norm": 2.533823683933743,
      "language_loss": 0.84315681,
      "learning_rate": 3.170396867149377e-06,
      "loss": 0.86550122,
      "num_input_tokens_seen": 57709415,
      "step": 2678,
      "time_per_iteration": 2.599835157394409
    },
    {
      "auxiliary_loss_clip": 0.01399348,
      "auxiliary_loss_mlp": 0.01036028,
      "balance_loss_clip": 1.05526435,
      "balance_loss_mlp": 1.02689099,
      "epoch": 0.3221307040221247,
      "flos": 20117014231680.0,
      "grad_norm": 1.928019297885561,
      "language_loss": 0.86072004,
      "learning_rate": 3.1697651179754653e-06,
      "loss": 0.88507378,
      "num_input_tokens_seen": 57728075,
      "step": 2679,
      "time_per_iteration": 2.762666940689087
    },
    {
      "auxiliary_loss_clip": 0.01356509,
      "auxiliary_loss_mlp": 0.01035434,
      "balance_loss_clip": 1.06065023,
      "balance_loss_mlp": 1.02793241,
      "epoch": 0.3222509469127638,
      "flos": 23987789602560.0,
      "grad_norm": 1.8598915589200222,
      "language_loss": 0.72788143,
      "learning_rate": 3.1691331913491153e-06,
      "loss": 0.75180089,
      "num_input_tokens_seen": 57750645,
      "step": 2680,
      "time_per_iteration": 2.8999311923980713
    },
    {
      "auxiliary_loss_clip": 0.01201075,
      "auxiliary_loss_mlp": 0.01027191,
      "balance_loss_clip": 1.05786967,
      "balance_loss_mlp": 1.01919782,
      "epoch": 0.32237118980340285,
      "flos": 17675735397120.0,
      "grad_norm": 11.200623426292253,
      "language_loss": 0.85053432,
      "learning_rate": 3.1685010873661898e-06,
      "loss": 0.87281698,
      "num_input_tokens_seen": 57769820,
      "step": 2681,
      "time_per_iteration": 2.606022357940674
    },
    {
      "auxiliary_loss_clip": 0.01247704,
      "auxiliary_loss_mlp": 0.01028926,
      "balance_loss_clip": 1.0533576,
      "balance_loss_mlp": 1.01999104,
      "epoch": 0.32249143269404196,
      "flos": 23147982645120.0,
      "grad_norm": 2.0127288814647577,
      "language_loss": 0.79814029,
      "learning_rate": 3.167868806122578e-06,
      "loss": 0.82090664,
      "num_input_tokens_seen": 57788870,
      "step": 2682,
      "time_per_iteration": 2.6618988513946533
    },
    {
      "auxiliary_loss_clip": 0.01301077,
      "auxiliary_loss_mlp": 0.0103391,
      "balance_loss_clip": 1.05601668,
      "balance_loss_mlp": 1.02511191,
      "epoch": 0.32261167558468107,
      "flos": 24422308427520.0,
      "grad_norm": 2.7006781681756604,
      "language_loss": 0.66573679,
      "learning_rate": 3.1672363477141968e-06,
      "loss": 0.68908668,
      "num_input_tokens_seen": 57808165,
      "step": 2683,
      "time_per_iteration": 2.839826822280884
    },
    {
      "auxiliary_loss_clip": 0.01213098,
      "auxiliary_loss_mlp": 0.01035442,
      "balance_loss_clip": 1.05555391,
      "balance_loss_mlp": 1.02612031,
      "epoch": 0.3227319184753201,
      "flos": 30367175852160.0,
      "grad_norm": 3.659116136755836,
      "language_loss": 0.85059512,
      "learning_rate": 3.1666037122369903e-06,
      "loss": 0.87308049,
      "num_input_tokens_seen": 57828825,
      "step": 2684,
      "time_per_iteration": 2.79789400100708
    },
    {
      "auxiliary_loss_clip": 0.01245079,
      "auxiliary_loss_mlp": 0.0103479,
      "balance_loss_clip": 1.05150712,
      "balance_loss_mlp": 1.02622509,
      "epoch": 0.32285216136595923,
      "flos": 16946174257920.0,
      "grad_norm": 2.7499796876889984,
      "language_loss": 0.87176812,
      "learning_rate": 3.165970899786928e-06,
      "loss": 0.89456677,
      "num_input_tokens_seen": 57846740,
      "step": 2685,
      "time_per_iteration": 2.673337697982788
    },
    {
      "auxiliary_loss_clip": 0.01361558,
      "auxiliary_loss_mlp": 0.01037689,
      "balance_loss_clip": 1.0579257,
      "balance_loss_mlp": 1.02918923,
      "epoch": 0.32297240425659834,
      "flos": 21981532383360.0,
      "grad_norm": 1.8335723795070467,
      "language_loss": 0.75211835,
      "learning_rate": 3.1653379104600067e-06,
      "loss": 0.77611077,
      "num_input_tokens_seen": 57866885,
      "step": 2686,
      "time_per_iteration": 2.743190288543701
    },
    {
      "auxiliary_loss_clip": 0.01251255,
      "auxiliary_loss_mlp": 0.0104586,
      "balance_loss_clip": 1.05506945,
      "balance_loss_mlp": 1.03787541,
      "epoch": 0.3230926471472374,
      "flos": 22748045639040.0,
      "grad_norm": 1.8390135993318508,
      "language_loss": 0.69537568,
      "learning_rate": 3.164704744352251e-06,
      "loss": 0.71834677,
      "num_input_tokens_seen": 57887690,
      "step": 2687,
      "time_per_iteration": 2.9319770336151123
    },
    {
      "auxiliary_loss_clip": 0.01246461,
      "auxiliary_loss_mlp": 0.01031302,
      "balance_loss_clip": 1.05719376,
      "balance_loss_mlp": 1.02328491,
      "epoch": 0.3232128900378765,
      "flos": 16942977947520.0,
      "grad_norm": 1.6668472030021257,
      "language_loss": 0.81029308,
      "learning_rate": 3.164071401559713e-06,
      "loss": 0.8330707,
      "num_input_tokens_seen": 57905090,
      "step": 2688,
      "time_per_iteration": 3.5935428142547607
    },
    {
      "auxiliary_loss_clip": 0.01300622,
      "auxiliary_loss_mlp": 0.01034603,
      "balance_loss_clip": 1.05454493,
      "balance_loss_mlp": 1.02652621,
      "epoch": 0.3233331329285156,
      "flos": 24023736138240.0,
      "grad_norm": 1.726403617664475,
      "language_loss": 0.70973527,
      "learning_rate": 3.1634378821784674e-06,
      "loss": 0.73308754,
      "num_input_tokens_seen": 57925305,
      "step": 2689,
      "time_per_iteration": 2.769500970840454
    },
    {
      "auxiliary_loss_clip": 0.01267223,
      "auxiliary_loss_mlp": 0.01035785,
      "balance_loss_clip": 1.06019187,
      "balance_loss_mlp": 1.02743423,
      "epoch": 0.3234533758191547,
      "flos": 18113845582080.0,
      "grad_norm": 3.741592053632237,
      "language_loss": 0.7437495,
      "learning_rate": 3.1628041863046208e-06,
      "loss": 0.7667796,
      "num_input_tokens_seen": 57942720,
      "step": 2690,
      "time_per_iteration": 3.652128219604492
    },
    {
      "auxiliary_loss_clip": 0.01205832,
      "auxiliary_loss_mlp": 0.01035016,
      "balance_loss_clip": 1.05795693,
      "balance_loss_mlp": 1.02627826,
      "epoch": 0.3235736187097938,
      "flos": 16946138344320.0,
      "grad_norm": 2.2730975824057853,
      "language_loss": 0.91826642,
      "learning_rate": 3.162170314034304e-06,
      "loss": 0.94067496,
      "num_input_tokens_seen": 57960135,
      "step": 2691,
      "time_per_iteration": 2.555877923965454
    },
    {
      "auxiliary_loss_clip": 0.01206944,
      "auxiliary_loss_mlp": 0.01033296,
      "balance_loss_clip": 1.06004953,
      "balance_loss_mlp": 1.02464187,
      "epoch": 0.3236938616004329,
      "flos": 22127150119680.0,
      "grad_norm": 1.728162886965683,
      "language_loss": 0.81147045,
      "learning_rate": 3.1615362654636738e-06,
      "loss": 0.8338728,
      "num_input_tokens_seen": 57980875,
      "step": 2692,
      "time_per_iteration": 2.6387429237365723
    },
    {
      "auxiliary_loss_clip": 0.01347572,
      "auxiliary_loss_mlp": 0.01028808,
      "balance_loss_clip": 1.05959427,
      "balance_loss_mlp": 1.02078557,
      "epoch": 0.32381410449107195,
      "flos": 17164618819200.0,
      "grad_norm": 1.804482876654117,
      "language_loss": 0.8752107,
      "learning_rate": 3.1609020406889163e-06,
      "loss": 0.89897454,
      "num_input_tokens_seen": 57998310,
      "step": 2693,
      "time_per_iteration": 2.648016929626465
    },
    {
      "auxiliary_loss_clip": 0.01302934,
      "auxiliary_loss_mlp": 0.01031595,
      "balance_loss_clip": 1.05276978,
      "balance_loss_mlp": 1.02321434,
      "epoch": 0.32393434738171106,
      "flos": 16578125550720.0,
      "grad_norm": 1.825475608215507,
      "language_loss": 0.85039198,
      "learning_rate": 3.1602676398062416e-06,
      "loss": 0.87373728,
      "num_input_tokens_seen": 58017220,
      "step": 2694,
      "time_per_iteration": 2.720710277557373
    },
    {
      "auxiliary_loss_clip": 0.01249433,
      "auxiliary_loss_mlp": 0.0102933,
      "balance_loss_clip": 1.05765831,
      "balance_loss_mlp": 1.02099144,
      "epoch": 0.3240545902723502,
      "flos": 25483612602240.0,
      "grad_norm": 3.255465086246207,
      "language_loss": 0.61510026,
      "learning_rate": 3.1596330629118886e-06,
      "loss": 0.6378879,
      "num_input_tokens_seen": 58037190,
      "step": 2695,
      "time_per_iteration": 2.662487268447876
    },
    {
      "auxiliary_loss_clip": 0.01395656,
      "auxiliary_loss_mlp": 0.01042929,
      "balance_loss_clip": 1.05285788,
      "balance_loss_mlp": 1.03470945,
      "epoch": 0.32417483316298923,
      "flos": 35845851634560.0,
      "grad_norm": 2.0221531881941455,
      "language_loss": 0.73507369,
      "learning_rate": 3.1589983101021223e-06,
      "loss": 0.75945956,
      "num_input_tokens_seen": 58055820,
      "step": 2696,
      "time_per_iteration": 3.7396678924560547
    },
    {
      "auxiliary_loss_clip": 0.0120652,
      "auxiliary_loss_mlp": 0.01030172,
      "balance_loss_clip": 1.05406713,
      "balance_loss_mlp": 1.02245903,
      "epoch": 0.32429507605362834,
      "flos": 30080501406720.0,
      "grad_norm": 2.1726652987806068,
      "language_loss": 0.84681112,
      "learning_rate": 3.1583633814732337e-06,
      "loss": 0.86917794,
      "num_input_tokens_seen": 58075340,
      "step": 2697,
      "time_per_iteration": 2.752586603164673
    },
    {
      "auxiliary_loss_clip": 0.0120156,
      "auxiliary_loss_mlp": 0.01035749,
      "balance_loss_clip": 1.05761719,
      "balance_loss_mlp": 1.02849209,
      "epoch": 0.3244153189442674,
      "flos": 18223265387520.0,
      "grad_norm": 2.4564126277819938,
      "language_loss": 0.72262216,
      "learning_rate": 3.157728277121541e-06,
      "loss": 0.74499524,
      "num_input_tokens_seen": 58093515,
      "step": 2698,
      "time_per_iteration": 3.486793279647827
    },
    {
      "auxiliary_loss_clip": 0.0119903,
      "auxiliary_loss_mlp": 0.01031247,
      "balance_loss_clip": 1.05582964,
      "balance_loss_mlp": 1.02266407,
      "epoch": 0.3245355618349065,
      "flos": 17710317216000.0,
      "grad_norm": 2.6760876769634763,
      "language_loss": 0.78489769,
      "learning_rate": 3.1570929971433897e-06,
      "loss": 0.80720043,
      "num_input_tokens_seen": 58109300,
      "step": 2699,
      "time_per_iteration": 2.624271869659424
    },
    {
      "auxiliary_loss_clip": 0.01250472,
      "auxiliary_loss_mlp": 0.0103615,
      "balance_loss_clip": 1.06001472,
      "balance_loss_mlp": 1.02862203,
      "epoch": 0.3246558047255456,
      "flos": 23440798316160.0,
      "grad_norm": 2.090111841393296,
      "language_loss": 0.83882022,
      "learning_rate": 3.1564575416351504e-06,
      "loss": 0.86168635,
      "num_input_tokens_seen": 58128000,
      "step": 2700,
      "time_per_iteration": 2.640345573425293
    },
    {
      "auxiliary_loss_clip": 0.01201705,
      "auxiliary_loss_mlp": 0.01033833,
      "balance_loss_clip": 1.05930185,
      "balance_loss_mlp": 1.02515411,
      "epoch": 0.32477604761618467,
      "flos": 21760861178880.0,
      "grad_norm": 2.528403943366054,
      "language_loss": 0.74045771,
      "learning_rate": 3.155821910693221e-06,
      "loss": 0.76281303,
      "num_input_tokens_seen": 58147415,
      "step": 2701,
      "time_per_iteration": 2.666980743408203
    },
    {
      "auxiliary_loss_clip": 0.01297906,
      "auxiliary_loss_mlp": 0.01027461,
      "balance_loss_clip": 1.05345881,
      "balance_loss_mlp": 1.01946783,
      "epoch": 0.3248962905068238,
      "flos": 19828328624640.0,
      "grad_norm": 1.753813393330224,
      "language_loss": 0.85898125,
      "learning_rate": 3.1551861044140275e-06,
      "loss": 0.88223493,
      "num_input_tokens_seen": 58167050,
      "step": 2702,
      "time_per_iteration": 2.6877663135528564
    },
    {
      "auxiliary_loss_clip": 0.01394427,
      "auxiliary_loss_mlp": 0.01032623,
      "balance_loss_clip": 1.05441284,
      "balance_loss_mlp": 1.0250504,
      "epoch": 0.3250165333974629,
      "flos": 23948215793280.0,
      "grad_norm": 1.9341699088037936,
      "language_loss": 0.77689505,
      "learning_rate": 3.15455012289402e-06,
      "loss": 0.80116558,
      "num_input_tokens_seen": 58186695,
      "step": 2703,
      "time_per_iteration": 2.7777981758117676
    },
    {
      "auxiliary_loss_clip": 0.01253299,
      "auxiliary_loss_mlp": 0.01023574,
      "balance_loss_clip": 1.06148016,
      "balance_loss_mlp": 1.01553345,
      "epoch": 0.32513677628810195,
      "flos": 23989333887360.0,
      "grad_norm": 2.1913580478920602,
      "language_loss": 0.84535629,
      "learning_rate": 3.153913966229677e-06,
      "loss": 0.86812502,
      "num_input_tokens_seen": 58205815,
      "step": 2704,
      "time_per_iteration": 2.6978988647460938
    },
    {
      "auxiliary_loss_clip": 0.01151295,
      "auxiliary_loss_mlp": 0.01002636,
      "balance_loss_clip": 1.02740276,
      "balance_loss_mlp": 1.00143778,
      "epoch": 0.32525701917874106,
      "flos": 70655790009600.0,
      "grad_norm": 0.6492005053943973,
      "language_loss": 0.50248885,
      "learning_rate": 3.1532776345175027e-06,
      "loss": 0.52402818,
      "num_input_tokens_seen": 58270960,
      "step": 2705,
      "time_per_iteration": 3.224557399749756
    },
    {
      "auxiliary_loss_clip": 0.01199718,
      "auxiliary_loss_mlp": 0.01028102,
      "balance_loss_clip": 1.05813551,
      "balance_loss_mlp": 1.01998663,
      "epoch": 0.32537726206938017,
      "flos": 19682639061120.0,
      "grad_norm": 2.036504734461978,
      "language_loss": 0.78612828,
      "learning_rate": 3.1526411278540285e-06,
      "loss": 0.80840647,
      "num_input_tokens_seen": 58289390,
      "step": 2706,
      "time_per_iteration": 2.6768252849578857
    },
    {
      "auxiliary_loss_clip": 0.01303388,
      "auxiliary_loss_mlp": 0.01025141,
      "balance_loss_clip": 1.05163574,
      "balance_loss_mlp": 1.01721406,
      "epoch": 0.3254975049600192,
      "flos": 28760999293440.0,
      "grad_norm": 2.64655289777107,
      "language_loss": 0.81937814,
      "learning_rate": 3.1520044463358116e-06,
      "loss": 0.84266341,
      "num_input_tokens_seen": 58306120,
      "step": 2707,
      "time_per_iteration": 2.8092055320739746
    },
    {
      "auxiliary_loss_clip": 0.01248236,
      "auxiliary_loss_mlp": 0.01027983,
      "balance_loss_clip": 1.05809903,
      "balance_loss_mlp": 1.02028537,
      "epoch": 0.32561774785065833,
      "flos": 18877378008960.0,
      "grad_norm": 1.7526615119352826,
      "language_loss": 0.80299735,
      "learning_rate": 3.151367590059436e-06,
      "loss": 0.82575953,
      "num_input_tokens_seen": 58324545,
      "step": 2708,
      "time_per_iteration": 2.5971014499664307
    },
    {
      "auxiliary_loss_clip": 0.01201646,
      "auxiliary_loss_mlp": 0.02570519,
      "balance_loss_clip": 1.05873036,
      "balance_loss_mlp": 1.00020051,
      "epoch": 0.32573799074129745,
      "flos": 23112107936640.0,
      "grad_norm": 2.120228681362872,
      "language_loss": 0.86753309,
      "learning_rate": 3.1507305591215117e-06,
      "loss": 0.90525472,
      "num_input_tokens_seen": 58342455,
      "step": 2709,
      "time_per_iteration": 2.6127495765686035
    },
    {
      "auxiliary_loss_clip": 0.01150317,
      "auxiliary_loss_mlp": 0.01002556,
      "balance_loss_clip": 1.02636123,
      "balance_loss_mlp": 1.00135231,
      "epoch": 0.3258582336319365,
      "flos": 71237650423680.0,
      "grad_norm": 0.6693941908245172,
      "language_loss": 0.55737078,
      "learning_rate": 3.150093353618677e-06,
      "loss": 0.5788995,
      "num_input_tokens_seen": 58407185,
      "step": 2710,
      "time_per_iteration": 3.2743947505950928
    },
    {
      "auxiliary_loss_clip": 0.01258712,
      "auxiliary_loss_mlp": 0.01033001,
      "balance_loss_clip": 1.05826187,
      "balance_loss_mlp": 1.02491879,
      "epoch": 0.3259784765225756,
      "flos": 22456020067200.0,
      "grad_norm": 2.3486047909484538,
      "language_loss": 0.88244325,
      "learning_rate": 3.149455973647596e-06,
      "loss": 0.90536034,
      "num_input_tokens_seen": 58425245,
      "step": 2711,
      "time_per_iteration": 2.6772823333740234
    },
    {
      "auxiliary_loss_clip": 0.01343333,
      "auxiliary_loss_mlp": 0.01035501,
      "balance_loss_clip": 1.04941642,
      "balance_loss_mlp": 1.02713227,
      "epoch": 0.32609871941321467,
      "flos": 20484811543680.0,
      "grad_norm": 2.2887707533302994,
      "language_loss": 0.77255648,
      "learning_rate": 3.1488184193049563e-06,
      "loss": 0.79634482,
      "num_input_tokens_seen": 58444780,
      "step": 2712,
      "time_per_iteration": 2.7534680366516113
    },
    {
      "auxiliary_loss_clip": 0.0120118,
      "auxiliary_loss_mlp": 0.01030221,
      "balance_loss_clip": 1.06007075,
      "balance_loss_mlp": 1.02250218,
      "epoch": 0.3262189623038538,
      "flos": 22416805393920.0,
      "grad_norm": 1.8789473669912855,
      "language_loss": 0.7227661,
      "learning_rate": 3.1481806906874767e-06,
      "loss": 0.74508011,
      "num_input_tokens_seen": 58466090,
      "step": 2713,
      "time_per_iteration": 2.6472201347351074
    },
    {
      "auxiliary_loss_clip": 0.0120073,
      "auxiliary_loss_mlp": 0.01035955,
      "balance_loss_clip": 1.05796003,
      "balance_loss_mlp": 1.02846289,
      "epoch": 0.3263392051944929,
      "flos": 20923496346240.0,
      "grad_norm": 1.5646970557830702,
      "language_loss": 0.87960982,
      "learning_rate": 3.147542787891899e-06,
      "loss": 0.9019767,
      "num_input_tokens_seen": 58485435,
      "step": 2714,
      "time_per_iteration": 3.5429935455322266
    },
    {
      "auxiliary_loss_clip": 0.01216298,
      "auxiliary_loss_mlp": 0.01030018,
      "balance_loss_clip": 1.06032062,
      "balance_loss_mlp": 1.02203119,
      "epoch": 0.32645944808513194,
      "flos": 24025172682240.0,
      "grad_norm": 2.2223682904386557,
      "language_loss": 0.75253254,
      "learning_rate": 3.1469047110149926e-06,
      "loss": 0.77499568,
      "num_input_tokens_seen": 58504175,
      "step": 2715,
      "time_per_iteration": 2.7069613933563232
    },
    {
      "auxiliary_loss_clip": 0.01388833,
      "auxiliary_loss_mlp": 0.01033122,
      "balance_loss_clip": 1.05148554,
      "balance_loss_mlp": 1.02455711,
      "epoch": 0.32657969097577105,
      "flos": 21032413361280.0,
      "grad_norm": 1.9303086890530632,
      "language_loss": 0.84997845,
      "learning_rate": 3.146266460153554e-06,
      "loss": 0.87419796,
      "num_input_tokens_seen": 58523885,
      "step": 2716,
      "time_per_iteration": 3.6838464736938477
    },
    {
      "auxiliary_loss_clip": 0.01296485,
      "auxiliary_loss_mlp": 0.02572833,
      "balance_loss_clip": 1.05482388,
      "balance_loss_mlp": 1.00006914,
      "epoch": 0.32669993386641016,
      "flos": 22710267509760.0,
      "grad_norm": 1.72140840737589,
      "language_loss": 0.80046725,
      "learning_rate": 3.145628035404404e-06,
      "loss": 0.83916044,
      "num_input_tokens_seen": 58543085,
      "step": 2717,
      "time_per_iteration": 2.6835856437683105
    },
    {
      "auxiliary_loss_clip": 0.01145539,
      "auxiliary_loss_mlp": 0.00999502,
      "balance_loss_clip": 1.02383518,
      "balance_loss_mlp": 0.99838156,
      "epoch": 0.3268201767570492,
      "flos": 72105718406400.0,
      "grad_norm": 0.8839941635247627,
      "language_loss": 0.57369941,
      "learning_rate": 3.1449894368643922e-06,
      "loss": 0.59514987,
      "num_input_tokens_seen": 58605400,
      "step": 2718,
      "time_per_iteration": 3.3026413917541504
    },
    {
      "auxiliary_loss_clip": 0.01349342,
      "auxiliary_loss_mlp": 0.0103125,
      "balance_loss_clip": 1.05751276,
      "balance_loss_mlp": 1.02365041,
      "epoch": 0.32694041964768833,
      "flos": 24535175938560.0,
      "grad_norm": 1.47649415651115,
      "language_loss": 0.7143749,
      "learning_rate": 3.1443506646303934e-06,
      "loss": 0.73818076,
      "num_input_tokens_seen": 58626700,
      "step": 2719,
      "time_per_iteration": 2.751838445663452
    },
    {
      "auxiliary_loss_clip": 0.01252996,
      "auxiliary_loss_mlp": 0.01029673,
      "balance_loss_clip": 1.05713952,
      "balance_loss_mlp": 1.02142406,
      "epoch": 0.32706066253832744,
      "flos": 33183003755520.0,
      "grad_norm": 1.8550433059288485,
      "language_loss": 0.66699737,
      "learning_rate": 3.1437117187993086e-06,
      "loss": 0.68982404,
      "num_input_tokens_seen": 58649020,
      "step": 2720,
      "time_per_iteration": 2.7447216510772705
    },
    {
      "auxiliary_loss_clip": 0.01344656,
      "auxiliary_loss_mlp": 0.01034246,
      "balance_loss_clip": 1.05015028,
      "balance_loss_mlp": 1.02615213,
      "epoch": 0.3271809054289665,
      "flos": 24061622008320.0,
      "grad_norm": 1.839564392190203,
      "language_loss": 0.80000794,
      "learning_rate": 3.143072599468065e-06,
      "loss": 0.82379699,
      "num_input_tokens_seen": 58668845,
      "step": 2721,
      "time_per_iteration": 2.731677293777466
    },
    {
      "auxiliary_loss_clip": 0.01300137,
      "auxiliary_loss_mlp": 0.01032463,
      "balance_loss_clip": 1.05931759,
      "balance_loss_mlp": 1.02476811,
      "epoch": 0.3273011483196056,
      "flos": 38253769712640.0,
      "grad_norm": 1.5821147848385706,
      "language_loss": 0.75952768,
      "learning_rate": 3.1424333067336174e-06,
      "loss": 0.78285366,
      "num_input_tokens_seen": 58691610,
      "step": 2722,
      "time_per_iteration": 3.6843934059143066
    },
    {
      "auxiliary_loss_clip": 0.01253598,
      "auxiliary_loss_mlp": 0.01032249,
      "balance_loss_clip": 1.05627835,
      "balance_loss_mlp": 1.02404761,
      "epoch": 0.3274213912102447,
      "flos": 29054389582080.0,
      "grad_norm": 1.8030696304466916,
      "language_loss": 0.78501964,
      "learning_rate": 3.141793840692945e-06,
      "loss": 0.80787814,
      "num_input_tokens_seen": 58712360,
      "step": 2723,
      "time_per_iteration": 2.728302478790283
    },
    {
      "auxiliary_loss_clip": 0.01294341,
      "auxiliary_loss_mlp": 0.01030307,
      "balance_loss_clip": 1.05418205,
      "balance_loss_mlp": 1.02168798,
      "epoch": 0.32754163410088377,
      "flos": 29133249891840.0,
      "grad_norm": 2.318011180572133,
      "language_loss": 0.62029028,
      "learning_rate": 3.1411542014430553e-06,
      "loss": 0.64353669,
      "num_input_tokens_seen": 58733440,
      "step": 2724,
      "time_per_iteration": 3.6193501949310303
    },
    {
      "auxiliary_loss_clip": 0.01348491,
      "auxiliary_loss_mlp": 0.01032972,
      "balance_loss_clip": 1.05036318,
      "balance_loss_mlp": 1.02588248,
      "epoch": 0.3276618769915229,
      "flos": 20631075724800.0,
      "grad_norm": 2.016418904797021,
      "language_loss": 0.8142038,
      "learning_rate": 3.1405143890809804e-06,
      "loss": 0.83801848,
      "num_input_tokens_seen": 58752735,
      "step": 2725,
      "time_per_iteration": 2.737915515899658
    },
    {
      "auxiliary_loss_clip": 0.0129545,
      "auxiliary_loss_mlp": 0.01027927,
      "balance_loss_clip": 1.05452204,
      "balance_loss_mlp": 1.02057219,
      "epoch": 0.327782119882162,
      "flos": 18657425076480.0,
      "grad_norm": 1.9383053045172085,
      "language_loss": 0.70025009,
      "learning_rate": 3.1398744037037796e-06,
      "loss": 0.72348392,
      "num_input_tokens_seen": 58772070,
      "step": 2726,
      "time_per_iteration": 2.769911050796509
    },
    {
      "auxiliary_loss_clip": 0.0129645,
      "auxiliary_loss_mlp": 0.01031796,
      "balance_loss_clip": 1.05591059,
      "balance_loss_mlp": 1.02369583,
      "epoch": 0.32790236277280105,
      "flos": 21795802133760.0,
      "grad_norm": 2.1700932790289866,
      "language_loss": 0.84379101,
      "learning_rate": 3.139234245408538e-06,
      "loss": 0.86707342,
      "num_input_tokens_seen": 58790950,
      "step": 2727,
      "time_per_iteration": 2.6687350273132324
    },
    {
      "auxiliary_loss_clip": 0.0134569,
      "auxiliary_loss_mlp": 0.0257051,
      "balance_loss_clip": 1.05394328,
      "balance_loss_mlp": 1.00012708,
      "epoch": 0.32802260566344016,
      "flos": 23331414424320.0,
      "grad_norm": 1.9607053788323388,
      "language_loss": 0.76161921,
      "learning_rate": 3.1385939142923666e-06,
      "loss": 0.80078125,
      "num_input_tokens_seen": 58813340,
      "step": 2728,
      "time_per_iteration": 2.8452231884002686
    },
    {
      "auxiliary_loss_clip": 0.01298003,
      "auxiliary_loss_mlp": 0.01035327,
      "balance_loss_clip": 1.05303669,
      "balance_loss_mlp": 1.02651691,
      "epoch": 0.3281428485540792,
      "flos": 24206988349440.0,
      "grad_norm": 2.464837611701616,
      "language_loss": 0.78399169,
      "learning_rate": 3.137953410452405e-06,
      "loss": 0.80732495,
      "num_input_tokens_seen": 58833610,
      "step": 2729,
      "time_per_iteration": 2.687645673751831
    },
    {
      "auxiliary_loss_clip": 0.012965,
      "auxiliary_loss_mlp": 0.0103235,
      "balance_loss_clip": 1.05385065,
      "balance_loss_mlp": 1.02448785,
      "epoch": 0.3282630914447183,
      "flos": 34128962380800.0,
      "grad_norm": 1.8075910602192178,
      "language_loss": 0.74741364,
      "learning_rate": 3.1373127339858146e-06,
      "loss": 0.77070206,
      "num_input_tokens_seen": 58856210,
      "step": 2730,
      "time_per_iteration": 2.7853593826293945
    },
    {
      "auxiliary_loss_clip": 0.01345884,
      "auxiliary_loss_mlp": 0.01034032,
      "balance_loss_clip": 1.05258679,
      "balance_loss_mlp": 1.02596176,
      "epoch": 0.32838333433535744,
      "flos": 27600726170880.0,
      "grad_norm": 1.9594304689114797,
      "language_loss": 0.7499603,
      "learning_rate": 3.136671884989787e-06,
      "loss": 0.77375948,
      "num_input_tokens_seen": 58876120,
      "step": 2731,
      "time_per_iteration": 2.769310235977173
    },
    {
      "auxiliary_loss_clip": 0.01351986,
      "auxiliary_loss_mlp": 0.0103396,
      "balance_loss_clip": 1.04738784,
      "balance_loss_mlp": 1.02468538,
      "epoch": 0.3285035772259965,
      "flos": 12349500935040.0,
      "grad_norm": 2.2160594775492095,
      "language_loss": 0.86412191,
      "learning_rate": 3.1360308635615383e-06,
      "loss": 0.88798136,
      "num_input_tokens_seen": 58894660,
      "step": 2732,
      "time_per_iteration": 2.769449234008789
    },
    {
      "auxiliary_loss_clip": 0.013077,
      "auxiliary_loss_mlp": 0.01035985,
      "balance_loss_clip": 1.0573976,
      "balance_loss_mlp": 1.02685928,
      "epoch": 0.3286238201166356,
      "flos": 24316084932480.0,
      "grad_norm": 2.0447630394741596,
      "language_loss": 0.7873773,
      "learning_rate": 3.135389669798311e-06,
      "loss": 0.81081414,
      "num_input_tokens_seen": 58912720,
      "step": 2733,
      "time_per_iteration": 2.738898515701294
    },
    {
      "auxiliary_loss_clip": 0.01250879,
      "auxiliary_loss_mlp": 0.02570728,
      "balance_loss_clip": 1.05641246,
      "balance_loss_mlp": 1.00008988,
      "epoch": 0.3287440630072747,
      "flos": 21392812471680.0,
      "grad_norm": 1.8727325527428638,
      "language_loss": 0.79877555,
      "learning_rate": 3.134748303797373e-06,
      "loss": 0.83699167,
      "num_input_tokens_seen": 58930090,
      "step": 2734,
      "time_per_iteration": 2.659327507019043
    },
    {
      "auxiliary_loss_clip": 0.01307379,
      "auxiliary_loss_mlp": 0.01034852,
      "balance_loss_clip": 1.05179,
      "balance_loss_mlp": 1.0255661,
      "epoch": 0.32886430589791377,
      "flos": 23732536579200.0,
      "grad_norm": 2.1531305285056654,
      "language_loss": 0.80702233,
      "learning_rate": 3.1341067656560203e-06,
      "loss": 0.83044469,
      "num_input_tokens_seen": 58947935,
      "step": 2735,
      "time_per_iteration": 2.7330806255340576
    },
    {
      "auxiliary_loss_clip": 0.01310011,
      "auxiliary_loss_mlp": 0.01035025,
      "balance_loss_clip": 1.05593848,
      "balance_loss_mlp": 1.02612603,
      "epoch": 0.3289845487885529,
      "flos": 22418708814720.0,
      "grad_norm": 1.8315908627645825,
      "language_loss": 0.86562455,
      "learning_rate": 3.133465055471572e-06,
      "loss": 0.88907492,
      "num_input_tokens_seen": 58967720,
      "step": 2736,
      "time_per_iteration": 2.7080581188201904
    },
    {
      "auxiliary_loss_clip": 0.01342602,
      "auxiliary_loss_mlp": 0.01037163,
      "balance_loss_clip": 1.05312371,
      "balance_loss_mlp": 1.02912831,
      "epoch": 0.329104791679192,
      "flos": 19682603147520.0,
      "grad_norm": 2.3480426532730667,
      "language_loss": 0.6631425,
      "learning_rate": 3.1328231733413767e-06,
      "loss": 0.68694013,
      "num_input_tokens_seen": 58984360,
      "step": 2737,
      "time_per_iteration": 2.690516948699951
    },
    {
      "auxiliary_loss_clip": 0.01251128,
      "auxiliary_loss_mlp": 0.01032275,
      "balance_loss_clip": 1.05760932,
      "balance_loss_mlp": 1.02418637,
      "epoch": 0.32922503456983104,
      "flos": 15997234803840.0,
      "grad_norm": 2.1347583773438994,
      "language_loss": 0.90849668,
      "learning_rate": 3.1321811193628067e-06,
      "loss": 0.93133068,
      "num_input_tokens_seen": 59002505,
      "step": 2738,
      "time_per_iteration": 2.59450626373291
    },
    {
      "auxiliary_loss_clip": 0.01251459,
      "auxiliary_loss_mlp": 0.02575666,
      "balance_loss_clip": 1.05952215,
      "balance_loss_mlp": 1.00012183,
      "epoch": 0.32934527746047015,
      "flos": 26834069260800.0,
      "grad_norm": 2.6309933939341663,
      "language_loss": 0.70081508,
      "learning_rate": 3.131538893633261e-06,
      "loss": 0.73908633,
      "num_input_tokens_seen": 59022065,
      "step": 2739,
      "time_per_iteration": 2.695283889770508
    },
    {
      "auxiliary_loss_clip": 0.01201483,
      "auxiliary_loss_mlp": 0.01037338,
      "balance_loss_clip": 1.0597024,
      "balance_loss_mlp": 1.02913618,
      "epoch": 0.32946552035110926,
      "flos": 23403774372480.0,
      "grad_norm": 2.036177148804156,
      "language_loss": 0.77839041,
      "learning_rate": 3.130896496250165e-06,
      "loss": 0.80077857,
      "num_input_tokens_seen": 59041890,
      "step": 2740,
      "time_per_iteration": 3.492018461227417
    },
    {
      "auxiliary_loss_clip": 0.01200263,
      "auxiliary_loss_mlp": 0.01031387,
      "balance_loss_clip": 1.05745006,
      "balance_loss_mlp": 1.02289343,
      "epoch": 0.3295857632417483,
      "flos": 14172470029440.0,
      "grad_norm": 2.046314384363834,
      "language_loss": 0.86986434,
      "learning_rate": 3.1302539273109693e-06,
      "loss": 0.89218092,
      "num_input_tokens_seen": 59058715,
      "step": 2741,
      "time_per_iteration": 3.417073965072632
    },
    {
      "auxiliary_loss_clip": 0.0129533,
      "auxiliary_loss_mlp": 0.01038514,
      "balance_loss_clip": 1.05558896,
      "balance_loss_mlp": 1.02956176,
      "epoch": 0.32970600613238743,
      "flos": 22196708807040.0,
      "grad_norm": 1.7229827257563408,
      "language_loss": 0.80462146,
      "learning_rate": 3.1296111869131513e-06,
      "loss": 0.8279599,
      "num_input_tokens_seen": 59076140,
      "step": 2742,
      "time_per_iteration": 2.6722474098205566
    },
    {
      "auxiliary_loss_clip": 0.01200283,
      "auxiliary_loss_mlp": 0.01031737,
      "balance_loss_clip": 1.05807734,
      "balance_loss_mlp": 1.02386308,
      "epoch": 0.32982624902302654,
      "flos": 22053784590720.0,
      "grad_norm": 1.899533047807885,
      "language_loss": 0.85800767,
      "learning_rate": 3.1289682751542153e-06,
      "loss": 0.88032788,
      "num_input_tokens_seen": 59095700,
      "step": 2743,
      "time_per_iteration": 2.6564061641693115
    },
    {
      "auxiliary_loss_clip": 0.01250834,
      "auxiliary_loss_mlp": 0.01030285,
      "balance_loss_clip": 1.05633307,
      "balance_loss_mlp": 1.02208388,
      "epoch": 0.3299464919136656,
      "flos": 18661626967680.0,
      "grad_norm": 2.125558754619266,
      "language_loss": 0.71352637,
      "learning_rate": 3.1283251921316883e-06,
      "loss": 0.73633754,
      "num_input_tokens_seen": 59113445,
      "step": 2744,
      "time_per_iteration": 2.683009147644043
    },
    {
      "auxiliary_loss_clip": 0.01396257,
      "auxiliary_loss_mlp": 0.01034135,
      "balance_loss_clip": 1.05543947,
      "balance_loss_mlp": 1.02586174,
      "epoch": 0.3300667348043047,
      "flos": 13407357404160.0,
      "grad_norm": 2.063073767723093,
      "language_loss": 0.80892402,
      "learning_rate": 3.1276819379431277e-06,
      "loss": 0.83322787,
      "num_input_tokens_seen": 59131535,
      "step": 2745,
      "time_per_iteration": 2.818302631378174
    },
    {
      "auxiliary_loss_clip": 0.01217891,
      "auxiliary_loss_mlp": 0.0257545,
      "balance_loss_clip": 1.05799115,
      "balance_loss_mlp": 1.00004363,
      "epoch": 0.33018697769494376,
      "flos": 15742556398080.0,
      "grad_norm": 2.071200183266099,
      "language_loss": 0.74970502,
      "learning_rate": 3.1270385126861134e-06,
      "loss": 0.78763843,
      "num_input_tokens_seen": 59149520,
      "step": 2746,
      "time_per_iteration": 2.669538974761963
    },
    {
      "auxiliary_loss_clip": 0.01203758,
      "auxiliary_loss_mlp": 0.01025743,
      "balance_loss_clip": 1.05990231,
      "balance_loss_mlp": 1.01673043,
      "epoch": 0.3303072205855829,
      "flos": 18258601392000.0,
      "grad_norm": 2.0256429720276397,
      "language_loss": 0.82112813,
      "learning_rate": 3.1263949164582533e-06,
      "loss": 0.84342313,
      "num_input_tokens_seen": 59169170,
      "step": 2747,
      "time_per_iteration": 2.609295129776001
    },
    {
      "auxiliary_loss_clip": 0.01201697,
      "auxiliary_loss_mlp": 0.01032528,
      "balance_loss_clip": 1.05671811,
      "balance_loss_mlp": 1.02392077,
      "epoch": 0.330427463476222,
      "flos": 17749424148480.0,
      "grad_norm": 2.2024766073859494,
      "language_loss": 0.78701353,
      "learning_rate": 3.1257511493571797e-06,
      "loss": 0.80935585,
      "num_input_tokens_seen": 59187675,
      "step": 2748,
      "time_per_iteration": 3.483334541320801
    },
    {
      "auxiliary_loss_clip": 0.01346632,
      "auxiliary_loss_mlp": 0.01028716,
      "balance_loss_clip": 1.05289924,
      "balance_loss_mlp": 1.02080631,
      "epoch": 0.33054770636686104,
      "flos": 27162580072320.0,
      "grad_norm": 2.7192951768915212,
      "language_loss": 0.78951049,
      "learning_rate": 3.125107211480552e-06,
      "loss": 0.81326395,
      "num_input_tokens_seen": 59207610,
      "step": 2749,
      "time_per_iteration": 2.739971160888672
    },
    {
      "auxiliary_loss_clip": 0.01355638,
      "auxiliary_loss_mlp": 0.01029245,
      "balance_loss_clip": 1.050264,
      "balance_loss_mlp": 1.02110314,
      "epoch": 0.33066794925750015,
      "flos": 20117193799680.0,
      "grad_norm": 1.6422971777030027,
      "language_loss": 0.8042841,
      "learning_rate": 3.124463102926054e-06,
      "loss": 0.82813293,
      "num_input_tokens_seen": 59226945,
      "step": 2750,
      "time_per_iteration": 3.708165168762207
    },
    {
      "auxiliary_loss_clip": 0.01137722,
      "auxiliary_loss_mlp": 0.01007603,
      "balance_loss_clip": 1.01864874,
      "balance_loss_mlp": 1.00641656,
      "epoch": 0.33078819214813926,
      "flos": 70642609718400.0,
      "grad_norm": 0.847733545070313,
      "language_loss": 0.61625183,
      "learning_rate": 3.1238188237913984e-06,
      "loss": 0.63770509,
      "num_input_tokens_seen": 59291485,
      "step": 2751,
      "time_per_iteration": 3.347686290740967
    },
    {
      "auxiliary_loss_clip": 0.01206867,
      "auxiliary_loss_mlp": 0.01032653,
      "balance_loss_clip": 1.06046271,
      "balance_loss_mlp": 1.02372432,
      "epoch": 0.3309084350387783,
      "flos": 21141940907520.0,
      "grad_norm": 2.5170871832964057,
      "language_loss": 0.76035041,
      "learning_rate": 3.1231743741743202e-06,
      "loss": 0.7827456,
      "num_input_tokens_seen": 59310990,
      "step": 2752,
      "time_per_iteration": 2.6039533615112305
    },
    {
      "auxiliary_loss_clip": 0.01244747,
      "auxiliary_loss_mlp": 0.01028585,
      "balance_loss_clip": 1.05110145,
      "balance_loss_mlp": 1.02018058,
      "epoch": 0.3310286779294174,
      "flos": 14209350318720.0,
      "grad_norm": 2.1894620456584453,
      "language_loss": 0.83552361,
      "learning_rate": 3.122529754172582e-06,
      "loss": 0.85825694,
      "num_input_tokens_seen": 59327875,
      "step": 2753,
      "time_per_iteration": 2.645589590072632
    },
    {
      "auxiliary_loss_clip": 0.01247636,
      "auxiliary_loss_mlp": 0.01036174,
      "balance_loss_clip": 1.05691874,
      "balance_loss_mlp": 1.02783513,
      "epoch": 0.33114892082005654,
      "flos": 20778130005120.0,
      "grad_norm": 8.858543105136327,
      "language_loss": 0.72013545,
      "learning_rate": 3.1218849638839736e-06,
      "loss": 0.74297357,
      "num_input_tokens_seen": 59347135,
      "step": 2754,
      "time_per_iteration": 2.656952381134033
    },
    {
      "auxiliary_loss_clip": 0.01339291,
      "auxiliary_loss_mlp": 0.0102793,
      "balance_loss_clip": 1.04602444,
      "balance_loss_mlp": 1.01866174,
      "epoch": 0.3312691637106956,
      "flos": 17090750499840.0,
      "grad_norm": 4.682141996399791,
      "language_loss": 0.7841866,
      "learning_rate": 3.121240003406307e-06,
      "loss": 0.80785882,
      "num_input_tokens_seen": 59365985,
      "step": 2755,
      "time_per_iteration": 2.746295928955078
    },
    {
      "auxiliary_loss_clip": 0.01354288,
      "auxiliary_loss_mlp": 0.01032618,
      "balance_loss_clip": 1.0560329,
      "balance_loss_mlp": 1.02363014,
      "epoch": 0.3313894066013347,
      "flos": 29456230008960.0,
      "grad_norm": 2.169601903256514,
      "language_loss": 0.72642744,
      "learning_rate": 3.120594872837425e-06,
      "loss": 0.75029647,
      "num_input_tokens_seen": 59384655,
      "step": 2756,
      "time_per_iteration": 2.78584361076355
    },
    {
      "auxiliary_loss_clip": 0.01140993,
      "auxiliary_loss_mlp": 0.0251621,
      "balance_loss_clip": 1.02070618,
      "balance_loss_mlp": 0.99983913,
      "epoch": 0.3315096494919738,
      "flos": 61419242280960.0,
      "grad_norm": 0.8286886989355703,
      "language_loss": 0.62330782,
      "learning_rate": 3.1199495722751906e-06,
      "loss": 0.6598798,
      "num_input_tokens_seen": 59444185,
      "step": 2757,
      "time_per_iteration": 3.2382733821868896
    },
    {
      "auxiliary_loss_clip": 0.01397284,
      "auxiliary_loss_mlp": 0.01032363,
      "balance_loss_clip": 1.0496093,
      "balance_loss_mlp": 1.02335644,
      "epoch": 0.33162989238261287,
      "flos": 21653057485440.0,
      "grad_norm": 2.2873842194143257,
      "language_loss": 0.84079999,
      "learning_rate": 3.1193041018174972e-06,
      "loss": 0.86509645,
      "num_input_tokens_seen": 59464900,
      "step": 2758,
      "time_per_iteration": 2.8033053874969482
    },
    {
      "auxiliary_loss_clip": 0.01253868,
      "auxiliary_loss_mlp": 0.0103681,
      "balance_loss_clip": 1.05713415,
      "balance_loss_mlp": 1.02797675,
      "epoch": 0.331750135273252,
      "flos": 22674787850880.0,
      "grad_norm": 2.0783511928603713,
      "language_loss": 0.94733453,
      "learning_rate": 3.118658461562261e-06,
      "loss": 0.97024131,
      "num_input_tokens_seen": 59481000,
      "step": 2759,
      "time_per_iteration": 2.6488711833953857
    },
    {
      "auxiliary_loss_clip": 0.01302518,
      "auxiliary_loss_mlp": 0.01033567,
      "balance_loss_clip": 1.05972767,
      "balance_loss_mlp": 1.02461481,
      "epoch": 0.33187037816389103,
      "flos": 22746896403840.0,
      "grad_norm": 1.6187275731717328,
      "language_loss": 0.85175478,
      "learning_rate": 3.118012651607426e-06,
      "loss": 0.87511563,
      "num_input_tokens_seen": 59502605,
      "step": 2760,
      "time_per_iteration": 2.731895923614502
    },
    {
      "auxiliary_loss_clip": 0.01201385,
      "auxiliary_loss_mlp": 0.01036488,
      "balance_loss_clip": 1.05918026,
      "balance_loss_mlp": 1.02788711,
      "epoch": 0.33199062105453014,
      "flos": 19203769918080.0,
      "grad_norm": 1.943628110729952,
      "language_loss": 0.83164418,
      "learning_rate": 3.1173666720509603e-06,
      "loss": 0.85402286,
      "num_input_tokens_seen": 59519540,
      "step": 2761,
      "time_per_iteration": 2.594526529312134
    },
    {
      "auxiliary_loss_clip": 0.01302974,
      "auxiliary_loss_mlp": 0.01032939,
      "balance_loss_clip": 1.05466187,
      "balance_loss_mlp": 1.02474892,
      "epoch": 0.33211086394516925,
      "flos": 31577006764800.0,
      "grad_norm": 1.7998656683018657,
      "language_loss": 0.6842587,
      "learning_rate": 3.116720522990859e-06,
      "loss": 0.70761776,
      "num_input_tokens_seen": 59540415,
      "step": 2762,
      "time_per_iteration": 2.7679738998413086
    },
    {
      "auxiliary_loss_clip": 0.01448017,
      "auxiliary_loss_mlp": 0.01032694,
      "balance_loss_clip": 1.05100858,
      "balance_loss_mlp": 1.02448022,
      "epoch": 0.3322311068358083,
      "flos": 17932496791680.0,
      "grad_norm": 2.1012007059063604,
      "language_loss": 0.62210685,
      "learning_rate": 3.116074204525142e-06,
      "loss": 0.64691401,
      "num_input_tokens_seen": 59558590,
      "step": 2763,
      "time_per_iteration": 2.7465224266052246
    },
    {
      "auxiliary_loss_clip": 0.01245958,
      "auxiliary_loss_mlp": 0.01034675,
      "balance_loss_clip": 1.0556283,
      "balance_loss_mlp": 1.02615178,
      "epoch": 0.3323513497264474,
      "flos": 32269831269120.0,
      "grad_norm": 1.8524393975877405,
      "language_loss": 0.83665496,
      "learning_rate": 3.1154277167518553e-06,
      "loss": 0.85946125,
      "num_input_tokens_seen": 59580205,
      "step": 2764,
      "time_per_iteration": 2.7482433319091797
    },
    {
      "auxiliary_loss_clip": 0.01185874,
      "auxiliary_loss_mlp": 0.01004212,
      "balance_loss_clip": 1.01715136,
      "balance_loss_mlp": 1.00279951,
      "epoch": 0.33247159261708653,
      "flos": 52668674588160.0,
      "grad_norm": 0.7764877702506137,
      "language_loss": 0.59524548,
      "learning_rate": 3.114781059769072e-06,
      "loss": 0.61714637,
      "num_input_tokens_seen": 59631530,
      "step": 2765,
      "time_per_iteration": 3.1152126789093018
    },
    {
      "auxiliary_loss_clip": 0.0129781,
      "auxiliary_loss_mlp": 0.01028205,
      "balance_loss_clip": 1.05627608,
      "balance_loss_mlp": 1.01908565,
      "epoch": 0.3325918355077256,
      "flos": 27125232906240.0,
      "grad_norm": 5.271492523949398,
      "language_loss": 0.67877269,
      "learning_rate": 3.1141342336748874e-06,
      "loss": 0.7020328,
      "num_input_tokens_seen": 59651090,
      "step": 2766,
      "time_per_iteration": 3.6468052864074707
    },
    {
      "auxiliary_loss_clip": 0.01241552,
      "auxiliary_loss_mlp": 0.01034364,
      "balance_loss_clip": 1.05538034,
      "balance_loss_mlp": 1.02646613,
      "epoch": 0.3327120783983647,
      "flos": 23664414435840.0,
      "grad_norm": 1.5464679333151938,
      "language_loss": 0.82202059,
      "learning_rate": 3.1134872385674253e-06,
      "loss": 0.84477973,
      "num_input_tokens_seen": 59675245,
      "step": 2767,
      "time_per_iteration": 3.6118650436401367
    },
    {
      "auxiliary_loss_clip": 0.01300825,
      "auxiliary_loss_mlp": 0.01034747,
      "balance_loss_clip": 1.05155706,
      "balance_loss_mlp": 1.02605689,
      "epoch": 0.3328323212890038,
      "flos": 19171378828800.0,
      "grad_norm": 1.7434703282554898,
      "language_loss": 0.86048424,
      "learning_rate": 3.1128400745448353e-06,
      "loss": 0.88383996,
      "num_input_tokens_seen": 59694625,
      "step": 2768,
      "time_per_iteration": 2.7182064056396484
    },
    {
      "auxiliary_loss_clip": 0.0125155,
      "auxiliary_loss_mlp": 0.01034054,
      "balance_loss_clip": 1.05755782,
      "balance_loss_mlp": 1.02488041,
      "epoch": 0.33295256417964286,
      "flos": 37706347463040.0,
      "grad_norm": 2.019712212760729,
      "language_loss": 0.62909806,
      "learning_rate": 3.11219274170529e-06,
      "loss": 0.65195405,
      "num_input_tokens_seen": 59716435,
      "step": 2769,
      "time_per_iteration": 2.7237935066223145
    },
    {
      "auxiliary_loss_clip": 0.01291717,
      "auxiliary_loss_mlp": 0.01032802,
      "balance_loss_clip": 1.05194306,
      "balance_loss_mlp": 1.02487421,
      "epoch": 0.333072807070282,
      "flos": 26505989412480.0,
      "grad_norm": 1.7154468268285585,
      "language_loss": 0.81531727,
      "learning_rate": 3.1115452401469903e-06,
      "loss": 0.83856243,
      "num_input_tokens_seen": 59736835,
      "step": 2770,
      "time_per_iteration": 2.7129788398742676
    },
    {
      "auxiliary_loss_clip": 0.01393764,
      "auxiliary_loss_mlp": 0.01034169,
      "balance_loss_clip": 1.04916573,
      "balance_loss_mlp": 1.02608633,
      "epoch": 0.3331930499609211,
      "flos": 21430913823360.0,
      "grad_norm": 1.978102616197759,
      "language_loss": 0.86409819,
      "learning_rate": 3.1108975699681613e-06,
      "loss": 0.88837743,
      "num_input_tokens_seen": 59754230,
      "step": 2771,
      "time_per_iteration": 2.744023561477661
    },
    {
      "auxiliary_loss_clip": 0.01347182,
      "auxiliary_loss_mlp": 0.01034362,
      "balance_loss_clip": 1.05319035,
      "balance_loss_mlp": 1.02649474,
      "epoch": 0.33331329285156014,
      "flos": 20659947281280.0,
      "grad_norm": 1.7370548161557668,
      "language_loss": 0.71527684,
      "learning_rate": 3.1102497312670542e-06,
      "loss": 0.73909223,
      "num_input_tokens_seen": 59772235,
      "step": 2772,
      "time_per_iteration": 2.740532159805298
    },
    {
      "auxiliary_loss_clip": 0.01286842,
      "auxiliary_loss_mlp": 0.01029762,
      "balance_loss_clip": 1.04918337,
      "balance_loss_mlp": 1.02181709,
      "epoch": 0.33343353574219925,
      "flos": 28001596930560.0,
      "grad_norm": 2.015949721084627,
      "language_loss": 0.80705106,
      "learning_rate": 3.109601724141946e-06,
      "loss": 0.83021712,
      "num_input_tokens_seen": 59791230,
      "step": 2773,
      "time_per_iteration": 2.7140095233917236
    },
    {
      "auxiliary_loss_clip": 0.01204902,
      "auxiliary_loss_mlp": 0.0103746,
      "balance_loss_clip": 1.05242705,
      "balance_loss_mlp": 1.02877581,
      "epoch": 0.33355377863283836,
      "flos": 23764963582080.0,
      "grad_norm": 2.6874592582794268,
      "language_loss": 0.68311036,
      "learning_rate": 3.108953548691138e-06,
      "loss": 0.70553398,
      "num_input_tokens_seen": 59811315,
      "step": 2774,
      "time_per_iteration": 3.5832653045654297
    },
    {
      "auxiliary_loss_clip": 0.01200962,
      "auxiliary_loss_mlp": 0.01030833,
      "balance_loss_clip": 1.05795896,
      "balance_loss_mlp": 1.02257228,
      "epoch": 0.3336740215234774,
      "flos": 37779677078400.0,
      "grad_norm": 2.663432802853727,
      "language_loss": 0.72415268,
      "learning_rate": 3.108305205012959e-06,
      "loss": 0.74647069,
      "num_input_tokens_seen": 59832010,
      "step": 2775,
      "time_per_iteration": 3.607879877090454
    },
    {
      "auxiliary_loss_clip": 0.01298734,
      "auxiliary_loss_mlp": 0.01029506,
      "balance_loss_clip": 1.05512369,
      "balance_loss_mlp": 1.02222824,
      "epoch": 0.3337942644141165,
      "flos": 25519056347520.0,
      "grad_norm": 2.0458596462054026,
      "language_loss": 0.87452739,
      "learning_rate": 3.107656693205761e-06,
      "loss": 0.8978098,
      "num_input_tokens_seen": 59851450,
      "step": 2776,
      "time_per_iteration": 2.6674318313598633
    },
    {
      "auxiliary_loss_clip": 0.01206726,
      "auxiliary_loss_mlp": 0.010357,
      "balance_loss_clip": 1.05939186,
      "balance_loss_mlp": 1.02637792,
      "epoch": 0.3339145073047556,
      "flos": 25989844930560.0,
      "grad_norm": 2.4603400159097664,
      "language_loss": 0.70731485,
      "learning_rate": 3.107008013367924e-06,
      "loss": 0.72973907,
      "num_input_tokens_seen": 59870245,
      "step": 2777,
      "time_per_iteration": 2.624476671218872
    },
    {
      "auxiliary_loss_clip": 0.0134245,
      "auxiliary_loss_mlp": 0.01027669,
      "balance_loss_clip": 1.0503819,
      "balance_loss_mlp": 1.01944315,
      "epoch": 0.3340347501953947,
      "flos": 19062569554560.0,
      "grad_norm": 2.2425945038124597,
      "language_loss": 0.86661565,
      "learning_rate": 3.1063591655978507e-06,
      "loss": 0.89031684,
      "num_input_tokens_seen": 59886195,
      "step": 2778,
      "time_per_iteration": 2.7107934951782227
    },
    {
      "auxiliary_loss_clip": 0.01295523,
      "auxiliary_loss_mlp": 0.01032023,
      "balance_loss_clip": 1.04507422,
      "balance_loss_mlp": 1.02296925,
      "epoch": 0.3341549930860338,
      "flos": 18109715518080.0,
      "grad_norm": 4.58374583264801,
      "language_loss": 0.79991364,
      "learning_rate": 3.105710149993972e-06,
      "loss": 0.82318914,
      "num_input_tokens_seen": 59905525,
      "step": 2779,
      "time_per_iteration": 2.7472386360168457
    },
    {
      "auxiliary_loss_clip": 0.01201781,
      "auxiliary_loss_mlp": 0.01035747,
      "balance_loss_clip": 1.05801034,
      "balance_loss_mlp": 1.02773571,
      "epoch": 0.33427523597667286,
      "flos": 22674967418880.0,
      "grad_norm": 2.5387504433761943,
      "language_loss": 0.85855955,
      "learning_rate": 3.1050609666547427e-06,
      "loss": 0.88093483,
      "num_input_tokens_seen": 59925085,
      "step": 2780,
      "time_per_iteration": 2.6382105350494385
    },
    {
      "auxiliary_loss_clip": 0.01352055,
      "auxiliary_loss_mlp": 0.01033893,
      "balance_loss_clip": 1.05358946,
      "balance_loss_mlp": 1.02544117,
      "epoch": 0.33439547886731197,
      "flos": 22638338524800.0,
      "grad_norm": 1.8736338027229247,
      "language_loss": 0.77823442,
      "learning_rate": 3.104411615678644e-06,
      "loss": 0.80209392,
      "num_input_tokens_seen": 59943935,
      "step": 2781,
      "time_per_iteration": 2.734347105026245
    },
    {
      "auxiliary_loss_clip": 0.01304948,
      "auxiliary_loss_mlp": 0.0104344,
      "balance_loss_clip": 1.0578028,
      "balance_loss_mlp": 1.03448164,
      "epoch": 0.3345157217579511,
      "flos": 24096383395200.0,
      "grad_norm": 3.4621394168369406,
      "language_loss": 0.7357372,
      "learning_rate": 3.1037620971641803e-06,
      "loss": 0.75922108,
      "num_input_tokens_seen": 59963725,
      "step": 2782,
      "time_per_iteration": 2.8004729747772217
    },
    {
      "auxiliary_loss_clip": 0.01201522,
      "auxiliary_loss_mlp": 0.01030479,
      "balance_loss_clip": 1.0580771,
      "balance_loss_mlp": 1.0215379,
      "epoch": 0.33463596464859013,
      "flos": 18989491334400.0,
      "grad_norm": 2.619045902386856,
      "language_loss": 0.65488327,
      "learning_rate": 3.1031124112098844e-06,
      "loss": 0.6772033,
      "num_input_tokens_seen": 59981935,
      "step": 2783,
      "time_per_iteration": 2.5992393493652344
    },
    {
      "auxiliary_loss_clip": 0.01300142,
      "auxiliary_loss_mlp": 0.01037325,
      "balance_loss_clip": 1.05449033,
      "balance_loss_mlp": 1.02853894,
      "epoch": 0.33475620753922924,
      "flos": 20375607219840.0,
      "grad_norm": 2.018403684885105,
      "language_loss": 0.72516191,
      "learning_rate": 3.1024625579143127e-06,
      "loss": 0.74853659,
      "num_input_tokens_seen": 59999455,
      "step": 2784,
      "time_per_iteration": 2.6751391887664795
    },
    {
      "auxiliary_loss_clip": 0.01200145,
      "auxiliary_loss_mlp": 0.01039066,
      "balance_loss_clip": 1.05736971,
      "balance_loss_mlp": 1.03028297,
      "epoch": 0.33487645042986836,
      "flos": 18182578256640.0,
      "grad_norm": 5.625520016850965,
      "language_loss": 0.7325536,
      "learning_rate": 3.101812537376048e-06,
      "loss": 0.7549457,
      "num_input_tokens_seen": 60018475,
      "step": 2785,
      "time_per_iteration": 2.6273956298828125
    },
    {
      "auxiliary_loss_clip": 0.01290044,
      "auxiliary_loss_mlp": 0.02570789,
      "balance_loss_clip": 1.05139744,
      "balance_loss_mlp": 1.0000273,
      "epoch": 0.3349966933205074,
      "flos": 25848824135040.0,
      "grad_norm": 1.955391173297287,
      "language_loss": 0.84733456,
      "learning_rate": 3.1011623496936973e-06,
      "loss": 0.88594294,
      "num_input_tokens_seen": 60036770,
      "step": 2786,
      "time_per_iteration": 2.7413854598999023
    },
    {
      "auxiliary_loss_clip": 0.01195992,
      "auxiliary_loss_mlp": 0.0103395,
      "balance_loss_clip": 1.05591655,
      "balance_loss_mlp": 1.02568316,
      "epoch": 0.3351169362111465,
      "flos": 28111447699200.0,
      "grad_norm": 1.9557450623242916,
      "language_loss": 0.69575524,
      "learning_rate": 3.100511994965893e-06,
      "loss": 0.71805465,
      "num_input_tokens_seen": 60056725,
      "step": 2787,
      "time_per_iteration": 2.623093843460083
    },
    {
      "auxiliary_loss_clip": 0.01244021,
      "auxiliary_loss_mlp": 0.01037898,
      "balance_loss_clip": 1.05621099,
      "balance_loss_mlp": 1.03007245,
      "epoch": 0.33523717910178563,
      "flos": 22673315393280.0,
      "grad_norm": 1.7538028981508496,
      "language_loss": 0.84761012,
      "learning_rate": 3.0998614732912947e-06,
      "loss": 0.8704294,
      "num_input_tokens_seen": 60076100,
      "step": 2788,
      "time_per_iteration": 2.693640947341919
    },
    {
      "auxiliary_loss_clip": 0.01248804,
      "auxiliary_loss_mlp": 0.01032555,
      "balance_loss_clip": 1.05791688,
      "balance_loss_mlp": 1.02378082,
      "epoch": 0.3353574219924247,
      "flos": 15669801400320.0,
      "grad_norm": 2.1441176688974637,
      "language_loss": 0.68325448,
      "learning_rate": 3.0992107847685855e-06,
      "loss": 0.70606804,
      "num_input_tokens_seen": 60093815,
      "step": 2789,
      "time_per_iteration": 2.6067380905151367
    },
    {
      "auxiliary_loss_clip": 0.01298339,
      "auxiliary_loss_mlp": 0.01039334,
      "balance_loss_clip": 1.05856013,
      "balance_loss_mlp": 1.02983356,
      "epoch": 0.3354776648830638,
      "flos": 24790644443520.0,
      "grad_norm": 1.8194510368540773,
      "language_loss": 0.79316336,
      "learning_rate": 3.0985599294964736e-06,
      "loss": 0.81654012,
      "num_input_tokens_seen": 60113370,
      "step": 2790,
      "time_per_iteration": 2.6973791122436523
    },
    {
      "auxiliary_loss_clip": 0.01271499,
      "auxiliary_loss_mlp": 0.01039355,
      "balance_loss_clip": 1.0556314,
      "balance_loss_mlp": 1.03031325,
      "epoch": 0.33559790777370285,
      "flos": 28694852398080.0,
      "grad_norm": 1.8876852404589808,
      "language_loss": 0.70185524,
      "learning_rate": 3.097908907573695e-06,
      "loss": 0.72496372,
      "num_input_tokens_seen": 60131350,
      "step": 2791,
      "time_per_iteration": 2.7446839809417725
    },
    {
      "auxiliary_loss_clip": 0.0144526,
      "auxiliary_loss_mlp": 0.01030872,
      "balance_loss_clip": 1.05347109,
      "balance_loss_mlp": 1.02296877,
      "epoch": 0.33571815066434196,
      "flos": 22235779825920.0,
      "grad_norm": 2.0192186812961954,
      "language_loss": 0.89545643,
      "learning_rate": 3.0972577190990067e-06,
      "loss": 0.92021775,
      "num_input_tokens_seen": 60149830,
      "step": 2792,
      "time_per_iteration": 3.706181049346924
    },
    {
      "auxiliary_loss_clip": 0.0135201,
      "auxiliary_loss_mlp": 0.01036673,
      "balance_loss_clip": 1.0535835,
      "balance_loss_mlp": 1.02830482,
      "epoch": 0.3358383935549811,
      "flos": 23842279607040.0,
      "grad_norm": 1.8553399103126984,
      "language_loss": 0.80052805,
      "learning_rate": 3.096606364171196e-06,
      "loss": 0.82441485,
      "num_input_tokens_seen": 60169620,
      "step": 2793,
      "time_per_iteration": 3.6682770252227783
    },
    {
      "auxiliary_loss_clip": 0.01341204,
      "auxiliary_loss_mlp": 0.01032857,
      "balance_loss_clip": 1.05255628,
      "balance_loss_mlp": 1.02449441,
      "epoch": 0.33595863644562013,
      "flos": 22267308988800.0,
      "grad_norm": 1.920542284244653,
      "language_loss": 0.85274148,
      "learning_rate": 3.0959548428890703e-06,
      "loss": 0.87648207,
      "num_input_tokens_seen": 60188490,
      "step": 2794,
      "time_per_iteration": 2.688107967376709
    },
    {
      "auxiliary_loss_clip": 0.01247473,
      "auxiliary_loss_mlp": 0.01030394,
      "balance_loss_clip": 1.05815566,
      "balance_loss_mlp": 1.02224016,
      "epoch": 0.33607887933625924,
      "flos": 20119779578880.0,
      "grad_norm": 2.2162673016810976,
      "language_loss": 0.84005433,
      "learning_rate": 3.095303155351468e-06,
      "loss": 0.86283302,
      "num_input_tokens_seen": 60208695,
      "step": 2795,
      "time_per_iteration": 2.669416904449463
    },
    {
      "auxiliary_loss_clip": 0.01391975,
      "auxiliary_loss_mlp": 0.01036128,
      "balance_loss_clip": 1.05039001,
      "balance_loss_mlp": 1.0282836,
      "epoch": 0.33619912222689835,
      "flos": 19318109886720.0,
      "grad_norm": 2.6231057846577297,
      "language_loss": 0.80003983,
      "learning_rate": 3.0946513016572464e-06,
      "loss": 0.82432085,
      "num_input_tokens_seen": 60227600,
      "step": 2796,
      "time_per_iteration": 2.7578086853027344
    },
    {
      "auxiliary_loss_clip": 0.01256084,
      "auxiliary_loss_mlp": 0.01031671,
      "balance_loss_clip": 1.05824137,
      "balance_loss_mlp": 1.02236044,
      "epoch": 0.3363193651175374,
      "flos": 16800664262400.0,
      "grad_norm": 2.2957509561373026,
      "language_loss": 0.76776075,
      "learning_rate": 3.0939992819052938e-06,
      "loss": 0.79063833,
      "num_input_tokens_seen": 60245110,
      "step": 2797,
      "time_per_iteration": 2.5690016746520996
    },
    {
      "auxiliary_loss_clip": 0.01298414,
      "auxiliary_loss_mlp": 0.01035703,
      "balance_loss_clip": 1.05548191,
      "balance_loss_mlp": 1.02758503,
      "epoch": 0.3364396080081765,
      "flos": 23550289948800.0,
      "grad_norm": 2.1555898322291354,
      "language_loss": 0.81108975,
      "learning_rate": 3.0933470961945193e-06,
      "loss": 0.83443093,
      "num_input_tokens_seen": 60263405,
      "step": 2798,
      "time_per_iteration": 2.7319107055664062
    },
    {
      "auxiliary_loss_clip": 0.01293056,
      "auxiliary_loss_mlp": 0.01027229,
      "balance_loss_clip": 1.05501771,
      "balance_loss_mlp": 1.0194149,
      "epoch": 0.3365598508988156,
      "flos": 28037902602240.0,
      "grad_norm": 2.556095907357704,
      "language_loss": 0.68244994,
      "learning_rate": 3.0926947446238597e-06,
      "loss": 0.70565283,
      "num_input_tokens_seen": 60282975,
      "step": 2799,
      "time_per_iteration": 2.7062857151031494
    },
    {
      "auxiliary_loss_clip": 0.012558,
      "auxiliary_loss_mlp": 0.01034257,
      "balance_loss_clip": 1.05601263,
      "balance_loss_mlp": 1.02554917,
      "epoch": 0.3366800937894547,
      "flos": 16982767238400.0,
      "grad_norm": 2.418136690683408,
      "language_loss": 0.82298017,
      "learning_rate": 3.092042227292276e-06,
      "loss": 0.84588075,
      "num_input_tokens_seen": 60299810,
      "step": 2800,
      "time_per_iteration": 3.579735517501831
    },
    {
      "auxiliary_loss_clip": 0.01200587,
      "auxiliary_loss_mlp": 0.01030937,
      "balance_loss_clip": 1.05957985,
      "balance_loss_mlp": 1.02318811,
      "epoch": 0.3368003366800938,
      "flos": 23915321913600.0,
      "grad_norm": 1.6584475596709194,
      "language_loss": 0.88412666,
      "learning_rate": 3.0913895442987557e-06,
      "loss": 0.90644193,
      "num_input_tokens_seen": 60320775,
      "step": 2801,
      "time_per_iteration": 3.609792709350586
    },
    {
      "auxiliary_loss_clip": 0.01344038,
      "auxiliary_loss_mlp": 0.02574172,
      "balance_loss_clip": 1.05435681,
      "balance_loss_mlp": 0.99999923,
      "epoch": 0.3369205795707329,
      "flos": 24791219061120.0,
      "grad_norm": 1.549887470672772,
      "language_loss": 0.8596772,
      "learning_rate": 3.090736695742308e-06,
      "loss": 0.89885926,
      "num_input_tokens_seen": 60341905,
      "step": 2802,
      "time_per_iteration": 2.992711067199707
    },
    {
      "auxiliary_loss_clip": 0.01393148,
      "auxiliary_loss_mlp": 0.01028257,
      "balance_loss_clip": 1.05032766,
      "balance_loss_mlp": 1.02048469,
      "epoch": 0.33704082246137196,
      "flos": 17931096161280.0,
      "grad_norm": 2.1945931637305005,
      "language_loss": 0.52512771,
      "learning_rate": 3.0900836817219713e-06,
      "loss": 0.54934174,
      "num_input_tokens_seen": 60358335,
      "step": 2803,
      "time_per_iteration": 2.7522132396698
    },
    {
      "auxiliary_loss_clip": 0.01200145,
      "auxiliary_loss_mlp": 0.01027212,
      "balance_loss_clip": 1.0582372,
      "balance_loss_mlp": 1.01946294,
      "epoch": 0.33716106535201107,
      "flos": 21286517149440.0,
      "grad_norm": 2.1331154628984668,
      "language_loss": 0.83926344,
      "learning_rate": 3.089430502336807e-06,
      "loss": 0.86153704,
      "num_input_tokens_seen": 60378305,
      "step": 2804,
      "time_per_iteration": 2.6406409740448
    },
    {
      "auxiliary_loss_clip": 0.0125389,
      "auxiliary_loss_mlp": 0.01027484,
      "balance_loss_clip": 1.05743003,
      "balance_loss_mlp": 1.01864433,
      "epoch": 0.3372813082426502,
      "flos": 18402962152320.0,
      "grad_norm": 2.879870179361866,
      "language_loss": 0.9048714,
      "learning_rate": 3.088777157685902e-06,
      "loss": 0.92768514,
      "num_input_tokens_seen": 60393895,
      "step": 2805,
      "time_per_iteration": 2.640377998352051
    },
    {
      "auxiliary_loss_clip": 0.01295234,
      "auxiliary_loss_mlp": 0.01030919,
      "balance_loss_clip": 1.05652869,
      "balance_loss_mlp": 1.02276516,
      "epoch": 0.33740155113328923,
      "flos": 17201391367680.0,
      "grad_norm": 1.775621810347205,
      "language_loss": 0.85765034,
      "learning_rate": 3.088123647868367e-06,
      "loss": 0.88091189,
      "num_input_tokens_seen": 60410445,
      "step": 2806,
      "time_per_iteration": 2.586916446685791
    },
    {
      "auxiliary_loss_clip": 0.01256519,
      "auxiliary_loss_mlp": 0.01033005,
      "balance_loss_clip": 1.05726624,
      "balance_loss_mlp": 1.02441597,
      "epoch": 0.33752179402392835,
      "flos": 29058950609280.0,
      "grad_norm": 4.708248592912433,
      "language_loss": 0.81795371,
      "learning_rate": 3.0874699729833405e-06,
      "loss": 0.84084886,
      "num_input_tokens_seen": 60431815,
      "step": 2807,
      "time_per_iteration": 2.713743209838867
    },
    {
      "auxiliary_loss_clip": 0.01294874,
      "auxiliary_loss_mlp": 0.01026229,
      "balance_loss_clip": 1.05557394,
      "balance_loss_mlp": 1.01837897,
      "epoch": 0.3376420369145674,
      "flos": 25080730680960.0,
      "grad_norm": 1.5892024279808348,
      "language_loss": 0.7992177,
      "learning_rate": 3.086816133129983e-06,
      "loss": 0.82242864,
      "num_input_tokens_seen": 60452075,
      "step": 2808,
      "time_per_iteration": 2.7191998958587646
    },
    {
      "auxiliary_loss_clip": 0.01206637,
      "auxiliary_loss_mlp": 0.01036902,
      "balance_loss_clip": 1.06467474,
      "balance_loss_mlp": 1.0287776,
      "epoch": 0.3377622798052065,
      "flos": 27490624007040.0,
      "grad_norm": 3.157989242402198,
      "language_loss": 0.76340902,
      "learning_rate": 3.0861621284074826e-06,
      "loss": 0.78584445,
      "num_input_tokens_seen": 60472600,
      "step": 2809,
      "time_per_iteration": 2.6715030670166016
    },
    {
      "auxiliary_loss_clip": 0.01216108,
      "auxiliary_loss_mlp": 0.01031855,
      "balance_loss_clip": 1.05881977,
      "balance_loss_mlp": 1.02385557,
      "epoch": 0.3378825226958456,
      "flos": 21975211589760.0,
      "grad_norm": 1.593514767918034,
      "language_loss": 0.72983742,
      "learning_rate": 3.085507958915051e-06,
      "loss": 0.75231707,
      "num_input_tokens_seen": 60491030,
      "step": 2810,
      "time_per_iteration": 2.662234306335449
    },
    {
      "auxiliary_loss_clip": 0.01296243,
      "auxiliary_loss_mlp": 0.01025704,
      "balance_loss_clip": 1.05559945,
      "balance_loss_mlp": 1.01782143,
      "epoch": 0.3380027655864847,
      "flos": 42523189200000.0,
      "grad_norm": 2.5701637048061117,
      "language_loss": 0.71652627,
      "learning_rate": 3.084853624751925e-06,
      "loss": 0.73974574,
      "num_input_tokens_seen": 60512615,
      "step": 2811,
      "time_per_iteration": 2.9217023849487305
    },
    {
      "auxiliary_loss_clip": 0.01353239,
      "auxiliary_loss_mlp": 0.01032007,
      "balance_loss_clip": 1.05834317,
      "balance_loss_mlp": 1.02399063,
      "epoch": 0.3381230084771238,
      "flos": 26725080418560.0,
      "grad_norm": 1.9002776208774446,
      "language_loss": 0.8558706,
      "learning_rate": 3.0841991260173668e-06,
      "loss": 0.87972307,
      "num_input_tokens_seen": 60532520,
      "step": 2812,
      "time_per_iteration": 2.7624831199645996
    },
    {
      "auxiliary_loss_clip": 0.01204173,
      "auxiliary_loss_mlp": 0.01032448,
      "balance_loss_clip": 1.06061471,
      "balance_loss_mlp": 1.02441335,
      "epoch": 0.3382432513677629,
      "flos": 22710375250560.0,
      "grad_norm": 1.9026400019455083,
      "language_loss": 0.80556357,
      "learning_rate": 3.0835444628106634e-06,
      "loss": 0.82792974,
      "num_input_tokens_seen": 60551500,
      "step": 2813,
      "time_per_iteration": 2.6516847610473633
    },
    {
      "auxiliary_loss_clip": 0.01199441,
      "auxiliary_loss_mlp": 0.02572286,
      "balance_loss_clip": 1.05791736,
      "balance_loss_mlp": 0.99997997,
      "epoch": 0.33836349425840195,
      "flos": 22122409524480.0,
      "grad_norm": 1.8450959070441355,
      "language_loss": 0.8335017,
      "learning_rate": 3.082889635231126e-06,
      "loss": 0.87121892,
      "num_input_tokens_seen": 60570160,
      "step": 2814,
      "time_per_iteration": 2.6424148082733154
    },
    {
      "auxiliary_loss_clip": 0.01302915,
      "auxiliary_loss_mlp": 0.01028198,
      "balance_loss_clip": 1.05555987,
      "balance_loss_mlp": 1.01978791,
      "epoch": 0.33848373714904106,
      "flos": 27308090067840.0,
      "grad_norm": 2.709681895240856,
      "language_loss": 0.77549541,
      "learning_rate": 3.0822346433780925e-06,
      "loss": 0.79880655,
      "num_input_tokens_seen": 60590885,
      "step": 2815,
      "time_per_iteration": 2.746055841445923
    },
    {
      "auxiliary_loss_clip": 0.0125323,
      "auxiliary_loss_mlp": 0.01032669,
      "balance_loss_clip": 1.05635333,
      "balance_loss_mlp": 1.02447343,
      "epoch": 0.3386039800396802,
      "flos": 25848716394240.0,
      "grad_norm": 2.751785574958954,
      "language_loss": 0.87506437,
      "learning_rate": 3.0815794873509237e-06,
      "loss": 0.89792341,
      "num_input_tokens_seen": 60609170,
      "step": 2816,
      "time_per_iteration": 2.728621482849121
    },
    {
      "auxiliary_loss_clip": 0.01201671,
      "auxiliary_loss_mlp": 0.01031387,
      "balance_loss_clip": 1.05895519,
      "balance_loss_mlp": 1.02279782,
      "epoch": 0.33872422293031923,
      "flos": 18880646146560.0,
      "grad_norm": 2.2851441945425397,
      "language_loss": 0.72633129,
      "learning_rate": 3.0809241672490066e-06,
      "loss": 0.74866188,
      "num_input_tokens_seen": 60627340,
      "step": 2817,
      "time_per_iteration": 2.5971593856811523
    },
    {
      "auxiliary_loss_clip": 0.01298089,
      "auxiliary_loss_mlp": 0.01035597,
      "balance_loss_clip": 1.05523443,
      "balance_loss_mlp": 1.02749085,
      "epoch": 0.33884446582095834,
      "flos": 23146977064320.0,
      "grad_norm": 1.6565066312209926,
      "language_loss": 0.84890008,
      "learning_rate": 3.080268683171753e-06,
      "loss": 0.87223697,
      "num_input_tokens_seen": 60647630,
      "step": 2818,
      "time_per_iteration": 3.653149366378784
    },
    {
      "auxiliary_loss_clip": 0.01250069,
      "auxiliary_loss_mlp": 0.0102994,
      "balance_loss_clip": 1.05475175,
      "balance_loss_mlp": 1.02245355,
      "epoch": 0.33896470871159745,
      "flos": 15997342544640.0,
      "grad_norm": 1.9957402452401236,
      "language_loss": 0.89259076,
      "learning_rate": 3.0796130352185985e-06,
      "loss": 0.91539085,
      "num_input_tokens_seen": 60664485,
      "step": 2819,
      "time_per_iteration": 3.490203619003296
    },
    {
      "auxiliary_loss_clip": 0.01290964,
      "auxiliary_loss_mlp": 0.02574613,
      "balance_loss_clip": 1.04815495,
      "balance_loss_mlp": 1.00000346,
      "epoch": 0.3390849516022365,
      "flos": 34495754112000.0,
      "grad_norm": 4.566977908397807,
      "language_loss": 0.66949272,
      "learning_rate": 3.0789572234890057e-06,
      "loss": 0.70814848,
      "num_input_tokens_seen": 60686125,
      "step": 2820,
      "time_per_iteration": 2.8559441566467285
    },
    {
      "auxiliary_loss_clip": 0.013021,
      "auxiliary_loss_mlp": 0.0103376,
      "balance_loss_clip": 1.06087303,
      "balance_loss_mlp": 1.02517116,
      "epoch": 0.3392051944928756,
      "flos": 16180307447040.0,
      "grad_norm": 1.8873803002495546,
      "language_loss": 0.77251768,
      "learning_rate": 3.0783012480824596e-06,
      "loss": 0.79587626,
      "num_input_tokens_seen": 60705270,
      "step": 2821,
      "time_per_iteration": 2.7080438137054443
    },
    {
      "auxiliary_loss_clip": 0.01198167,
      "auxiliary_loss_mlp": 0.01038844,
      "balance_loss_clip": 1.05635285,
      "balance_loss_mlp": 1.03013599,
      "epoch": 0.33932543738351467,
      "flos": 17086656349440.0,
      "grad_norm": 4.811029889216106,
      "language_loss": 0.74348807,
      "learning_rate": 3.077645109098471e-06,
      "loss": 0.76585817,
      "num_input_tokens_seen": 60721540,
      "step": 2822,
      "time_per_iteration": 2.6358115673065186
    },
    {
      "auxiliary_loss_clip": 0.01340399,
      "auxiliary_loss_mlp": 0.01032482,
      "balance_loss_clip": 1.05388868,
      "balance_loss_mlp": 1.02432179,
      "epoch": 0.3394456802741538,
      "flos": 22126970551680.0,
      "grad_norm": 1.660770272088191,
      "language_loss": 0.7236433,
      "learning_rate": 3.076988806636577e-06,
      "loss": 0.74737215,
      "num_input_tokens_seen": 60739300,
      "step": 2823,
      "time_per_iteration": 2.7425107955932617
    },
    {
      "auxiliary_loss_clip": 0.01303298,
      "auxiliary_loss_mlp": 0.02576644,
      "balance_loss_clip": 1.05895877,
      "balance_loss_mlp": 0.99998045,
      "epoch": 0.3395659231647929,
      "flos": 25226887121280.0,
      "grad_norm": 2.0491568020610758,
      "language_loss": 0.88920105,
      "learning_rate": 3.0763323407963377e-06,
      "loss": 0.92800045,
      "num_input_tokens_seen": 60758910,
      "step": 2824,
      "time_per_iteration": 2.6998302936553955
    },
    {
      "auxiliary_loss_clip": 0.0124951,
      "auxiliary_loss_mlp": 0.01029751,
      "balance_loss_clip": 1.05601192,
      "balance_loss_mlp": 1.0215261,
      "epoch": 0.33968616605543195,
      "flos": 29096477343360.0,
      "grad_norm": 1.8844807428221013,
      "language_loss": 0.80271935,
      "learning_rate": 3.075675711677337e-06,
      "loss": 0.82551205,
      "num_input_tokens_seen": 60779005,
      "step": 2825,
      "time_per_iteration": 2.7269704341888428
    },
    {
      "auxiliary_loss_clip": 0.01293676,
      "auxiliary_loss_mlp": 0.01031432,
      "balance_loss_clip": 1.05855143,
      "balance_loss_mlp": 1.02360034,
      "epoch": 0.33980640894607106,
      "flos": 21433966479360.0,
      "grad_norm": 3.4446712259318475,
      "language_loss": 0.78205991,
      "learning_rate": 3.0750189193791865e-06,
      "loss": 0.80531096,
      "num_input_tokens_seen": 60798590,
      "step": 2826,
      "time_per_iteration": 3.5523312091827393
    },
    {
      "auxiliary_loss_clip": 0.01248826,
      "auxiliary_loss_mlp": 0.01031671,
      "balance_loss_clip": 1.05568397,
      "balance_loss_mlp": 1.02295053,
      "epoch": 0.33992665183671017,
      "flos": 32490035596800.0,
      "grad_norm": 2.090682257376321,
      "language_loss": 0.70584714,
      "learning_rate": 3.0743619640015203e-06,
      "loss": 0.72865212,
      "num_input_tokens_seen": 60818840,
      "step": 2827,
      "time_per_iteration": 3.6334452629089355
    },
    {
      "auxiliary_loss_clip": 0.01303341,
      "auxiliary_loss_mlp": 0.01032582,
      "balance_loss_clip": 1.05250978,
      "balance_loss_mlp": 1.02445757,
      "epoch": 0.3400468947273492,
      "flos": 17055414495360.0,
      "grad_norm": 1.8422748635004635,
      "language_loss": 0.92346609,
      "learning_rate": 3.073704845643999e-06,
      "loss": 0.94682539,
      "num_input_tokens_seen": 60835965,
      "step": 2828,
      "time_per_iteration": 2.660728693008423
    },
    {
      "auxiliary_loss_clip": 0.01253035,
      "auxiliary_loss_mlp": 0.01033998,
      "balance_loss_clip": 1.05548024,
      "balance_loss_mlp": 1.0251472,
      "epoch": 0.34016713761798834,
      "flos": 16872988296960.0,
      "grad_norm": 2.7938925364671094,
      "language_loss": 0.7860617,
      "learning_rate": 3.0730475644063063e-06,
      "loss": 0.80893207,
      "num_input_tokens_seen": 60851065,
      "step": 2829,
      "time_per_iteration": 2.6082005500793457
    },
    {
      "auxiliary_loss_clip": 0.0120249,
      "auxiliary_loss_mlp": 0.02569882,
      "balance_loss_clip": 1.05074823,
      "balance_loss_mlp": 0.99997818,
      "epoch": 0.34028738050862745,
      "flos": 21907161273600.0,
      "grad_norm": 1.8633370378373313,
      "language_loss": 0.64551264,
      "learning_rate": 3.072390120388151e-06,
      "loss": 0.68323636,
      "num_input_tokens_seen": 60869390,
      "step": 2830,
      "time_per_iteration": 2.699586868286133
    },
    {
      "auxiliary_loss_clip": 0.01247545,
      "auxiliary_loss_mlp": 0.01036073,
      "balance_loss_clip": 1.0557965,
      "balance_loss_mlp": 1.02677464,
      "epoch": 0.3404076233992665,
      "flos": 22746034477440.0,
      "grad_norm": 2.0490007837624358,
      "language_loss": 0.71676713,
      "learning_rate": 3.071732513689267e-06,
      "loss": 0.73960334,
      "num_input_tokens_seen": 60887925,
      "step": 2831,
      "time_per_iteration": 2.6384952068328857
    },
    {
      "auxiliary_loss_clip": 0.01258113,
      "auxiliary_loss_mlp": 0.01028639,
      "balance_loss_clip": 1.06201804,
      "balance_loss_mlp": 1.0199312,
      "epoch": 0.3405278662899056,
      "flos": 17052361839360.0,
      "grad_norm": 2.325105304614131,
      "language_loss": 0.67657918,
      "learning_rate": 3.0710747444094134e-06,
      "loss": 0.69944668,
      "num_input_tokens_seen": 60905955,
      "step": 2832,
      "time_per_iteration": 2.61584210395813
    },
    {
      "auxiliary_loss_clip": 0.01299803,
      "auxiliary_loss_mlp": 0.01032187,
      "balance_loss_clip": 1.0580169,
      "balance_loss_mlp": 1.02380681,
      "epoch": 0.3406481091805447,
      "flos": 42813131783040.0,
      "grad_norm": 1.9671626811567173,
      "language_loss": 0.65230381,
      "learning_rate": 3.070416812648372e-06,
      "loss": 0.67562366,
      "num_input_tokens_seen": 60929405,
      "step": 2833,
      "time_per_iteration": 2.8860719203948975
    },
    {
      "auxiliary_loss_clip": 0.01347546,
      "auxiliary_loss_mlp": 0.01030713,
      "balance_loss_clip": 1.049505,
      "balance_loss_mlp": 1.02189779,
      "epoch": 0.3407683520711838,
      "flos": 26761457917440.0,
      "grad_norm": 2.7036207958834506,
      "language_loss": 0.64696276,
      "learning_rate": 3.069758718505951e-06,
      "loss": 0.67074531,
      "num_input_tokens_seen": 60951145,
      "step": 2834,
      "time_per_iteration": 2.7567684650421143
    },
    {
      "auxiliary_loss_clip": 0.01202627,
      "auxiliary_loss_mlp": 0.01028605,
      "balance_loss_clip": 1.06062388,
      "balance_loss_mlp": 1.02048635,
      "epoch": 0.3408885949618229,
      "flos": 28767643309440.0,
      "grad_norm": 1.9529698311175585,
      "language_loss": 0.80214089,
      "learning_rate": 3.0691004620819836e-06,
      "loss": 0.82445318,
      "num_input_tokens_seen": 60971275,
      "step": 2835,
      "time_per_iteration": 2.6620278358459473
    },
    {
      "auxiliary_loss_clip": 0.01302684,
      "auxiliary_loss_mlp": 0.01008036,
      "balance_loss_clip": 1.02286744,
      "balance_loss_mlp": 1.0065341,
      "epoch": 0.341008837852462,
      "flos": 63576252881280.0,
      "grad_norm": 0.8083668566183136,
      "language_loss": 0.60224771,
      "learning_rate": 3.0684420434763254e-06,
      "loss": 0.62535489,
      "num_input_tokens_seen": 61037460,
      "step": 2836,
      "time_per_iteration": 3.36918568611145
    },
    {
      "auxiliary_loss_clip": 0.01342219,
      "auxiliary_loss_mlp": 0.01029184,
      "balance_loss_clip": 1.05455899,
      "balance_loss_mlp": 1.02140307,
      "epoch": 0.34112908074310105,
      "flos": 20812173120000.0,
      "grad_norm": 1.9520546707809927,
      "language_loss": 0.76898718,
      "learning_rate": 3.06778346278886e-06,
      "loss": 0.79270118,
      "num_input_tokens_seen": 61056295,
      "step": 2837,
      "time_per_iteration": 2.8621397018432617
    },
    {
      "auxiliary_loss_clip": 0.01202164,
      "auxiliary_loss_mlp": 0.01031828,
      "balance_loss_clip": 1.06100523,
      "balance_loss_mlp": 1.02337003,
      "epoch": 0.34124932363374016,
      "flos": 24976446520320.0,
      "grad_norm": 1.6385479522757052,
      "language_loss": 0.78992945,
      "learning_rate": 3.0671247201194906e-06,
      "loss": 0.81226933,
      "num_input_tokens_seen": 61078430,
      "step": 2838,
      "time_per_iteration": 2.655972957611084
    },
    {
      "auxiliary_loss_clip": 0.01349351,
      "auxiliary_loss_mlp": 0.01024821,
      "balance_loss_clip": 1.05456686,
      "balance_loss_mlp": 1.01666141,
      "epoch": 0.3413695665243792,
      "flos": 28402970480640.0,
      "grad_norm": 1.746291028028605,
      "language_loss": 0.75365794,
      "learning_rate": 3.066465815568151e-06,
      "loss": 0.77739966,
      "num_input_tokens_seen": 61099260,
      "step": 2839,
      "time_per_iteration": 2.775007963180542
    },
    {
      "auxiliary_loss_clip": 0.01249987,
      "auxiliary_loss_mlp": 0.01026512,
      "balance_loss_clip": 1.05468464,
      "balance_loss_mlp": 1.01823926,
      "epoch": 0.34148980941501833,
      "flos": 25302012416640.0,
      "grad_norm": 2.371282391575246,
      "language_loss": 0.68747717,
      "learning_rate": 3.0658067492347947e-06,
      "loss": 0.71024215,
      "num_input_tokens_seen": 61121900,
      "step": 2840,
      "time_per_iteration": 2.6964564323425293
    },
    {
      "auxiliary_loss_clip": 0.01488878,
      "auxiliary_loss_mlp": 0.01030524,
      "balance_loss_clip": 1.0485158,
      "balance_loss_mlp": 1.02223873,
      "epoch": 0.34161005230565744,
      "flos": 17530081747200.0,
      "grad_norm": 2.3658192162742435,
      "language_loss": 0.67217636,
      "learning_rate": 3.065147521219402e-06,
      "loss": 0.69737035,
      "num_input_tokens_seen": 61141155,
      "step": 2841,
      "time_per_iteration": 2.7631824016571045
    },
    {
      "auxiliary_loss_clip": 0.01294178,
      "auxiliary_loss_mlp": 0.01028806,
      "balance_loss_clip": 1.05532622,
      "balance_loss_mlp": 1.0211854,
      "epoch": 0.3417302951962965,
      "flos": 43650101566080.0,
      "grad_norm": 1.9523049982021068,
      "language_loss": 0.74436152,
      "learning_rate": 3.064488131621977e-06,
      "loss": 0.76759136,
      "num_input_tokens_seen": 61164480,
      "step": 2842,
      "time_per_iteration": 2.8839781284332275
    },
    {
      "auxiliary_loss_clip": 0.01242044,
      "auxiliary_loss_mlp": 0.01031183,
      "balance_loss_clip": 1.05292928,
      "balance_loss_mlp": 1.02276039,
      "epoch": 0.3418505380869356,
      "flos": 30882207012480.0,
      "grad_norm": 1.7370054330186282,
      "language_loss": 0.73704726,
      "learning_rate": 3.063828580542549e-06,
      "loss": 0.75977945,
      "num_input_tokens_seen": 61185675,
      "step": 2843,
      "time_per_iteration": 2.6840124130249023
    },
    {
      "auxiliary_loss_clip": 0.01296013,
      "auxiliary_loss_mlp": 0.01030074,
      "balance_loss_clip": 1.05410957,
      "balance_loss_mlp": 1.02252758,
      "epoch": 0.3419707809775747,
      "flos": 19463871277440.0,
      "grad_norm": 2.2215956593803714,
      "language_loss": 0.73463941,
      "learning_rate": 3.0631688680811706e-06,
      "loss": 0.75790024,
      "num_input_tokens_seen": 61205300,
      "step": 2844,
      "time_per_iteration": 3.632139205932617
    },
    {
      "auxiliary_loss_clip": 0.01199621,
      "auxiliary_loss_mlp": 0.01029978,
      "balance_loss_clip": 1.05813038,
      "balance_loss_mlp": 1.02138352,
      "epoch": 0.3420910238682138,
      "flos": 28727818104960.0,
      "grad_norm": 2.407393847541024,
      "language_loss": 0.75839919,
      "learning_rate": 3.062508994337921e-06,
      "loss": 0.7806952,
      "num_input_tokens_seen": 61224905,
      "step": 2845,
      "time_per_iteration": 2.6265995502471924
    },
    {
      "auxiliary_loss_clip": 0.01246338,
      "auxiliary_loss_mlp": 0.01028562,
      "balance_loss_clip": 1.0550642,
      "balance_loss_mlp": 1.02016997,
      "epoch": 0.3422112667588529,
      "flos": 21397265758080.0,
      "grad_norm": 2.1942321686988753,
      "language_loss": 0.79553413,
      "learning_rate": 3.0618489594129013e-06,
      "loss": 0.81828308,
      "num_input_tokens_seen": 61243045,
      "step": 2846,
      "time_per_iteration": 3.5991854667663574
    },
    {
      "auxiliary_loss_clip": 0.01355145,
      "auxiliary_loss_mlp": 0.01030289,
      "balance_loss_clip": 1.05739653,
      "balance_loss_mlp": 1.02204561,
      "epoch": 0.342331509649492,
      "flos": 13881450038400.0,
      "grad_norm": 6.168506178877179,
      "language_loss": 0.71447265,
      "learning_rate": 3.061188763406239e-06,
      "loss": 0.73832703,
      "num_input_tokens_seen": 61259190,
      "step": 2847,
      "time_per_iteration": 2.7855207920074463
    },
    {
      "auxiliary_loss_clip": 0.01300141,
      "auxiliary_loss_mlp": 0.01030504,
      "balance_loss_clip": 1.05668342,
      "balance_loss_mlp": 1.02229023,
      "epoch": 0.34245175254013105,
      "flos": 28621450955520.0,
      "grad_norm": 2.8795020495272,
      "language_loss": 0.82370239,
      "learning_rate": 3.060528406418085e-06,
      "loss": 0.84700882,
      "num_input_tokens_seen": 61279040,
      "step": 2848,
      "time_per_iteration": 2.7231194972991943
    },
    {
      "auxiliary_loss_clip": 0.01297538,
      "auxiliary_loss_mlp": 0.01034378,
      "balance_loss_clip": 1.0565294,
      "balance_loss_mlp": 1.02690339,
      "epoch": 0.34257199543077016,
      "flos": 34127058960000.0,
      "grad_norm": 1.718029328777169,
      "language_loss": 0.61779535,
      "learning_rate": 3.0598678885486145e-06,
      "loss": 0.64111447,
      "num_input_tokens_seen": 61301580,
      "step": 2849,
      "time_per_iteration": 2.8023905754089355
    },
    {
      "auxiliary_loss_clip": 0.01256942,
      "auxiliary_loss_mlp": 0.02568102,
      "balance_loss_clip": 1.0489614,
      "balance_loss_mlp": 1.00002313,
      "epoch": 0.34269223832140927,
      "flos": 19974018188160.0,
      "grad_norm": 1.7060818850665407,
      "language_loss": 0.74576873,
      "learning_rate": 3.0592072098980282e-06,
      "loss": 0.78401917,
      "num_input_tokens_seen": 61321240,
      "step": 2850,
      "time_per_iteration": 2.791290760040283
    },
    {
      "auxiliary_loss_clip": 0.01300389,
      "auxiliary_loss_mlp": 0.0104062,
      "balance_loss_clip": 1.05659628,
      "balance_loss_mlp": 1.03215027,
      "epoch": 0.3428124812120483,
      "flos": 27235658292480.0,
      "grad_norm": 3.517453095316934,
      "language_loss": 0.73346376,
      "learning_rate": 3.0585463705665514e-06,
      "loss": 0.75687385,
      "num_input_tokens_seen": 61341615,
      "step": 2851,
      "time_per_iteration": 2.74783992767334
    },
    {
      "auxiliary_loss_clip": 0.0134765,
      "auxiliary_loss_mlp": 0.01027351,
      "balance_loss_clip": 1.05221713,
      "balance_loss_mlp": 1.01920271,
      "epoch": 0.34293272410268744,
      "flos": 24570871079040.0,
      "grad_norm": 2.1911681322027463,
      "language_loss": 0.71083188,
      "learning_rate": 3.0578853706544304e-06,
      "loss": 0.73458183,
      "num_input_tokens_seen": 61359005,
      "step": 2852,
      "time_per_iteration": 3.66605806350708
    },
    {
      "auxiliary_loss_clip": 0.01355384,
      "auxiliary_loss_mlp": 0.02569147,
      "balance_loss_clip": 1.05518365,
      "balance_loss_mlp": 1.00014806,
      "epoch": 0.34305296699332655,
      "flos": 21506865131520.0,
      "grad_norm": 2.5346313080248914,
      "language_loss": 0.65659481,
      "learning_rate": 3.0572242102619404e-06,
      "loss": 0.69584012,
      "num_input_tokens_seen": 61376160,
      "step": 2853,
      "time_per_iteration": 2.7188103199005127
    },
    {
      "auxiliary_loss_clip": 0.01297008,
      "auxiliary_loss_mlp": 0.01030132,
      "balance_loss_clip": 1.05779171,
      "balance_loss_mlp": 1.02271152,
      "epoch": 0.3431732098839656,
      "flos": 24056665931520.0,
      "grad_norm": 1.880083563564818,
      "language_loss": 0.8067044,
      "learning_rate": 3.0565628894893784e-06,
      "loss": 0.82997584,
      "num_input_tokens_seen": 61396795,
      "step": 2854,
      "time_per_iteration": 3.6470959186553955
    },
    {
      "auxiliary_loss_clip": 0.01245735,
      "auxiliary_loss_mlp": 0.01032411,
      "balance_loss_clip": 1.05618906,
      "balance_loss_mlp": 1.0247097,
      "epoch": 0.3432934527746047,
      "flos": 16800879744000.0,
      "grad_norm": 1.7589289009453013,
      "language_loss": 0.7493031,
      "learning_rate": 3.0559014084370655e-06,
      "loss": 0.77208459,
      "num_input_tokens_seen": 61415320,
      "step": 2855,
      "time_per_iteration": 2.6362407207489014
    },
    {
      "auxiliary_loss_clip": 0.01305029,
      "auxiliary_loss_mlp": 0.01033282,
      "balance_loss_clip": 1.05777144,
      "balance_loss_mlp": 1.02461553,
      "epoch": 0.34341369566524377,
      "flos": 23439720908160.0,
      "grad_norm": 1.817655543699375,
      "language_loss": 0.78906167,
      "learning_rate": 3.055239767205349e-06,
      "loss": 0.81244481,
      "num_input_tokens_seen": 61437070,
      "step": 2856,
      "time_per_iteration": 2.6955840587615967
    },
    {
      "auxiliary_loss_clip": 0.01249436,
      "auxiliary_loss_mlp": 0.01030523,
      "balance_loss_clip": 1.06122398,
      "balance_loss_mlp": 1.02222037,
      "epoch": 0.3435339385558829,
      "flos": 17267466435840.0,
      "grad_norm": 1.7596597196699444,
      "language_loss": 0.78241777,
      "learning_rate": 3.054577965894599e-06,
      "loss": 0.80521739,
      "num_input_tokens_seen": 61453215,
      "step": 2857,
      "time_per_iteration": 2.627566337585449
    },
    {
      "auxiliary_loss_clip": 0.01215179,
      "auxiliary_loss_mlp": 0.0103201,
      "balance_loss_clip": 1.06020951,
      "balance_loss_mlp": 1.02393627,
      "epoch": 0.343654181446522,
      "flos": 22199366413440.0,
      "grad_norm": 2.432306442392321,
      "language_loss": 0.70391345,
      "learning_rate": 3.0539160046052094e-06,
      "loss": 0.72638535,
      "num_input_tokens_seen": 61472915,
      "step": 2858,
      "time_per_iteration": 2.67649245262146
    },
    {
      "auxiliary_loss_clip": 0.01296361,
      "auxiliary_loss_mlp": 0.01021229,
      "balance_loss_clip": 1.05562878,
      "balance_loss_mlp": 1.01306343,
      "epoch": 0.34377442433716104,
      "flos": 19901801894400.0,
      "grad_norm": 2.269103280764902,
      "language_loss": 0.70520192,
      "learning_rate": 3.0532538834376003e-06,
      "loss": 0.72837782,
      "num_input_tokens_seen": 61492475,
      "step": 2859,
      "time_per_iteration": 2.71311092376709
    },
    {
      "auxiliary_loss_clip": 0.01253983,
      "auxiliary_loss_mlp": 0.01032498,
      "balance_loss_clip": 1.05755031,
      "balance_loss_mlp": 1.02429628,
      "epoch": 0.34389466722780015,
      "flos": 22197678474240.0,
      "grad_norm": 1.7507060213573395,
      "language_loss": 0.78002769,
      "learning_rate": 3.0525916024922143e-06,
      "loss": 0.80289251,
      "num_input_tokens_seen": 61511660,
      "step": 2860,
      "time_per_iteration": 2.6404237747192383
    },
    {
      "auxiliary_loss_clip": 0.01299193,
      "auxiliary_loss_mlp": 0.01030679,
      "balance_loss_clip": 1.05523217,
      "balance_loss_mlp": 1.02315092,
      "epoch": 0.34401491011843927,
      "flos": 18624567110400.0,
      "grad_norm": 2.7706428263664553,
      "language_loss": 0.83950269,
      "learning_rate": 3.0519291618695193e-06,
      "loss": 0.86280143,
      "num_input_tokens_seen": 61529060,
      "step": 2861,
      "time_per_iteration": 2.705951452255249
    },
    {
      "auxiliary_loss_clip": 0.01343416,
      "auxiliary_loss_mlp": 0.01031021,
      "balance_loss_clip": 1.05178261,
      "balance_loss_mlp": 1.0230782,
      "epoch": 0.3441351530090783,
      "flos": 17858197509120.0,
      "grad_norm": 1.5241441360545758,
      "language_loss": 0.76151222,
      "learning_rate": 3.0512665616700065e-06,
      "loss": 0.78525662,
      "num_input_tokens_seen": 61548125,
      "step": 2862,
      "time_per_iteration": 2.7042758464813232
    },
    {
      "auxiliary_loss_clip": 0.01392442,
      "auxiliary_loss_mlp": 0.01029598,
      "balance_loss_clip": 1.04903865,
      "balance_loss_mlp": 1.02204311,
      "epoch": 0.34425539589971743,
      "flos": 23112754381440.0,
      "grad_norm": 1.8767471386832204,
      "language_loss": 0.89043045,
      "learning_rate": 3.0506038019941933e-06,
      "loss": 0.91465086,
      "num_input_tokens_seen": 61568135,
      "step": 2863,
      "time_per_iteration": 2.850890874862671
    },
    {
      "auxiliary_loss_clip": 0.01344466,
      "auxiliary_loss_mlp": 0.01031129,
      "balance_loss_clip": 1.05469656,
      "balance_loss_mlp": 1.02281404,
      "epoch": 0.34437563879035654,
      "flos": 21907699977600.0,
      "grad_norm": 2.8541793409845844,
      "language_loss": 0.68242025,
      "learning_rate": 3.049940882942617e-06,
      "loss": 0.70617622,
      "num_input_tokens_seen": 61586920,
      "step": 2864,
      "time_per_iteration": 2.7351346015930176
    },
    {
      "auxiliary_loss_clip": 0.01197173,
      "auxiliary_loss_mlp": 0.01029353,
      "balance_loss_clip": 1.0564661,
      "balance_loss_mlp": 1.02171206,
      "epoch": 0.3444958816809956,
      "flos": 23076915586560.0,
      "grad_norm": 1.7515300265296787,
      "language_loss": 0.80567169,
      "learning_rate": 3.0492778046158448e-06,
      "loss": 0.82793695,
      "num_input_tokens_seen": 61608340,
      "step": 2865,
      "time_per_iteration": 2.660358190536499
    },
    {
      "auxiliary_loss_clip": 0.0124273,
      "auxiliary_loss_mlp": 0.01027736,
      "balance_loss_clip": 1.05847704,
      "balance_loss_mlp": 1.02042818,
      "epoch": 0.3446161245716347,
      "flos": 21908633731200.0,
      "grad_norm": 2.283570285180506,
      "language_loss": 0.76906985,
      "learning_rate": 3.0486145671144633e-06,
      "loss": 0.79177451,
      "num_input_tokens_seen": 61628130,
      "step": 2866,
      "time_per_iteration": 2.692950487136841
    },
    {
      "auxiliary_loss_clip": 0.01442197,
      "auxiliary_loss_mlp": 0.01032638,
      "balance_loss_clip": 1.05166888,
      "balance_loss_mlp": 1.02469313,
      "epoch": 0.3447363674622738,
      "flos": 25112834461440.0,
      "grad_norm": 2.5124677619776783,
      "language_loss": 0.76629072,
      "learning_rate": 3.047951170539086e-06,
      "loss": 0.79103911,
      "num_input_tokens_seen": 61647755,
      "step": 2867,
      "time_per_iteration": 2.8293051719665527
    },
    {
      "auxiliary_loss_clip": 0.0135478,
      "auxiliary_loss_mlp": 0.01025822,
      "balance_loss_clip": 1.06390595,
      "balance_loss_mlp": 1.01877367,
      "epoch": 0.3448566103529129,
      "flos": 11984684451840.0,
      "grad_norm": 1.856365461727799,
      "language_loss": 0.84168553,
      "learning_rate": 3.047287614990349e-06,
      "loss": 0.86549151,
      "num_input_tokens_seen": 61665675,
      "step": 2868,
      "time_per_iteration": 2.68662691116333
    },
    {
      "auxiliary_loss_clip": 0.01295198,
      "auxiliary_loss_mlp": 0.01033003,
      "balance_loss_clip": 1.05649853,
      "balance_loss_mlp": 1.02455127,
      "epoch": 0.344976853243552,
      "flos": 40187882465280.0,
      "grad_norm": 2.300300943102262,
      "language_loss": 0.62307304,
      "learning_rate": 3.046623900568914e-06,
      "loss": 0.64635503,
      "num_input_tokens_seen": 61688240,
      "step": 2869,
      "time_per_iteration": 2.7912259101867676
    },
    {
      "auxiliary_loss_clip": 0.01304642,
      "auxiliary_loss_mlp": 0.01027074,
      "balance_loss_clip": 1.05873775,
      "balance_loss_mlp": 1.01861,
      "epoch": 0.34509709613419104,
      "flos": 28723652127360.0,
      "grad_norm": 3.2359562199309555,
      "language_loss": 0.69886613,
      "learning_rate": 3.045960027375465e-06,
      "loss": 0.72218323,
      "num_input_tokens_seen": 61706075,
      "step": 2870,
      "time_per_iteration": 3.9531455039978027
    },
    {
      "auxiliary_loss_clip": 0.0125205,
      "auxiliary_loss_mlp": 0.01032673,
      "balance_loss_clip": 1.05641341,
      "balance_loss_mlp": 1.02435255,
      "epoch": 0.34521733902483015,
      "flos": 29967597982080.0,
      "grad_norm": 3.158265182135427,
      "language_loss": 0.82837856,
      "learning_rate": 3.045295995510711e-06,
      "loss": 0.85122579,
      "num_input_tokens_seen": 61723045,
      "step": 2871,
      "time_per_iteration": 2.7042510509490967
    },
    {
      "auxiliary_loss_clip": 0.01304449,
      "auxiliary_loss_mlp": 0.01035667,
      "balance_loss_clip": 1.0610218,
      "balance_loss_mlp": 1.02863717,
      "epoch": 0.34533758191546926,
      "flos": 27923059843200.0,
      "grad_norm": 2.502218940855827,
      "language_loss": 0.7349174,
      "learning_rate": 3.0446318050753865e-06,
      "loss": 0.75831854,
      "num_input_tokens_seen": 61743525,
      "step": 2872,
      "time_per_iteration": 3.68770432472229
    },
    {
      "auxiliary_loss_clip": 0.01245306,
      "auxiliary_loss_mlp": 0.01025328,
      "balance_loss_clip": 1.05523014,
      "balance_loss_mlp": 1.01739478,
      "epoch": 0.3454578248061083,
      "flos": 27125879351040.0,
      "grad_norm": 2.001408784185906,
      "language_loss": 0.77418417,
      "learning_rate": 3.0439674561702474e-06,
      "loss": 0.79689056,
      "num_input_tokens_seen": 61763025,
      "step": 2873,
      "time_per_iteration": 2.7185468673706055
    },
    {
      "auxiliary_loss_clip": 0.01242911,
      "auxiliary_loss_mlp": 0.010304,
      "balance_loss_clip": 1.05783534,
      "balance_loss_mlp": 1.02299142,
      "epoch": 0.3455780676967474,
      "flos": 19024899166080.0,
      "grad_norm": 2.183170458335052,
      "language_loss": 0.88232958,
      "learning_rate": 3.043302948896076e-06,
      "loss": 0.90506274,
      "num_input_tokens_seen": 61781630,
      "step": 2874,
      "time_per_iteration": 2.6122987270355225
    },
    {
      "auxiliary_loss_clip": 0.01393698,
      "auxiliary_loss_mlp": 0.010284,
      "balance_loss_clip": 1.05439997,
      "balance_loss_mlp": 1.0206573,
      "epoch": 0.34569831058738654,
      "flos": 34496005507200.0,
      "grad_norm": 1.806954175569307,
      "language_loss": 0.60424829,
      "learning_rate": 3.0426382833536756e-06,
      "loss": 0.62846929,
      "num_input_tokens_seen": 61804985,
      "step": 2875,
      "time_per_iteration": 2.9287376403808594
    },
    {
      "auxiliary_loss_clip": 0.01344807,
      "auxiliary_loss_mlp": 0.01037179,
      "balance_loss_clip": 1.05255508,
      "balance_loss_mlp": 1.02866125,
      "epoch": 0.3458185534780256,
      "flos": 31138681098240.0,
      "grad_norm": 3.6052410968422124,
      "language_loss": 0.77794123,
      "learning_rate": 3.041973459643877e-06,
      "loss": 0.80176115,
      "num_input_tokens_seen": 61824440,
      "step": 2876,
      "time_per_iteration": 2.772815465927124
    },
    {
      "auxiliary_loss_clip": 0.0139636,
      "auxiliary_loss_mlp": 0.0102963,
      "balance_loss_clip": 1.05062151,
      "balance_loss_mlp": 1.02150869,
      "epoch": 0.3459387963686647,
      "flos": 32452508862720.0,
      "grad_norm": 2.063082479889827,
      "language_loss": 0.6724571,
      "learning_rate": 3.0413084778675334e-06,
      "loss": 0.69671702,
      "num_input_tokens_seen": 61845690,
      "step": 2877,
      "time_per_iteration": 2.874943494796753
    },
    {
      "auxiliary_loss_clip": 0.01294589,
      "auxiliary_loss_mlp": 0.0256529,
      "balance_loss_clip": 1.05235481,
      "balance_loss_mlp": 1.00005674,
      "epoch": 0.3460590392593038,
      "flos": 24675658030080.0,
      "grad_norm": 3.232096635245014,
      "language_loss": 0.84104788,
      "learning_rate": 3.0406433381255214e-06,
      "loss": 0.87964666,
      "num_input_tokens_seen": 61863725,
      "step": 2878,
      "time_per_iteration": 3.541861057281494
    },
    {
      "auxiliary_loss_clip": 0.01247441,
      "auxiliary_loss_mlp": 0.01031549,
      "balance_loss_clip": 1.06137574,
      "balance_loss_mlp": 1.02358246,
      "epoch": 0.34617928214994287,
      "flos": 18807316531200.0,
      "grad_norm": 3.578625584412755,
      "language_loss": 0.82272971,
      "learning_rate": 3.0399780405187425e-06,
      "loss": 0.84551954,
      "num_input_tokens_seen": 61882720,
      "step": 2879,
      "time_per_iteration": 3.5735225677490234
    },
    {
      "auxiliary_loss_clip": 0.0124609,
      "auxiliary_loss_mlp": 0.01028198,
      "balance_loss_clip": 1.0562067,
      "balance_loss_mlp": 1.02059233,
      "epoch": 0.346299525040582,
      "flos": 24857653265280.0,
      "grad_norm": 1.787639165991668,
      "language_loss": 0.78503442,
      "learning_rate": 3.0393125851481216e-06,
      "loss": 0.80777729,
      "num_input_tokens_seen": 61902595,
      "step": 2880,
      "time_per_iteration": 2.6958537101745605
    },
    {
      "auxiliary_loss_clip": 0.01344586,
      "auxiliary_loss_mlp": 0.01027705,
      "balance_loss_clip": 1.05385017,
      "balance_loss_mlp": 1.02018893,
      "epoch": 0.3464197679312211,
      "flos": 16434914025600.0,
      "grad_norm": 2.3212336242194986,
      "language_loss": 0.86662054,
      "learning_rate": 3.038646972114608e-06,
      "loss": 0.89034349,
      "num_input_tokens_seen": 61918920,
      "step": 2881,
      "time_per_iteration": 2.6876232624053955
    },
    {
      "auxiliary_loss_clip": 0.01342013,
      "auxiliary_loss_mlp": 0.01029353,
      "balance_loss_clip": 1.05608046,
      "balance_loss_mlp": 1.02152658,
      "epoch": 0.34654001082186014,
      "flos": 22382474970240.0,
      "grad_norm": 1.5964690922880143,
      "language_loss": 0.67680919,
      "learning_rate": 3.037981201519174e-06,
      "loss": 0.7005229,
      "num_input_tokens_seen": 61939520,
      "step": 2882,
      "time_per_iteration": 2.733980178833008
    },
    {
      "auxiliary_loss_clip": 0.01250268,
      "auxiliary_loss_mlp": 0.01031755,
      "balance_loss_clip": 1.06183374,
      "balance_loss_mlp": 1.023947,
      "epoch": 0.34666025371249926,
      "flos": 19573901614080.0,
      "grad_norm": 1.8904672104716467,
      "language_loss": 0.70919919,
      "learning_rate": 3.0373152734628175e-06,
      "loss": 0.73201942,
      "num_input_tokens_seen": 61957800,
      "step": 2883,
      "time_per_iteration": 2.613339424133301
    },
    {
      "auxiliary_loss_clip": 0.01241704,
      "auxiliary_loss_mlp": 0.01031207,
      "balance_loss_clip": 1.05365396,
      "balance_loss_mlp": 1.02411723,
      "epoch": 0.34678049660313837,
      "flos": 15267637751040.0,
      "grad_norm": 1.975680279799148,
      "language_loss": 0.76342285,
      "learning_rate": 3.0366491880465584e-06,
      "loss": 0.78615189,
      "num_input_tokens_seen": 61975820,
      "step": 2884,
      "time_per_iteration": 2.612025022506714
    },
    {
      "auxiliary_loss_clip": 0.01204697,
      "auxiliary_loss_mlp": 0.01024879,
      "balance_loss_clip": 1.06276393,
      "balance_loss_mlp": 1.01614118,
      "epoch": 0.3469007394937774,
      "flos": 21181550630400.0,
      "grad_norm": 1.6325619020318005,
      "language_loss": 0.82266486,
      "learning_rate": 3.035982945371443e-06,
      "loss": 0.84496069,
      "num_input_tokens_seen": 61997515,
      "step": 2885,
      "time_per_iteration": 2.700660467147827
    },
    {
      "auxiliary_loss_clip": 0.01308684,
      "auxiliary_loss_mlp": 0.01027681,
      "balance_loss_clip": 1.0584439,
      "balance_loss_mlp": 1.01965535,
      "epoch": 0.34702098238441653,
      "flos": 22375471818240.0,
      "grad_norm": 2.4815026347840563,
      "language_loss": 0.85827732,
      "learning_rate": 3.035316545538537e-06,
      "loss": 0.88164103,
      "num_input_tokens_seen": 62016310,
      "step": 2886,
      "time_per_iteration": 2.6735472679138184
    },
    {
      "auxiliary_loss_clip": 0.01298341,
      "auxiliary_loss_mlp": 0.01035879,
      "balance_loss_clip": 1.06119943,
      "balance_loss_mlp": 1.02814829,
      "epoch": 0.3471412252750556,
      "flos": 22929430343040.0,
      "grad_norm": 2.0153273931868716,
      "language_loss": 0.80154729,
      "learning_rate": 3.034649988648935e-06,
      "loss": 0.82488954,
      "num_input_tokens_seen": 62036075,
      "step": 2887,
      "time_per_iteration": 2.7298243045806885
    },
    {
      "auxiliary_loss_clip": 0.01298858,
      "auxiliary_loss_mlp": 0.01025001,
      "balance_loss_clip": 1.05598998,
      "balance_loss_mlp": 1.01666856,
      "epoch": 0.3472614681656947,
      "flos": 21324259365120.0,
      "grad_norm": 1.9382147174370683,
      "language_loss": 0.80804759,
      "learning_rate": 3.033983274803752e-06,
      "loss": 0.83128619,
      "num_input_tokens_seen": 62055865,
      "step": 2888,
      "time_per_iteration": 2.717029571533203
    },
    {
      "auxiliary_loss_clip": 0.01298808,
      "auxiliary_loss_mlp": 0.01029226,
      "balance_loss_clip": 1.05499673,
      "balance_loss_mlp": 1.02111959,
      "epoch": 0.3473817110563338,
      "flos": 23475739271040.0,
      "grad_norm": 3.141783589045261,
      "language_loss": 0.72760749,
      "learning_rate": 3.0333164041041283e-06,
      "loss": 0.75088787,
      "num_input_tokens_seen": 62072180,
      "step": 2889,
      "time_per_iteration": 2.7648487091064453
    },
    {
      "auxiliary_loss_clip": 0.01452178,
      "auxiliary_loss_mlp": 0.01033539,
      "balance_loss_clip": 1.05425119,
      "balance_loss_mlp": 1.02588856,
      "epoch": 0.34750195394697286,
      "flos": 22346025644160.0,
      "grad_norm": 1.7387916483392476,
      "language_loss": 0.72238022,
      "learning_rate": 3.032649376651228e-06,
      "loss": 0.74723732,
      "num_input_tokens_seen": 62091600,
      "step": 2890,
      "time_per_iteration": 2.8098206520080566
    },
    {
      "auxiliary_loss_clip": 0.01349666,
      "auxiliary_loss_mlp": 0.01031479,
      "balance_loss_clip": 1.05611324,
      "balance_loss_mlp": 1.02340269,
      "epoch": 0.347622196837612,
      "flos": 29095004885760.0,
      "grad_norm": 1.630708424129234,
      "language_loss": 0.76302791,
      "learning_rate": 3.031982192546238e-06,
      "loss": 0.78683937,
      "num_input_tokens_seen": 62114695,
      "step": 2891,
      "time_per_iteration": 2.811267375946045
    },
    {
      "auxiliary_loss_clip": 0.01250344,
      "auxiliary_loss_mlp": 0.01028581,
      "balance_loss_clip": 1.05764449,
      "balance_loss_mlp": 1.02099967,
      "epoch": 0.3477424397282511,
      "flos": 22455732758400.0,
      "grad_norm": 2.1834125088658154,
      "language_loss": 0.95394766,
      "learning_rate": 3.0313148518903696e-06,
      "loss": 0.9767369,
      "num_input_tokens_seen": 62134520,
      "step": 2892,
      "time_per_iteration": 2.675196409225464
    },
    {
      "auxiliary_loss_clip": 0.0130872,
      "auxiliary_loss_mlp": 0.01031222,
      "balance_loss_clip": 1.06150341,
      "balance_loss_mlp": 1.02349138,
      "epoch": 0.34786268261889014,
      "flos": 15778790242560.0,
      "grad_norm": 3.5922252334808937,
      "language_loss": 0.80942374,
      "learning_rate": 3.030647354784859e-06,
      "loss": 0.83282316,
      "num_input_tokens_seen": 62151560,
      "step": 2893,
      "time_per_iteration": 2.679007053375244
    },
    {
      "auxiliary_loss_clip": 0.01340159,
      "auxiliary_loss_mlp": 0.01026233,
      "balance_loss_clip": 1.05248487,
      "balance_loss_mlp": 1.01864588,
      "epoch": 0.34798292550952925,
      "flos": 20777627214720.0,
      "grad_norm": 2.0219135173934824,
      "language_loss": 0.77761233,
      "learning_rate": 3.029979701330964e-06,
      "loss": 0.80127627,
      "num_input_tokens_seen": 62170985,
      "step": 2894,
      "time_per_iteration": 2.77066969871521
    },
    {
      "auxiliary_loss_clip": 0.01303838,
      "auxiliary_loss_mlp": 0.01029298,
      "balance_loss_clip": 1.05746698,
      "balance_loss_mlp": 1.02122784,
      "epoch": 0.34810316840016836,
      "flos": 19937820257280.0,
      "grad_norm": 2.0941642795206508,
      "language_loss": 0.80150127,
      "learning_rate": 3.029311891629966e-06,
      "loss": 0.82483256,
      "num_input_tokens_seen": 62189440,
      "step": 2895,
      "time_per_iteration": 2.6482183933258057
    },
    {
      "auxiliary_loss_clip": 0.01297254,
      "auxiliary_loss_mlp": 0.0102529,
      "balance_loss_clip": 1.05899191,
      "balance_loss_mlp": 1.01720142,
      "epoch": 0.3482234112908074,
      "flos": 23623296341760.0,
      "grad_norm": 2.144632102169098,
      "language_loss": 0.74256647,
      "learning_rate": 3.0286439257831744e-06,
      "loss": 0.76579189,
      "num_input_tokens_seen": 62208910,
      "step": 2896,
      "time_per_iteration": 2.7423503398895264
    },
    {
      "auxiliary_loss_clip": 0.01203669,
      "auxiliary_loss_mlp": 0.01036787,
      "balance_loss_clip": 1.06010556,
      "balance_loss_mlp": 1.02841878,
      "epoch": 0.3483436541814465,
      "flos": 23986712194560.0,
      "grad_norm": 2.5929317651061594,
      "language_loss": 0.7171495,
      "learning_rate": 3.0279758038919156e-06,
      "loss": 0.73955411,
      "num_input_tokens_seen": 62227135,
      "step": 2897,
      "time_per_iteration": 3.6103627681732178
    },
    {
      "auxiliary_loss_clip": 0.01251128,
      "auxiliary_loss_mlp": 0.01027516,
      "balance_loss_clip": 1.06073165,
      "balance_loss_mlp": 1.01942182,
      "epoch": 0.34846389707208564,
      "flos": 22638338524800.0,
      "grad_norm": 1.9585543513408568,
      "language_loss": 0.78764915,
      "learning_rate": 3.0273075260575455e-06,
      "loss": 0.81043565,
      "num_input_tokens_seen": 62246035,
      "step": 2898,
      "time_per_iteration": 3.5402910709381104
    },
    {
      "auxiliary_loss_clip": 0.01304913,
      "auxiliary_loss_mlp": 0.01026388,
      "balance_loss_clip": 1.05710649,
      "balance_loss_mlp": 1.01770711,
      "epoch": 0.3485841399627247,
      "flos": 21792857218560.0,
      "grad_norm": 1.9284678994522868,
      "language_loss": 0.81073809,
      "learning_rate": 3.0266390923814396e-06,
      "loss": 0.83405107,
      "num_input_tokens_seen": 62264095,
      "step": 2899,
      "time_per_iteration": 2.7446138858795166
    },
    {
      "auxiliary_loss_clip": 0.01312145,
      "auxiliary_loss_mlp": 0.01028438,
      "balance_loss_clip": 1.06516707,
      "balance_loss_mlp": 1.02009296,
      "epoch": 0.3487043828533638,
      "flos": 17019036996480.0,
      "grad_norm": 1.8776095090856384,
      "language_loss": 0.81982636,
      "learning_rate": 3.0259705029650008e-06,
      "loss": 0.84323215,
      "num_input_tokens_seen": 62282025,
      "step": 2900,
      "time_per_iteration": 2.6701841354370117
    },
    {
      "auxiliary_loss_clip": 0.01254496,
      "auxiliary_loss_mlp": 0.01027235,
      "balance_loss_clip": 1.0598371,
      "balance_loss_mlp": 1.01977229,
      "epoch": 0.34882462574400286,
      "flos": 22601135013120.0,
      "grad_norm": 1.6488953251531697,
      "language_loss": 0.72806543,
      "learning_rate": 3.025301757909652e-06,
      "loss": 0.75088274,
      "num_input_tokens_seen": 62302220,
      "step": 2901,
      "time_per_iteration": 2.7365505695343018
    },
    {
      "auxiliary_loss_clip": 0.01345338,
      "auxiliary_loss_mlp": 0.02575989,
      "balance_loss_clip": 1.05547428,
      "balance_loss_mlp": 1.00009525,
      "epoch": 0.34894486863464197,
      "flos": 29861518141440.0,
      "grad_norm": 1.6366392487946995,
      "language_loss": 0.80826539,
      "learning_rate": 3.024632857316842e-06,
      "loss": 0.84747863,
      "num_input_tokens_seen": 62323535,
      "step": 2902,
      "time_per_iteration": 2.774203062057495
    },
    {
      "auxiliary_loss_clip": 0.01250702,
      "auxiliary_loss_mlp": 0.0102935,
      "balance_loss_clip": 1.05986404,
      "balance_loss_mlp": 1.02059364,
      "epoch": 0.3490651115252811,
      "flos": 22122265870080.0,
      "grad_norm": 1.7389674195137272,
      "language_loss": 0.77799773,
      "learning_rate": 3.0239638012880412e-06,
      "loss": 0.8007983,
      "num_input_tokens_seen": 62343430,
      "step": 2903,
      "time_per_iteration": 2.7017650604248047
    },
    {
      "auxiliary_loss_clip": 0.01391587,
      "auxiliary_loss_mlp": 0.0103023,
      "balance_loss_clip": 1.05099428,
      "balance_loss_mlp": 1.02164733,
      "epoch": 0.34918535441592014,
      "flos": 12676682943360.0,
      "grad_norm": 2.922504252998511,
      "language_loss": 0.81347024,
      "learning_rate": 3.0232945899247466e-06,
      "loss": 0.83768845,
      "num_input_tokens_seen": 62360365,
      "step": 2904,
      "time_per_iteration": 3.660767078399658
    },
    {
      "auxiliary_loss_clip": 0.01255085,
      "auxiliary_loss_mlp": 0.01029683,
      "balance_loss_clip": 1.05915058,
      "balance_loss_mlp": 1.02162468,
      "epoch": 0.34930559730655925,
      "flos": 23185617120000.0,
      "grad_norm": 2.2548856985989953,
      "language_loss": 0.77468705,
      "learning_rate": 3.022625223328476e-06,
      "loss": 0.79753476,
      "num_input_tokens_seen": 62382105,
      "step": 2905,
      "time_per_iteration": 2.7253222465515137
    },
    {
      "auxiliary_loss_clip": 0.01169996,
      "auxiliary_loss_mlp": 0.01029358,
      "balance_loss_clip": 1.06373179,
      "balance_loss_mlp": 1.02045918,
      "epoch": 0.34942584019719836,
      "flos": 22855023319680.0,
      "grad_norm": 1.37707164140349,
      "language_loss": 0.69126767,
      "learning_rate": 3.0219557016007723e-06,
      "loss": 0.71326125,
      "num_input_tokens_seen": 62402235,
      "step": 2906,
      "time_per_iteration": 3.599968194961548
    },
    {
      "auxiliary_loss_clip": 0.01254463,
      "auxiliary_loss_mlp": 0.01029837,
      "balance_loss_clip": 1.0623858,
      "balance_loss_mlp": 1.02144432,
      "epoch": 0.3495460830878374,
      "flos": 24426043441920.0,
      "grad_norm": 2.210288308125198,
      "language_loss": 0.7000367,
      "learning_rate": 3.021286024843202e-06,
      "loss": 0.72287971,
      "num_input_tokens_seen": 62420430,
      "step": 2907,
      "time_per_iteration": 2.6889326572418213
    },
    {
      "auxiliary_loss_clip": 0.01098174,
      "auxiliary_loss_mlp": 0.0100536,
      "balance_loss_clip": 1.02917361,
      "balance_loss_mlp": 1.00400662,
      "epoch": 0.3496663259784765,
      "flos": 70008749389440.0,
      "grad_norm": 1.0864220020905129,
      "language_loss": 0.64809883,
      "learning_rate": 3.0206161931573526e-06,
      "loss": 0.66913414,
      "num_input_tokens_seen": 62472980,
      "step": 2908,
      "time_per_iteration": 3.127211809158325
    },
    {
      "auxiliary_loss_clip": 0.01296771,
      "auxiliary_loss_mlp": 0.01029911,
      "balance_loss_clip": 1.05397987,
      "balance_loss_mlp": 1.02147746,
      "epoch": 0.34978656886911563,
      "flos": 28692805322880.0,
      "grad_norm": 1.8995439473427054,
      "language_loss": 0.93223166,
      "learning_rate": 3.0199462066448388e-06,
      "loss": 0.95549846,
      "num_input_tokens_seen": 62495175,
      "step": 2909,
      "time_per_iteration": 2.747288942337036
    },
    {
      "auxiliary_loss_clip": 0.0125543,
      "auxiliary_loss_mlp": 0.01034507,
      "balance_loss_clip": 1.06257486,
      "balance_loss_mlp": 1.02619171,
      "epoch": 0.3499068117597547,
      "flos": 21142156389120.0,
      "grad_norm": 2.264444772207905,
      "language_loss": 0.68924975,
      "learning_rate": 3.019276065407296e-06,
      "loss": 0.71214914,
      "num_input_tokens_seen": 62514295,
      "step": 2910,
      "time_per_iteration": 2.674916982650757
    },
    {
      "auxiliary_loss_clip": 0.01395562,
      "auxiliary_loss_mlp": 0.0103483,
      "balance_loss_clip": 1.05146444,
      "balance_loss_mlp": 1.02618098,
      "epoch": 0.3500270546503938,
      "flos": 22782699285120.0,
      "grad_norm": 1.7542974990852325,
      "language_loss": 0.80792344,
      "learning_rate": 3.018605769546385e-06,
      "loss": 0.83222735,
      "num_input_tokens_seen": 62534850,
      "step": 2911,
      "time_per_iteration": 2.838106155395508
    },
    {
      "auxiliary_loss_clip": 0.0125108,
      "auxiliary_loss_mlp": 0.0103038,
      "balance_loss_clip": 1.05665517,
      "balance_loss_mlp": 1.02198124,
      "epoch": 0.3501472975410329,
      "flos": 22894058424960.0,
      "grad_norm": 2.221595207219945,
      "language_loss": 0.79188776,
      "learning_rate": 3.017935319163788e-06,
      "loss": 0.81470239,
      "num_input_tokens_seen": 62553810,
      "step": 2912,
      "time_per_iteration": 2.6216273307800293
    },
    {
      "auxiliary_loss_clip": 0.01254419,
      "auxiliary_loss_mlp": 0.01031761,
      "balance_loss_clip": 1.0605607,
      "balance_loss_mlp": 1.02278483,
      "epoch": 0.35026754043167196,
      "flos": 25446588658560.0,
      "grad_norm": 1.8832301096803592,
      "language_loss": 0.70397651,
      "learning_rate": 3.017264714361213e-06,
      "loss": 0.72683835,
      "num_input_tokens_seen": 62573460,
      "step": 2913,
      "time_per_iteration": 2.8792366981506348
    },
    {
      "auxiliary_loss_clip": 0.0129837,
      "auxiliary_loss_mlp": 0.02575759,
      "balance_loss_clip": 1.05639052,
      "balance_loss_mlp": 1.00016546,
      "epoch": 0.3503877833223111,
      "flos": 19573757959680.0,
      "grad_norm": 2.199930549883388,
      "language_loss": 0.82045829,
      "learning_rate": 3.016593955240389e-06,
      "loss": 0.85919958,
      "num_input_tokens_seen": 62592150,
      "step": 2914,
      "time_per_iteration": 2.713514566421509
    },
    {
      "auxiliary_loss_clip": 0.01145843,
      "auxiliary_loss_mlp": 0.00997123,
      "balance_loss_clip": 1.02746487,
      "balance_loss_mlp": 0.99580592,
      "epoch": 0.3505080262129502,
      "flos": 65072075880960.0,
      "grad_norm": 0.8217401868922278,
      "language_loss": 0.63616377,
      "learning_rate": 3.015923041903071e-06,
      "loss": 0.65759349,
      "num_input_tokens_seen": 62658275,
      "step": 2915,
      "time_per_iteration": 3.2971999645233154
    },
    {
      "auxiliary_loss_clip": 0.01251588,
      "auxiliary_loss_mlp": 0.01030242,
      "balance_loss_clip": 1.06070924,
      "balance_loss_mlp": 1.02246416,
      "epoch": 0.35062826910358924,
      "flos": 29314562768640.0,
      "grad_norm": 2.2313131008273985,
      "language_loss": 0.83377379,
      "learning_rate": 3.0152519744510347e-06,
      "loss": 0.85659212,
      "num_input_tokens_seen": 62678075,
      "step": 2916,
      "time_per_iteration": 2.7556324005126953
    },
    {
      "auxiliary_loss_clip": 0.01348038,
      "auxiliary_loss_mlp": 0.01034165,
      "balance_loss_clip": 1.05224216,
      "balance_loss_mlp": 1.02595711,
      "epoch": 0.35074851199422835,
      "flos": 23987717775360.0,
      "grad_norm": 3.9064656588883357,
      "language_loss": 0.82569265,
      "learning_rate": 3.014580752986081e-06,
      "loss": 0.84951472,
      "num_input_tokens_seen": 62696950,
      "step": 2917,
      "time_per_iteration": 2.7637572288513184
    },
    {
      "auxiliary_loss_clip": 0.0139969,
      "auxiliary_loss_mlp": 0.0104078,
      "balance_loss_clip": 1.05637908,
      "balance_loss_mlp": 1.03222096,
      "epoch": 0.3508687548848674,
      "flos": 15224436668160.0,
      "grad_norm": 2.363804104669686,
      "language_loss": 0.78995574,
      "learning_rate": 3.0139093776100345e-06,
      "loss": 0.81436038,
      "num_input_tokens_seen": 62713540,
      "step": 2918,
      "time_per_iteration": 2.792764186859131
    },
    {
      "auxiliary_loss_clip": 0.01198148,
      "auxiliary_loss_mlp": 0.0103157,
      "balance_loss_clip": 1.05866432,
      "balance_loss_mlp": 1.02332699,
      "epoch": 0.3509889977755065,
      "flos": 21361750185600.0,
      "grad_norm": 1.6567816407604703,
      "language_loss": 0.75324911,
      "learning_rate": 3.013237848424741e-06,
      "loss": 0.77554625,
      "num_input_tokens_seen": 62732925,
      "step": 2919,
      "time_per_iteration": 2.667980432510376
    },
    {
      "auxiliary_loss_clip": 0.01304289,
      "auxiliary_loss_mlp": 0.01032801,
      "balance_loss_clip": 1.0607357,
      "balance_loss_mlp": 1.02467704,
      "epoch": 0.35110924066614563,
      "flos": 19135360465920.0,
      "grad_norm": 2.3453746823652937,
      "language_loss": 0.75217903,
      "learning_rate": 3.012566165532072e-06,
      "loss": 0.77554989,
      "num_input_tokens_seen": 62751715,
      "step": 2920,
      "time_per_iteration": 2.7625412940979004
    },
    {
      "auxiliary_loss_clip": 0.01360942,
      "auxiliary_loss_mlp": 0.01034964,
      "balance_loss_clip": 1.05101812,
      "balance_loss_mlp": 1.02657723,
      "epoch": 0.3512294835567847,
      "flos": 21980885938560.0,
      "grad_norm": 2.3020050930763025,
      "language_loss": 0.76573831,
      "learning_rate": 3.0118943290339207e-06,
      "loss": 0.78969741,
      "num_input_tokens_seen": 62771925,
      "step": 2921,
      "time_per_iteration": 2.8298134803771973
    },
    {
      "auxiliary_loss_clip": 0.01339494,
      "auxiliary_loss_mlp": 0.01030906,
      "balance_loss_clip": 1.04976106,
      "balance_loss_mlp": 1.02159619,
      "epoch": 0.3513497264474238,
      "flos": 17817294896640.0,
      "grad_norm": 2.137480331227174,
      "language_loss": 0.68013978,
      "learning_rate": 3.011222339032204e-06,
      "loss": 0.70384377,
      "num_input_tokens_seen": 62790075,
      "step": 2922,
      "time_per_iteration": 3.5904178619384766
    },
    {
      "auxiliary_loss_clip": 0.01200829,
      "auxiliary_loss_mlp": 0.01032088,
      "balance_loss_clip": 1.06027532,
      "balance_loss_mlp": 1.02383888,
      "epoch": 0.3514699693380629,
      "flos": 26943417239040.0,
      "grad_norm": 1.9085507078779673,
      "language_loss": 0.69318336,
      "learning_rate": 3.0105501956288626e-06,
      "loss": 0.71551251,
      "num_input_tokens_seen": 62810545,
      "step": 2923,
      "time_per_iteration": 2.6774535179138184
    },
    {
      "auxiliary_loss_clip": 0.01255439,
      "auxiliary_loss_mlp": 0.01032625,
      "balance_loss_clip": 1.05775332,
      "balance_loss_mlp": 1.02380943,
      "epoch": 0.35159021222870196,
      "flos": 15267565923840.0,
      "grad_norm": 2.0400598138346617,
      "language_loss": 0.72812152,
      "learning_rate": 3.0098778989258602e-06,
      "loss": 0.75100219,
      "num_input_tokens_seen": 62829155,
      "step": 2924,
      "time_per_iteration": 3.623885154724121
    },
    {
      "auxiliary_loss_clip": 0.01350631,
      "auxiliary_loss_mlp": 0.01035948,
      "balance_loss_clip": 1.05675507,
      "balance_loss_mlp": 1.02763939,
      "epoch": 0.35171045511934107,
      "flos": 13984154000640.0,
      "grad_norm": 2.8782503288087855,
      "language_loss": 0.88061786,
      "learning_rate": 3.009205449025183e-06,
      "loss": 0.90448362,
      "num_input_tokens_seen": 62845350,
      "step": 2925,
      "time_per_iteration": 2.7012813091278076
    },
    {
      "auxiliary_loss_clip": 0.01255185,
      "auxiliary_loss_mlp": 0.01032709,
      "balance_loss_clip": 1.05134571,
      "balance_loss_mlp": 1.02482939,
      "epoch": 0.3518306980099802,
      "flos": 14283434119680.0,
      "grad_norm": 1.9603742810008622,
      "language_loss": 0.63394737,
      "learning_rate": 3.008532846028842e-06,
      "loss": 0.65682626,
      "num_input_tokens_seen": 62862110,
      "step": 2926,
      "time_per_iteration": 2.679356098175049
    },
    {
      "auxiliary_loss_clip": 0.01201636,
      "auxiliary_loss_mlp": 0.01032599,
      "balance_loss_clip": 1.05974245,
      "balance_loss_mlp": 1.02376509,
      "epoch": 0.35195094090061924,
      "flos": 27052872958080.0,
      "grad_norm": 2.519872987984773,
      "language_loss": 0.71885729,
      "learning_rate": 3.0078600900388694e-06,
      "loss": 0.74119961,
      "num_input_tokens_seen": 62882415,
      "step": 2927,
      "time_per_iteration": 2.7495343685150146
    },
    {
      "auxiliary_loss_clip": 0.0133947,
      "auxiliary_loss_mlp": 0.01038234,
      "balance_loss_clip": 1.04775882,
      "balance_loss_mlp": 1.02905464,
      "epoch": 0.35207118379125835,
      "flos": 25629266252160.0,
      "grad_norm": 2.3858668801231513,
      "language_loss": 0.7390005,
      "learning_rate": 3.007187181157323e-06,
      "loss": 0.76277751,
      "num_input_tokens_seen": 62902425,
      "step": 2928,
      "time_per_iteration": 2.7449352741241455
    },
    {
      "auxiliary_loss_clip": 0.01431698,
      "auxiliary_loss_mlp": 0.01036354,
      "balance_loss_clip": 1.04583287,
      "balance_loss_mlp": 1.02801561,
      "epoch": 0.35219142668189746,
      "flos": 18004713085440.0,
      "grad_norm": 2.4188780185297794,
      "language_loss": 0.67821825,
      "learning_rate": 3.006514119486282e-06,
      "loss": 0.70289874,
      "num_input_tokens_seen": 62919255,
      "step": 2929,
      "time_per_iteration": 3.7007052898406982
    },
    {
      "auxiliary_loss_clip": 0.01341614,
      "auxiliary_loss_mlp": 0.0103434,
      "balance_loss_clip": 1.05168009,
      "balance_loss_mlp": 1.02597773,
      "epoch": 0.3523116695725365,
      "flos": 14028109269120.0,
      "grad_norm": 1.806421691674182,
      "language_loss": 0.69705331,
      "learning_rate": 3.005840905127849e-06,
      "loss": 0.7208128,
      "num_input_tokens_seen": 62936160,
      "step": 2930,
      "time_per_iteration": 2.749537229537964
    },
    {
      "auxiliary_loss_clip": 0.01201961,
      "auxiliary_loss_mlp": 0.01037396,
      "balance_loss_clip": 1.0615716,
      "balance_loss_mlp": 1.02931905,
      "epoch": 0.3524319124631756,
      "flos": 21433966479360.0,
      "grad_norm": 2.26353440054525,
      "language_loss": 0.87086499,
      "learning_rate": 3.0051675381841516e-06,
      "loss": 0.89325857,
      "num_input_tokens_seen": 62953470,
      "step": 2931,
      "time_per_iteration": 2.639583110809326
    },
    {
      "auxiliary_loss_clip": 0.01403831,
      "auxiliary_loss_mlp": 0.02573991,
      "balance_loss_clip": 1.04534459,
      "balance_loss_mlp": 1.00021839,
      "epoch": 0.3525521553538147,
      "flos": 26322773114880.0,
      "grad_norm": 1.7621734849814825,
      "language_loss": 0.76974648,
      "learning_rate": 3.0044940187573363e-06,
      "loss": 0.80952466,
      "num_input_tokens_seen": 62974480,
      "step": 2932,
      "time_per_iteration": 3.8181819915771484
    },
    {
      "auxiliary_loss_clip": 0.01249345,
      "auxiliary_loss_mlp": 0.01033008,
      "balance_loss_clip": 1.05483556,
      "balance_loss_mlp": 1.02474129,
      "epoch": 0.3526723982444538,
      "flos": 21543314457600.0,
      "grad_norm": 2.3978450971591454,
      "language_loss": 0.64934766,
      "learning_rate": 3.003820346949578e-06,
      "loss": 0.67217118,
      "num_input_tokens_seen": 62992560,
      "step": 2933,
      "time_per_iteration": 2.772486448287964
    },
    {
      "auxiliary_loss_clip": 0.011986,
      "auxiliary_loss_mlp": 0.01034771,
      "balance_loss_clip": 1.05647361,
      "balance_loss_mlp": 1.02656937,
      "epoch": 0.3527926411350929,
      "flos": 23733649900800.0,
      "grad_norm": 2.214218708568085,
      "language_loss": 0.79693305,
      "learning_rate": 3.003146522863071e-06,
      "loss": 0.8192668,
      "num_input_tokens_seen": 63013445,
      "step": 2934,
      "time_per_iteration": 2.5668749809265137
    },
    {
      "auxiliary_loss_clip": 0.01298729,
      "auxiliary_loss_mlp": 0.01033571,
      "balance_loss_clip": 1.05848837,
      "balance_loss_mlp": 1.02569723,
      "epoch": 0.35291288402573195,
      "flos": 30445461544320.0,
      "grad_norm": 2.244986959263357,
      "language_loss": 0.8617748,
      "learning_rate": 3.0024725466000345e-06,
      "loss": 0.88509774,
      "num_input_tokens_seen": 63033400,
      "step": 2935,
      "time_per_iteration": 2.7575132846832275
    },
    {
      "auxiliary_loss_clip": 0.01249105,
      "auxiliary_loss_mlp": 0.01026411,
      "balance_loss_clip": 1.05937684,
      "balance_loss_mlp": 1.01819754,
      "epoch": 0.35303312691637107,
      "flos": 23112179763840.0,
      "grad_norm": 1.7099310374129788,
      "language_loss": 0.7855444,
      "learning_rate": 3.0017984182627087e-06,
      "loss": 0.8082996,
      "num_input_tokens_seen": 63052725,
      "step": 2936,
      "time_per_iteration": 2.625941514968872
    },
    {
      "auxiliary_loss_clip": 0.01349204,
      "auxiliary_loss_mlp": 0.02574059,
      "balance_loss_clip": 1.05251586,
      "balance_loss_mlp": 1.00025392,
      "epoch": 0.3531533698070102,
      "flos": 21835699165440.0,
      "grad_norm": 1.9818999003388524,
      "language_loss": 0.82523179,
      "learning_rate": 3.00112413795336e-06,
      "loss": 0.86446428,
      "num_input_tokens_seen": 63072560,
      "step": 2937,
      "time_per_iteration": 2.746152400970459
    },
    {
      "auxiliary_loss_clip": 0.01296868,
      "auxiliary_loss_mlp": 0.01028187,
      "balance_loss_clip": 1.05103421,
      "balance_loss_mlp": 1.01951408,
      "epoch": 0.35327361269764923,
      "flos": 15778969810560.0,
      "grad_norm": 2.1066074923463294,
      "language_loss": 0.8044014,
      "learning_rate": 3.000449705774275e-06,
      "loss": 0.82765198,
      "num_input_tokens_seen": 63090800,
      "step": 2938,
      "time_per_iteration": 2.6534218788146973
    },
    {
      "auxiliary_loss_clip": 0.01249986,
      "auxiliary_loss_mlp": 0.01026797,
      "balance_loss_clip": 1.058918,
      "balance_loss_mlp": 1.01864946,
      "epoch": 0.35339385558828834,
      "flos": 22090413484800.0,
      "grad_norm": 1.9000180673906124,
      "language_loss": 0.71594846,
      "learning_rate": 2.9997751218277654e-06,
      "loss": 0.73871624,
      "num_input_tokens_seen": 63108955,
      "step": 2939,
      "time_per_iteration": 2.681589365005493
    },
    {
      "auxiliary_loss_clip": 0.01201356,
      "auxiliary_loss_mlp": 0.01027755,
      "balance_loss_clip": 1.06019497,
      "balance_loss_mlp": 1.01930285,
      "epoch": 0.35351409847892745,
      "flos": 24165008328960.0,
      "grad_norm": 2.1233293421578634,
      "language_loss": 0.78059334,
      "learning_rate": 2.999100386216166e-06,
      "loss": 0.80288446,
      "num_input_tokens_seen": 63127895,
      "step": 2940,
      "time_per_iteration": 2.6055612564086914
    },
    {
      "auxiliary_loss_clip": 0.0129766,
      "auxiliary_loss_mlp": 0.01029761,
      "balance_loss_clip": 1.05607712,
      "balance_loss_mlp": 1.0215292,
      "epoch": 0.3536343413695665,
      "flos": 27052298340480.0,
      "grad_norm": 1.731475548266655,
      "language_loss": 0.74133748,
      "learning_rate": 2.998425499041831e-06,
      "loss": 0.76461172,
      "num_input_tokens_seen": 63148410,
      "step": 2941,
      "time_per_iteration": 2.7447891235351562
    },
    {
      "auxiliary_loss_clip": 0.01140722,
      "auxiliary_loss_mlp": 0.01010327,
      "balance_loss_clip": 1.02226949,
      "balance_loss_mlp": 1.00913513,
      "epoch": 0.3537545842602056,
      "flos": 65991066370560.0,
      "grad_norm": 1.2735930314002386,
      "language_loss": 0.64562643,
      "learning_rate": 2.997750460407142e-06,
      "loss": 0.66713691,
      "num_input_tokens_seen": 63209765,
      "step": 2942,
      "time_per_iteration": 3.2823379039764404
    },
    {
      "auxiliary_loss_clip": 0.01350039,
      "auxiliary_loss_mlp": 0.01026124,
      "balance_loss_clip": 1.05077016,
      "balance_loss_mlp": 1.0169096,
      "epoch": 0.35387482715084473,
      "flos": 18436897526400.0,
      "grad_norm": 2.339887514320975,
      "language_loss": 0.70114267,
      "learning_rate": 2.997075270414501e-06,
      "loss": 0.7249043,
      "num_input_tokens_seen": 63226980,
      "step": 2943,
      "time_per_iteration": 2.8193156719207764
    },
    {
      "auxiliary_loss_clip": 0.01195145,
      "auxiliary_loss_mlp": 0.01008533,
      "balance_loss_clip": 1.02402544,
      "balance_loss_mlp": 1.00731683,
      "epoch": 0.3539950700414838,
      "flos": 65588579498880.0,
      "grad_norm": 0.7053579189208251,
      "language_loss": 0.57726681,
      "learning_rate": 2.9963999291663347e-06,
      "loss": 0.5993036,
      "num_input_tokens_seen": 63292760,
      "step": 2944,
      "time_per_iteration": 3.260542869567871
    },
    {
      "auxiliary_loss_clip": 0.01390379,
      "auxiliary_loss_mlp": 0.01032222,
      "balance_loss_clip": 1.05393124,
      "balance_loss_mlp": 1.02450335,
      "epoch": 0.3541153129321229,
      "flos": 20521655919360.0,
      "grad_norm": 2.361853224914991,
      "language_loss": 0.73921335,
      "learning_rate": 2.9957244367650915e-06,
      "loss": 0.7634393,
      "num_input_tokens_seen": 63309005,
      "step": 2945,
      "time_per_iteration": 2.759427785873413
    },
    {
      "auxiliary_loss_clip": 0.0140134,
      "auxiliary_loss_mlp": 0.0103167,
      "balance_loss_clip": 1.05495906,
      "balance_loss_mlp": 1.02385044,
      "epoch": 0.354235555822762,
      "flos": 19573578391680.0,
      "grad_norm": 1.7235788882184087,
      "language_loss": 0.8387177,
      "learning_rate": 2.9950487933132425e-06,
      "loss": 0.86304784,
      "num_input_tokens_seen": 63326420,
      "step": 2946,
      "time_per_iteration": 2.987109899520874
    },
    {
      "auxiliary_loss_clip": 0.01256034,
      "auxiliary_loss_mlp": 0.01033431,
      "balance_loss_clip": 1.05944991,
      "balance_loss_mlp": 1.02521491,
      "epoch": 0.35435579871340106,
      "flos": 20777268078720.0,
      "grad_norm": 2.4431819094064062,
      "language_loss": 0.7193771,
      "learning_rate": 2.994372998913283e-06,
      "loss": 0.74227178,
      "num_input_tokens_seen": 63344925,
      "step": 2947,
      "time_per_iteration": 2.682549238204956
    },
    {
      "auxiliary_loss_clip": 0.01297321,
      "auxiliary_loss_mlp": 0.01026116,
      "balance_loss_clip": 1.05748618,
      "balance_loss_mlp": 1.01834393,
      "epoch": 0.35447604160404017,
      "flos": 23951807153280.0,
      "grad_norm": 3.4698281544662,
      "language_loss": 0.62770754,
      "learning_rate": 2.99369705366773e-06,
      "loss": 0.65094197,
      "num_input_tokens_seen": 63365170,
      "step": 2948,
      "time_per_iteration": 3.67044997215271
    },
    {
      "auxiliary_loss_clip": 0.01292924,
      "auxiliary_loss_mlp": 0.01029159,
      "balance_loss_clip": 1.05479431,
      "balance_loss_mlp": 1.0215174,
      "epoch": 0.3545962844946792,
      "flos": 23435662671360.0,
      "grad_norm": 2.272329350863742,
      "language_loss": 0.82378089,
      "learning_rate": 2.9930209576791244e-06,
      "loss": 0.84700167,
      "num_input_tokens_seen": 63383645,
      "step": 2949,
      "time_per_iteration": 2.7504618167877197
    },
    {
      "auxiliary_loss_clip": 0.01242156,
      "auxiliary_loss_mlp": 0.01028602,
      "balance_loss_clip": 1.05500674,
      "balance_loss_mlp": 1.02072251,
      "epoch": 0.35471652738531834,
      "flos": 22085134185600.0,
      "grad_norm": 2.0012296292381095,
      "language_loss": 0.63810778,
      "learning_rate": 2.9923447110500285e-06,
      "loss": 0.66081536,
      "num_input_tokens_seen": 63402390,
      "step": 2950,
      "time_per_iteration": 3.55464243888855
    },
    {
      "auxiliary_loss_clip": 0.01241412,
      "auxiliary_loss_mlp": 0.01027069,
      "balance_loss_clip": 1.05502033,
      "balance_loss_mlp": 1.01890945,
      "epoch": 0.35483677027595745,
      "flos": 27341881787520.0,
      "grad_norm": 1.4177018290837673,
      "language_loss": 0.75303149,
      "learning_rate": 2.9916683138830295e-06,
      "loss": 0.7757163,
      "num_input_tokens_seen": 63423055,
      "step": 2951,
      "time_per_iteration": 2.7042813301086426
    },
    {
      "auxiliary_loss_clip": 0.0129459,
      "auxiliary_loss_mlp": 0.01030945,
      "balance_loss_clip": 1.05647469,
      "balance_loss_mlp": 1.022892,
      "epoch": 0.3549570131665965,
      "flos": 13516166678400.0,
      "grad_norm": 2.037661827915762,
      "language_loss": 0.81105608,
      "learning_rate": 2.9909917662807353e-06,
      "loss": 0.83431137,
      "num_input_tokens_seen": 63440855,
      "step": 2952,
      "time_per_iteration": 2.695880174636841
    },
    {
      "auxiliary_loss_clip": 0.01244415,
      "auxiliary_loss_mlp": 0.01029196,
      "balance_loss_clip": 1.05634296,
      "balance_loss_mlp": 1.02092004,
      "epoch": 0.3550772560572356,
      "flos": 20887549810560.0,
      "grad_norm": 2.3558241636632413,
      "language_loss": 0.69033897,
      "learning_rate": 2.9903150683457783e-06,
      "loss": 0.7130751,
      "num_input_tokens_seen": 63459400,
      "step": 2953,
      "time_per_iteration": 2.694754123687744
    },
    {
      "auxiliary_loss_clip": 0.01204998,
      "auxiliary_loss_mlp": 0.01035912,
      "balance_loss_clip": 1.05476391,
      "balance_loss_mlp": 1.02804434,
      "epoch": 0.3551974989478747,
      "flos": 20194042947840.0,
      "grad_norm": 2.0910234547330413,
      "language_loss": 0.65692914,
      "learning_rate": 2.9896382201808126e-06,
      "loss": 0.67933822,
      "num_input_tokens_seen": 63476800,
      "step": 2954,
      "time_per_iteration": 2.7765626907348633
    },
    {
      "auxiliary_loss_clip": 0.01199604,
      "auxiliary_loss_mlp": 0.01034808,
      "balance_loss_clip": 1.05809021,
      "balance_loss_mlp": 1.02623725,
      "epoch": 0.3553177418385138,
      "flos": 19828831415040.0,
      "grad_norm": 2.4342032491161096,
      "language_loss": 0.81209171,
      "learning_rate": 2.988961221888516e-06,
      "loss": 0.83443582,
      "num_input_tokens_seen": 63493475,
      "step": 2955,
      "time_per_iteration": 3.6689159870147705
    },
    {
      "auxiliary_loss_clip": 0.01338126,
      "auxiliary_loss_mlp": 0.01029265,
      "balance_loss_clip": 1.05035496,
      "balance_loss_mlp": 1.02126598,
      "epoch": 0.3554379847291529,
      "flos": 14829132516480.0,
      "grad_norm": 2.4899091048557205,
      "language_loss": 0.79314733,
      "learning_rate": 2.988284073571589e-06,
      "loss": 0.81682128,
      "num_input_tokens_seen": 63509560,
      "step": 2956,
      "time_per_iteration": 2.7968881130218506
    },
    {
      "auxiliary_loss_clip": 0.01251838,
      "auxiliary_loss_mlp": 0.02567856,
      "balance_loss_clip": 1.0592767,
      "balance_loss_mlp": 1.0001651,
      "epoch": 0.355558227619792,
      "flos": 20485350247680.0,
      "grad_norm": 2.344623167266212,
      "language_loss": 0.73605549,
      "learning_rate": 2.9876067753327528e-06,
      "loss": 0.77425253,
      "num_input_tokens_seen": 63527290,
      "step": 2957,
      "time_per_iteration": 2.6749179363250732
    },
    {
      "auxiliary_loss_clip": 0.01158613,
      "auxiliary_loss_mlp": 0.01027882,
      "balance_loss_clip": 1.05586815,
      "balance_loss_mlp": 1.019853,
      "epoch": 0.35567847051043106,
      "flos": 37663613256960.0,
      "grad_norm": 2.3004940671535348,
      "language_loss": 0.80557799,
      "learning_rate": 2.986929327274754e-06,
      "loss": 0.827443,
      "num_input_tokens_seen": 63547870,
      "step": 2958,
      "time_per_iteration": 3.6901650428771973
    },
    {
      "auxiliary_loss_clip": 0.01245049,
      "auxiliary_loss_mlp": 0.01031015,
      "balance_loss_clip": 1.05833793,
      "balance_loss_mlp": 1.0230577,
      "epoch": 0.35579871340107017,
      "flos": 26943058103040.0,
      "grad_norm": 1.6961911792437312,
      "language_loss": 0.79221648,
      "learning_rate": 2.9862517295003617e-06,
      "loss": 0.81497711,
      "num_input_tokens_seen": 63568285,
      "step": 2959,
      "time_per_iteration": 2.7463302612304688
    },
    {
      "auxiliary_loss_clip": 0.01346372,
      "auxiliary_loss_mlp": 0.01028545,
      "balance_loss_clip": 1.0520159,
      "balance_loss_mlp": 1.02043259,
      "epoch": 0.3559189562917093,
      "flos": 28293335193600.0,
      "grad_norm": 1.5558816365458525,
      "language_loss": 0.72387207,
      "learning_rate": 2.9855739821123654e-06,
      "loss": 0.74762118,
      "num_input_tokens_seen": 63589865,
      "step": 2960,
      "time_per_iteration": 2.7461211681365967
    },
    {
      "auxiliary_loss_clip": 0.01242475,
      "auxiliary_loss_mlp": 0.01030188,
      "balance_loss_clip": 1.05701363,
      "balance_loss_mlp": 1.02238607,
      "epoch": 0.35603919918234833,
      "flos": 25664063552640.0,
      "grad_norm": 1.9020468192867084,
      "language_loss": 0.82389641,
      "learning_rate": 2.98489608521358e-06,
      "loss": 0.846623,
      "num_input_tokens_seen": 63609805,
      "step": 2961,
      "time_per_iteration": 2.759153366088867
    },
    {
      "auxiliary_loss_clip": 0.012548,
      "auxiliary_loss_mlp": 0.02566659,
      "balance_loss_clip": 1.05948472,
      "balance_loss_mlp": 1.00024128,
      "epoch": 0.35615944207298744,
      "flos": 23000856537600.0,
      "grad_norm": 2.1627255666364675,
      "language_loss": 0.79832017,
      "learning_rate": 2.9842180389068425e-06,
      "loss": 0.83653474,
      "num_input_tokens_seen": 63627115,
      "step": 2962,
      "time_per_iteration": 2.6509788036346436
    },
    {
      "auxiliary_loss_clip": 0.01270821,
      "auxiliary_loss_mlp": 0.01004156,
      "balance_loss_clip": 1.04057419,
      "balance_loss_mlp": 1.00299346,
      "epoch": 0.35627968496362655,
      "flos": 68251283723520.0,
      "grad_norm": 0.8089236796170745,
      "language_loss": 0.59168649,
      "learning_rate": 2.98353984329501e-06,
      "loss": 0.61443621,
      "num_input_tokens_seen": 63691460,
      "step": 2963,
      "time_per_iteration": 3.30861234664917
    },
    {
      "auxiliary_loss_clip": 0.01294795,
      "auxiliary_loss_mlp": 0.01026723,
      "balance_loss_clip": 1.05708587,
      "balance_loss_mlp": 1.01875365,
      "epoch": 0.3563999278542656,
      "flos": 22641714403200.0,
      "grad_norm": 1.5421768292027147,
      "language_loss": 0.70437837,
      "learning_rate": 2.982861498480965e-06,
      "loss": 0.72759348,
      "num_input_tokens_seen": 63713840,
      "step": 2964,
      "time_per_iteration": 2.7424862384796143
    },
    {
      "auxiliary_loss_clip": 0.01341114,
      "auxiliary_loss_mlp": 0.01029739,
      "balance_loss_clip": 1.04865766,
      "balance_loss_mlp": 1.02194858,
      "epoch": 0.3565201707449047,
      "flos": 25952533678080.0,
      "grad_norm": 1.747494597500948,
      "language_loss": 0.83035445,
      "learning_rate": 2.9821830045676122e-06,
      "loss": 0.85406297,
      "num_input_tokens_seen": 63733540,
      "step": 2965,
      "time_per_iteration": 2.790914297103882
    },
    {
      "auxiliary_loss_clip": 0.01200699,
      "auxiliary_loss_mlp": 0.01030727,
      "balance_loss_clip": 1.06033838,
      "balance_loss_mlp": 1.02281785,
      "epoch": 0.3566404136355438,
      "flos": 28475725478400.0,
      "grad_norm": 1.810577063942442,
      "language_loss": 0.73600501,
      "learning_rate": 2.9815043616578793e-06,
      "loss": 0.75831926,
      "num_input_tokens_seen": 63754335,
      "step": 2966,
      "time_per_iteration": 2.713864326477051
    },
    {
      "auxiliary_loss_clip": 0.01345197,
      "auxiliary_loss_mlp": 0.01031221,
      "balance_loss_clip": 1.05172026,
      "balance_loss_mlp": 1.02256083,
      "epoch": 0.3567606565261829,
      "flos": 38363117690880.0,
      "grad_norm": 2.7182058955177824,
      "language_loss": 0.77022707,
      "learning_rate": 2.9808255698547145e-06,
      "loss": 0.79399133,
      "num_input_tokens_seen": 63777135,
      "step": 2967,
      "time_per_iteration": 2.8962297439575195
    },
    {
      "auxiliary_loss_clip": 0.01245552,
      "auxiliary_loss_mlp": 0.01026951,
      "balance_loss_clip": 1.05819809,
      "balance_loss_mlp": 1.01866555,
      "epoch": 0.356880899416822,
      "flos": 21981029592960.0,
      "grad_norm": 2.726898990487692,
      "language_loss": 0.795735,
      "learning_rate": 2.9801466292610913e-06,
      "loss": 0.81845999,
      "num_input_tokens_seen": 63797020,
      "step": 2968,
      "time_per_iteration": 2.7073915004730225
    },
    {
      "auxiliary_loss_clip": 0.01247444,
      "auxiliary_loss_mlp": 0.0103213,
      "balance_loss_clip": 1.0573951,
      "balance_loss_mlp": 1.02459598,
      "epoch": 0.35700114230746105,
      "flos": 18989132198400.0,
      "grad_norm": 2.1641190504190333,
      "language_loss": 0.81038547,
      "learning_rate": 2.979467539980003e-06,
      "loss": 0.83318126,
      "num_input_tokens_seen": 63813810,
      "step": 2969,
      "time_per_iteration": 2.6415023803710938
    },
    {
      "auxiliary_loss_clip": 0.01251995,
      "auxiliary_loss_mlp": 0.01029977,
      "balance_loss_clip": 1.05907428,
      "balance_loss_mlp": 1.02185285,
      "epoch": 0.35712138519810016,
      "flos": 19756112330880.0,
      "grad_norm": 1.8308594687598716,
      "language_loss": 0.76941317,
      "learning_rate": 2.978788302114468e-06,
      "loss": 0.79223287,
      "num_input_tokens_seen": 63830925,
      "step": 2970,
      "time_per_iteration": 2.7208497524261475
    },
    {
      "auxiliary_loss_clip": 0.01247784,
      "auxiliary_loss_mlp": 0.01032296,
      "balance_loss_clip": 1.05902362,
      "balance_loss_mlp": 1.02466679,
      "epoch": 0.35724162808873927,
      "flos": 35183012008320.0,
      "grad_norm": 2.687672071060123,
      "language_loss": 0.83277583,
      "learning_rate": 2.9781089157675255e-06,
      "loss": 0.85557663,
      "num_input_tokens_seen": 63849385,
      "step": 2971,
      "time_per_iteration": 2.8177216053009033
    },
    {
      "auxiliary_loss_clip": 0.01250867,
      "auxiliary_loss_mlp": 0.01026756,
      "balance_loss_clip": 1.06147063,
      "balance_loss_mlp": 1.01881111,
      "epoch": 0.3573618709793783,
      "flos": 25556726736000.0,
      "grad_norm": 3.8547058395599714,
      "language_loss": 0.88453019,
      "learning_rate": 2.977429381042238e-06,
      "loss": 0.90730637,
      "num_input_tokens_seen": 63870060,
      "step": 2972,
      "time_per_iteration": 2.7452831268310547
    },
    {
      "auxiliary_loss_clip": 0.01296845,
      "auxiliary_loss_mlp": 0.01031253,
      "balance_loss_clip": 1.05538273,
      "balance_loss_mlp": 1.0241363,
      "epoch": 0.35748211387001744,
      "flos": 29132352051840.0,
      "grad_norm": 2.6855890153259603,
      "language_loss": 0.88908601,
      "learning_rate": 2.9767496980416913e-06,
      "loss": 0.91236699,
      "num_input_tokens_seen": 63889355,
      "step": 2973,
      "time_per_iteration": 2.7648606300354004
    },
    {
      "auxiliary_loss_clip": 0.01291007,
      "auxiliary_loss_mlp": 0.01025684,
      "balance_loss_clip": 1.05235195,
      "balance_loss_mlp": 1.01772058,
      "epoch": 0.35760235676065655,
      "flos": 13954169122560.0,
      "grad_norm": 2.5106483207038033,
      "language_loss": 0.81181151,
      "learning_rate": 2.9760698668689914e-06,
      "loss": 0.83497834,
      "num_input_tokens_seen": 63905580,
      "step": 2974,
      "time_per_iteration": 3.7123777866363525
    },
    {
      "auxiliary_loss_clip": 0.01247694,
      "auxiliary_loss_mlp": 0.01024825,
      "balance_loss_clip": 1.05664515,
      "balance_loss_mlp": 1.01757145,
      "epoch": 0.3577225996512956,
      "flos": 44018688977280.0,
      "grad_norm": 1.9355895552522022,
      "language_loss": 0.71492374,
      "learning_rate": 2.975389887627269e-06,
      "loss": 0.73764896,
      "num_input_tokens_seen": 63928180,
      "step": 2975,
      "time_per_iteration": 2.894158363342285
    },
    {
      "auxiliary_loss_clip": 0.01351207,
      "auxiliary_loss_mlp": 0.01026476,
      "balance_loss_clip": 1.05627918,
      "balance_loss_mlp": 1.01879311,
      "epoch": 0.3578428425419347,
      "flos": 17055199013760.0,
      "grad_norm": 2.0979508680797223,
      "language_loss": 0.89808655,
      "learning_rate": 2.9747097604196764e-06,
      "loss": 0.92186332,
      "num_input_tokens_seen": 63944825,
      "step": 2976,
      "time_per_iteration": 3.742046356201172
    },
    {
      "auxiliary_loss_clip": 0.01301215,
      "auxiliary_loss_mlp": 0.01002212,
      "balance_loss_clip": 1.019889,
      "balance_loss_mlp": 1.00096059,
      "epoch": 0.3579630854325738,
      "flos": 71676550707840.0,
      "grad_norm": 0.6754156572029947,
      "language_loss": 0.56680131,
      "learning_rate": 2.9740294853493875e-06,
      "loss": 0.58983552,
      "num_input_tokens_seen": 64016385,
      "step": 2977,
      "time_per_iteration": 3.5512185096740723
    },
    {
      "auxiliary_loss_clip": 0.01398151,
      "auxiliary_loss_mlp": 0.01029634,
      "balance_loss_clip": 1.04983664,
      "balance_loss_mlp": 1.02216506,
      "epoch": 0.3580833283232129,
      "flos": 25046651652480.0,
      "grad_norm": 2.4463176358511047,
      "language_loss": 0.67477667,
      "learning_rate": 2.9733490625196008e-06,
      "loss": 0.69905448,
      "num_input_tokens_seen": 64036245,
      "step": 2978,
      "time_per_iteration": 2.822939395904541
    },
    {
      "auxiliary_loss_clip": 0.01342111,
      "auxiliary_loss_mlp": 0.01029363,
      "balance_loss_clip": 1.05362391,
      "balance_loss_mlp": 1.02194798,
      "epoch": 0.358203571213852,
      "flos": 13953127628160.0,
      "grad_norm": 3.08481580988248,
      "language_loss": 0.75779885,
      "learning_rate": 2.9726684920335353e-06,
      "loss": 0.78151357,
      "num_input_tokens_seen": 64054110,
      "step": 2979,
      "time_per_iteration": 2.700693368911743
    },
    {
      "auxiliary_loss_clip": 0.0120122,
      "auxiliary_loss_mlp": 0.02569941,
      "balance_loss_clip": 1.05955529,
      "balance_loss_mlp": 1.00014174,
      "epoch": 0.35832381410449105,
      "flos": 20302457172480.0,
      "grad_norm": 2.3539410676824986,
      "language_loss": 0.8217833,
      "learning_rate": 2.971987773994432e-06,
      "loss": 0.85949498,
      "num_input_tokens_seen": 64070295,
      "step": 2980,
      "time_per_iteration": 2.676730155944824
    },
    {
      "auxiliary_loss_clip": 0.01240335,
      "auxiliary_loss_mlp": 0.01032782,
      "balance_loss_clip": 1.05266678,
      "balance_loss_mlp": 1.02496147,
      "epoch": 0.35844405699513016,
      "flos": 16983234115200.0,
      "grad_norm": 2.0557398391177424,
      "language_loss": 0.82761037,
      "learning_rate": 2.9713069085055566e-06,
      "loss": 0.85034156,
      "num_input_tokens_seen": 64088605,
      "step": 2981,
      "time_per_iteration": 3.6268632411956787
    },
    {
      "auxiliary_loss_clip": 0.01344327,
      "auxiliary_loss_mlp": 0.01028505,
      "balance_loss_clip": 1.05288041,
      "balance_loss_mlp": 1.0206728,
      "epoch": 0.35856429988576927,
      "flos": 23216858974080.0,
      "grad_norm": 2.103435737586411,
      "language_loss": 0.79076028,
      "learning_rate": 2.9706258956701958e-06,
      "loss": 0.81448865,
      "num_input_tokens_seen": 64108595,
      "step": 2982,
      "time_per_iteration": 2.74326229095459
    },
    {
      "auxiliary_loss_clip": 0.01250913,
      "auxiliary_loss_mlp": 0.01031449,
      "balance_loss_clip": 1.05923295,
      "balance_loss_mlp": 1.02319956,
      "epoch": 0.3586845427764083,
      "flos": 23034576430080.0,
      "grad_norm": 2.624781458655542,
      "language_loss": 0.77545166,
      "learning_rate": 2.9699447355916575e-06,
      "loss": 0.79827535,
      "num_input_tokens_seen": 64127405,
      "step": 2983,
      "time_per_iteration": 2.7463860511779785
    },
    {
      "auxiliary_loss_clip": 0.0119661,
      "auxiliary_loss_mlp": 0.02564055,
      "balance_loss_clip": 1.05766177,
      "balance_loss_mlp": 1.00020957,
      "epoch": 0.35880478566704743,
      "flos": 20010682995840.0,
      "grad_norm": 2.2063978988529644,
      "language_loss": 0.74214208,
      "learning_rate": 2.969263428373275e-06,
      "loss": 0.7797488,
      "num_input_tokens_seen": 64145755,
      "step": 2984,
      "time_per_iteration": 3.598775863647461
    },
    {
      "auxiliary_loss_clip": 0.01297587,
      "auxiliary_loss_mlp": 0.01034413,
      "balance_loss_clip": 1.05490136,
      "balance_loss_mlp": 1.02615809,
      "epoch": 0.35892502855768654,
      "flos": 13699095667200.0,
      "grad_norm": 2.0234310595002194,
      "language_loss": 0.79588473,
      "learning_rate": 2.9685819741184007e-06,
      "loss": 0.81920475,
      "num_input_tokens_seen": 64164195,
      "step": 2985,
      "time_per_iteration": 2.924746036529541
    },
    {
      "auxiliary_loss_clip": 0.01337907,
      "auxiliary_loss_mlp": 0.01027385,
      "balance_loss_clip": 1.05148768,
      "balance_loss_mlp": 1.02008927,
      "epoch": 0.3590452714483256,
      "flos": 18114096977280.0,
      "grad_norm": 4.476017933994888,
      "language_loss": 0.68820989,
      "learning_rate": 2.967900372930411e-06,
      "loss": 0.71186274,
      "num_input_tokens_seen": 64182705,
      "step": 2986,
      "time_per_iteration": 2.74324631690979
    },
    {
      "auxiliary_loss_clip": 0.01293657,
      "auxiliary_loss_mlp": 0.01028032,
      "balance_loss_clip": 1.05351353,
      "balance_loss_mlp": 1.02005661,
      "epoch": 0.3591655143389647,
      "flos": 17749352321280.0,
      "grad_norm": 2.4639385955964275,
      "language_loss": 0.79286218,
      "learning_rate": 2.9672186249127046e-06,
      "loss": 0.81607908,
      "num_input_tokens_seen": 64202170,
      "step": 2987,
      "time_per_iteration": 2.683039665222168
    },
    {
      "auxiliary_loss_clip": 0.01296202,
      "auxiliary_loss_mlp": 0.01035248,
      "balance_loss_clip": 1.05632865,
      "balance_loss_mlp": 1.02795208,
      "epoch": 0.3592857572296038,
      "flos": 25224409082880.0,
      "grad_norm": 2.0683908266243907,
      "language_loss": 0.79263461,
      "learning_rate": 2.9665367301687014e-06,
      "loss": 0.81594908,
      "num_input_tokens_seen": 64220415,
      "step": 2988,
      "time_per_iteration": 2.8014633655548096
    },
    {
      "auxiliary_loss_clip": 0.01291307,
      "auxiliary_loss_mlp": 0.01028489,
      "balance_loss_clip": 1.05179775,
      "balance_loss_mlp": 1.02064478,
      "epoch": 0.3594060001202429,
      "flos": 29384408764800.0,
      "grad_norm": 1.9197467979577256,
      "language_loss": 0.76690525,
      "learning_rate": 2.965854688801845e-06,
      "loss": 0.7901032,
      "num_input_tokens_seen": 64242475,
      "step": 2989,
      "time_per_iteration": 2.770176649093628
    },
    {
      "auxiliary_loss_clip": 0.01243168,
      "auxiliary_loss_mlp": 0.01029926,
      "balance_loss_clip": 1.05149853,
      "balance_loss_mlp": 1.02237141,
      "epoch": 0.359526243010882,
      "flos": 17052900543360.0,
      "grad_norm": 1.8335903440015937,
      "language_loss": 0.76345295,
      "learning_rate": 2.9651725009156005e-06,
      "loss": 0.78618395,
      "num_input_tokens_seen": 64260220,
      "step": 2990,
      "time_per_iteration": 2.700124740600586
    },
    {
      "auxiliary_loss_clip": 0.01290241,
      "auxiliary_loss_mlp": 0.0103285,
      "balance_loss_clip": 1.05185652,
      "balance_loss_mlp": 1.0247556,
      "epoch": 0.3596464859015211,
      "flos": 22965089569920.0,
      "grad_norm": 1.6935712865448291,
      "language_loss": 0.74228418,
      "learning_rate": 2.964490166613454e-06,
      "loss": 0.76551509,
      "num_input_tokens_seen": 64280145,
      "step": 2991,
      "time_per_iteration": 2.680142402648926
    },
    {
      "auxiliary_loss_clip": 0.01088439,
      "auxiliary_loss_mlp": 0.01012515,
      "balance_loss_clip": 1.01912415,
      "balance_loss_mlp": 1.01123381,
      "epoch": 0.35976672879216015,
      "flos": 54739462590720.0,
      "grad_norm": 0.7570451811322899,
      "language_loss": 0.57678711,
      "learning_rate": 2.963807685998917e-06,
      "loss": 0.59779668,
      "num_input_tokens_seen": 64336010,
      "step": 2992,
      "time_per_iteration": 3.007030725479126
    },
    {
      "auxiliary_loss_clip": 0.01393766,
      "auxiliary_loss_mlp": 0.01029288,
      "balance_loss_clip": 1.0493921,
      "balance_loss_mlp": 1.02196229,
      "epoch": 0.35988697168279926,
      "flos": 43139020901760.0,
      "grad_norm": 1.5001568071570641,
      "language_loss": 0.7840237,
      "learning_rate": 2.9631250591755196e-06,
      "loss": 0.80825424,
      "num_input_tokens_seen": 64358725,
      "step": 2993,
      "time_per_iteration": 2.88545560836792
    },
    {
      "auxiliary_loss_clip": 0.01289979,
      "auxiliary_loss_mlp": 0.01026275,
      "balance_loss_clip": 1.05542612,
      "balance_loss_mlp": 1.01838934,
      "epoch": 0.36000721457343837,
      "flos": 35845600239360.0,
      "grad_norm": 2.0138246480398387,
      "language_loss": 0.58059299,
      "learning_rate": 2.962442286246817e-06,
      "loss": 0.60375553,
      "num_input_tokens_seen": 64381555,
      "step": 2994,
      "time_per_iteration": 2.8754172325134277
    },
    {
      "auxiliary_loss_clip": 0.01205264,
      "auxiliary_loss_mlp": 0.01024407,
      "balance_loss_clip": 1.0549475,
      "balance_loss_mlp": 1.01631916,
      "epoch": 0.3601274574640774,
      "flos": 18291100222080.0,
      "grad_norm": 2.27123205485388,
      "language_loss": 0.69320744,
      "learning_rate": 2.9617593673163853e-06,
      "loss": 0.71550417,
      "num_input_tokens_seen": 64400375,
      "step": 2995,
      "time_per_iteration": 2.6835734844207764
    },
    {
      "auxiliary_loss_clip": 0.01296024,
      "auxiliary_loss_mlp": 0.01021611,
      "balance_loss_clip": 1.04997492,
      "balance_loss_mlp": 1.01450658,
      "epoch": 0.36024770035471654,
      "flos": 13333955961600.0,
      "grad_norm": 2.198435730454935,
      "language_loss": 0.77367771,
      "learning_rate": 2.9610763024878216e-06,
      "loss": 0.79685402,
      "num_input_tokens_seen": 64415880,
      "step": 2996,
      "time_per_iteration": 2.642463207244873
    },
    {
      "auxiliary_loss_clip": 0.01294343,
      "auxiliary_loss_mlp": 0.01026839,
      "balance_loss_clip": 1.05295825,
      "balance_loss_mlp": 1.01968658,
      "epoch": 0.3603679432453556,
      "flos": 20267013427200.0,
      "grad_norm": 2.1090013596956085,
      "language_loss": 0.9168551,
      "learning_rate": 2.960393091864747e-06,
      "loss": 0.94006687,
      "num_input_tokens_seen": 64434260,
      "step": 2997,
      "time_per_iteration": 2.7425074577331543
    },
    {
      "auxiliary_loss_clip": 0.01290701,
      "auxiliary_loss_mlp": 0.01028904,
      "balance_loss_clip": 1.05153012,
      "balance_loss_mlp": 1.02125943,
      "epoch": 0.3604881861359947,
      "flos": 22451135817600.0,
      "grad_norm": 1.793550116242748,
      "language_loss": 0.75340343,
      "learning_rate": 2.959709735550804e-06,
      "loss": 0.77659941,
      "num_input_tokens_seen": 64453855,
      "step": 2998,
      "time_per_iteration": 2.7236011028289795
    },
    {
      "auxiliary_loss_clip": 0.0139803,
      "auxiliary_loss_mlp": 0.01028926,
      "balance_loss_clip": 1.05282235,
      "balance_loss_mlp": 1.02115917,
      "epoch": 0.3606084290266338,
      "flos": 22054251467520.0,
      "grad_norm": 2.725967434014118,
      "language_loss": 0.7605828,
      "learning_rate": 2.9590262336496575e-06,
      "loss": 0.78485233,
      "num_input_tokens_seen": 64473585,
      "step": 2999,
      "time_per_iteration": 2.8498809337615967
    },
    {
      "auxiliary_loss_clip": 0.01352342,
      "auxiliary_loss_mlp": 0.0103678,
      "balance_loss_clip": 1.05842185,
      "balance_loss_mlp": 1.02756488,
      "epoch": 0.36072867191727287,
      "flos": 15632921111040.0,
      "grad_norm": 2.0539637534491426,
      "language_loss": 0.85623461,
      "learning_rate": 2.9583425862649936e-06,
      "loss": 0.88012576,
      "num_input_tokens_seen": 64491720,
      "step": 3000,
      "time_per_iteration": 3.739971399307251
    },
    {
      "auxiliary_loss_clip": 0.01201165,
      "auxiliary_loss_mlp": 0.01027691,
      "balance_loss_clip": 1.05922151,
      "balance_loss_mlp": 1.01888728,
      "epoch": 0.360848914807912,
      "flos": 19677000625920.0,
      "grad_norm": 2.840433286976049,
      "language_loss": 0.73826516,
      "learning_rate": 2.9576587935005215e-06,
      "loss": 0.76055372,
      "num_input_tokens_seen": 64509800,
      "step": 3001,
      "time_per_iteration": 2.7730135917663574
    },
    {
      "auxiliary_loss_clip": 0.01249806,
      "auxiliary_loss_mlp": 0.01033606,
      "balance_loss_clip": 1.05686629,
      "balance_loss_mlp": 1.025249,
      "epoch": 0.3609691576985511,
      "flos": 18877808972160.0,
      "grad_norm": 2.2565212171712563,
      "language_loss": 0.72006863,
      "learning_rate": 2.9569748554599713e-06,
      "loss": 0.74290276,
      "num_input_tokens_seen": 64525410,
      "step": 3002,
      "time_per_iteration": 3.5440452098846436
    },
    {
      "auxiliary_loss_clip": 0.01293903,
      "auxiliary_loss_mlp": 0.01023571,
      "balance_loss_clip": 1.0547936,
      "balance_loss_mlp": 1.01609349,
      "epoch": 0.36108940058919015,
      "flos": 42224088648960.0,
      "grad_norm": 2.089164691606238,
      "language_loss": 0.73638737,
      "learning_rate": 2.956290772247097e-06,
      "loss": 0.75956202,
      "num_input_tokens_seen": 64544085,
      "step": 3003,
      "time_per_iteration": 2.912515878677368
    },
    {
      "auxiliary_loss_clip": 0.01385755,
      "auxiliary_loss_mlp": 0.01028971,
      "balance_loss_clip": 1.05164647,
      "balance_loss_mlp": 1.02089429,
      "epoch": 0.36120964347982926,
      "flos": 23185150243200.0,
      "grad_norm": 1.9388513617752037,
      "language_loss": 0.73510969,
      "learning_rate": 2.9556065439656724e-06,
      "loss": 0.75925696,
      "num_input_tokens_seen": 64563135,
      "step": 3004,
      "time_per_iteration": 2.7971277236938477
    },
    {
      "auxiliary_loss_clip": 0.01436204,
      "auxiliary_loss_mlp": 0.01029684,
      "balance_loss_clip": 1.04540372,
      "balance_loss_mlp": 1.02212274,
      "epoch": 0.36132988637046837,
      "flos": 18113055482880.0,
      "grad_norm": 2.011154059835425,
      "language_loss": 0.8158623,
      "learning_rate": 2.9549221707194952e-06,
      "loss": 0.84052122,
      "num_input_tokens_seen": 64581985,
      "step": 3005,
      "time_per_iteration": 2.822597026824951
    },
    {
      "auxiliary_loss_clip": 0.01249265,
      "auxiliary_loss_mlp": 0.01030467,
      "balance_loss_clip": 1.05724609,
      "balance_loss_mlp": 1.02185392,
      "epoch": 0.3614501292611074,
      "flos": 27813101333760.0,
      "grad_norm": 1.8795961937630459,
      "language_loss": 0.72690231,
      "learning_rate": 2.954237652612384e-06,
      "loss": 0.74969959,
      "num_input_tokens_seen": 64601035,
      "step": 3006,
      "time_per_iteration": 2.7321994304656982
    },
    {
      "auxiliary_loss_clip": 0.01292665,
      "auxiliary_loss_mlp": 0.01031637,
      "balance_loss_clip": 1.05515373,
      "balance_loss_mlp": 1.0243113,
      "epoch": 0.36157037215174653,
      "flos": 22634926732800.0,
      "grad_norm": 2.177821564539628,
      "language_loss": 0.84912735,
      "learning_rate": 2.9535529897481796e-06,
      "loss": 0.87237036,
      "num_input_tokens_seen": 64618580,
      "step": 3007,
      "time_per_iteration": 3.5835440158843994
    },
    {
      "auxiliary_loss_clip": 0.01197952,
      "auxiliary_loss_mlp": 0.01031777,
      "balance_loss_clip": 1.05743086,
      "balance_loss_mlp": 1.02313423,
      "epoch": 0.36169061504238564,
      "flos": 12600839376000.0,
      "grad_norm": 2.4831276487998934,
      "language_loss": 0.76854181,
      "learning_rate": 2.9528681822307446e-06,
      "loss": 0.79083914,
      "num_input_tokens_seen": 64635430,
      "step": 3008,
      "time_per_iteration": 2.6247260570526123
    },
    {
      "auxiliary_loss_clip": 0.0124294,
      "auxiliary_loss_mlp": 0.02566019,
      "balance_loss_clip": 1.05599511,
      "balance_loss_mlp": 1.00010741,
      "epoch": 0.3618108579330247,
      "flos": 26684644682880.0,
      "grad_norm": 2.79137794792398,
      "language_loss": 0.82201338,
      "learning_rate": 2.952183230163964e-06,
      "loss": 0.86010301,
      "num_input_tokens_seen": 64655005,
      "step": 3009,
      "time_per_iteration": 3.676154375076294
    },
    {
      "auxiliary_loss_clip": 0.01332445,
      "auxiliary_loss_mlp": 0.01026581,
      "balance_loss_clip": 1.04766309,
      "balance_loss_mlp": 1.01871371,
      "epoch": 0.3619311008236638,
      "flos": 22817029708800.0,
      "grad_norm": 2.035801867893786,
      "language_loss": 0.72918665,
      "learning_rate": 2.9514981336517448e-06,
      "loss": 0.75277692,
      "num_input_tokens_seen": 64674775,
      "step": 3010,
      "time_per_iteration": 2.7914249897003174
    },
    {
      "auxiliary_loss_clip": 0.01244905,
      "auxiliary_loss_mlp": 0.01027784,
      "balance_loss_clip": 1.05776656,
      "balance_loss_mlp": 1.02007127,
      "epoch": 0.36205134371430286,
      "flos": 25919603884800.0,
      "grad_norm": 1.9423730687581724,
      "language_loss": 0.81206191,
      "learning_rate": 2.950812892798015e-06,
      "loss": 0.83478886,
      "num_input_tokens_seen": 64695670,
      "step": 3011,
      "time_per_iteration": 2.6941518783569336
    },
    {
      "auxiliary_loss_clip": 0.01385287,
      "auxiliary_loss_mlp": 0.02566233,
      "balance_loss_clip": 1.0524652,
      "balance_loss_mlp": 1.00021386,
      "epoch": 0.362171586604942,
      "flos": 26139592730880.0,
      "grad_norm": 2.4026555458428867,
      "language_loss": 0.87345302,
      "learning_rate": 2.9501275077067256e-06,
      "loss": 0.91296816,
      "num_input_tokens_seen": 64716290,
      "step": 3012,
      "time_per_iteration": 2.881466865539551
    },
    {
      "auxiliary_loss_clip": 0.01432478,
      "auxiliary_loss_mlp": 0.01028791,
      "balance_loss_clip": 1.04581702,
      "balance_loss_mlp": 1.02129281,
      "epoch": 0.3622918294955811,
      "flos": 28074208273920.0,
      "grad_norm": 1.5374911280372703,
      "language_loss": 0.88589048,
      "learning_rate": 2.949441978481848e-06,
      "loss": 0.91050321,
      "num_input_tokens_seen": 64737190,
      "step": 3013,
      "time_per_iteration": 2.896239757537842
    },
    {
      "auxiliary_loss_clip": 0.01256776,
      "auxiliary_loss_mlp": 0.01032063,
      "balance_loss_clip": 1.05307305,
      "balance_loss_mlp": 1.02373624,
      "epoch": 0.36241207238622014,
      "flos": 19828005402240.0,
      "grad_norm": 2.2143687702046755,
      "language_loss": 0.79846817,
      "learning_rate": 2.9487563052273778e-06,
      "loss": 0.82135653,
      "num_input_tokens_seen": 64753950,
      "step": 3014,
      "time_per_iteration": 2.766615867614746
    },
    {
      "auxiliary_loss_clip": 0.01253971,
      "auxiliary_loss_mlp": 0.01028816,
      "balance_loss_clip": 1.06404591,
      "balance_loss_mlp": 1.02070427,
      "epoch": 0.36253231527685925,
      "flos": 21397158017280.0,
      "grad_norm": 4.296978598433576,
      "language_loss": 0.85878909,
      "learning_rate": 2.94807048804733e-06,
      "loss": 0.88161695,
      "num_input_tokens_seen": 64773570,
      "step": 3015,
      "time_per_iteration": 2.696566104888916
    },
    {
      "auxiliary_loss_clip": 0.01258739,
      "auxiliary_loss_mlp": 0.01034341,
      "balance_loss_clip": 1.05248213,
      "balance_loss_mlp": 1.0267117,
      "epoch": 0.36265255816749836,
      "flos": 18362885552640.0,
      "grad_norm": 1.7954972925220913,
      "language_loss": 0.90143383,
      "learning_rate": 2.9473845270457434e-06,
      "loss": 0.92436469,
      "num_input_tokens_seen": 64790385,
      "step": 3016,
      "time_per_iteration": 2.7397186756134033
    },
    {
      "auxiliary_loss_clip": 0.01292738,
      "auxiliary_loss_mlp": 0.01034533,
      "balance_loss_clip": 1.05378234,
      "balance_loss_mlp": 1.02680266,
      "epoch": 0.3627728010581374,
      "flos": 18660046769280.0,
      "grad_norm": 2.212279746964265,
      "language_loss": 0.70544267,
      "learning_rate": 2.946698422326677e-06,
      "loss": 0.7287153,
      "num_input_tokens_seen": 64807845,
      "step": 3017,
      "time_per_iteration": 2.6429286003112793
    },
    {
      "auxiliary_loss_clip": 0.0139314,
      "auxiliary_loss_mlp": 0.01030659,
      "balance_loss_clip": 1.04951453,
      "balance_loss_mlp": 1.02242208,
      "epoch": 0.36289304394877653,
      "flos": 27524272072320.0,
      "grad_norm": 3.0565742289723374,
      "language_loss": 0.79979849,
      "learning_rate": 2.946012173994213e-06,
      "loss": 0.82403648,
      "num_input_tokens_seen": 64827630,
      "step": 3018,
      "time_per_iteration": 2.7804157733917236
    },
    {
      "auxiliary_loss_clip": 0.0124135,
      "auxiliary_loss_mlp": 0.01027163,
      "balance_loss_clip": 1.05876505,
      "balance_loss_mlp": 1.01959968,
      "epoch": 0.36301328683941564,
      "flos": 34533244932480.0,
      "grad_norm": 9.732205862403287,
      "language_loss": 0.67635572,
      "learning_rate": 2.945325782152454e-06,
      "loss": 0.69904089,
      "num_input_tokens_seen": 64850665,
      "step": 3019,
      "time_per_iteration": 2.8001904487609863
    },
    {
      "auxiliary_loss_clip": 0.01296219,
      "auxiliary_loss_mlp": 0.01025262,
      "balance_loss_clip": 1.05083871,
      "balance_loss_mlp": 1.01815748,
      "epoch": 0.3631335297300547,
      "flos": 19025976574080.0,
      "grad_norm": 2.079715427287812,
      "language_loss": 0.79204625,
      "learning_rate": 2.9446392469055257e-06,
      "loss": 0.81526107,
      "num_input_tokens_seen": 64868700,
      "step": 3020,
      "time_per_iteration": 2.6733763217926025
    },
    {
      "auxiliary_loss_clip": 0.01337574,
      "auxiliary_loss_mlp": 0.01032349,
      "balance_loss_clip": 1.05477262,
      "balance_loss_mlp": 1.02462411,
      "epoch": 0.3632537726206938,
      "flos": 19536769929600.0,
      "grad_norm": 6.627986391157917,
      "language_loss": 0.79758775,
      "learning_rate": 2.9439525683575745e-06,
      "loss": 0.82128692,
      "num_input_tokens_seen": 64887620,
      "step": 3021,
      "time_per_iteration": 2.735973596572876
    },
    {
      "auxiliary_loss_clip": 0.01203697,
      "auxiliary_loss_mlp": 0.01030334,
      "balance_loss_clip": 1.06089973,
      "balance_loss_mlp": 1.02194142,
      "epoch": 0.3633740155113329,
      "flos": 21068611292160.0,
      "grad_norm": 1.976809300905705,
      "language_loss": 0.75215805,
      "learning_rate": 2.9432657466127694e-06,
      "loss": 0.77449834,
      "num_input_tokens_seen": 64907190,
      "step": 3022,
      "time_per_iteration": 2.6092965602874756
    },
    {
      "auxiliary_loss_clip": 0.01393143,
      "auxiliary_loss_mlp": 0.01029024,
      "balance_loss_clip": 1.05638289,
      "balance_loss_mlp": 1.02169251,
      "epoch": 0.36349425840197197,
      "flos": 20298722158080.0,
      "grad_norm": 5.820645062830646,
      "language_loss": 0.76772952,
      "learning_rate": 2.9425787817753007e-06,
      "loss": 0.79195118,
      "num_input_tokens_seen": 64925850,
      "step": 3023,
      "time_per_iteration": 2.7539424896240234
    },
    {
      "auxiliary_loss_clip": 0.01254937,
      "auxiliary_loss_mlp": 0.01029326,
      "balance_loss_clip": 1.05444574,
      "balance_loss_mlp": 1.02140427,
      "epoch": 0.3636145012926111,
      "flos": 29716762331520.0,
      "grad_norm": 1.5958796630460161,
      "language_loss": 0.71572816,
      "learning_rate": 2.94189167394938e-06,
      "loss": 0.73857081,
      "num_input_tokens_seen": 64948285,
      "step": 3024,
      "time_per_iteration": 2.780282974243164
    },
    {
      "auxiliary_loss_clip": 0.01196945,
      "auxiliary_loss_mlp": 0.01031373,
      "balance_loss_clip": 1.05810452,
      "balance_loss_mlp": 1.02320075,
      "epoch": 0.3637347441832502,
      "flos": 21431847576960.0,
      "grad_norm": 2.0205102231882015,
      "language_loss": 0.81170332,
      "learning_rate": 2.941204423239241e-06,
      "loss": 0.83398652,
      "num_input_tokens_seen": 64967160,
      "step": 3025,
      "time_per_iteration": 3.6828622817993164
    },
    {
      "auxiliary_loss_clip": 0.01242781,
      "auxiliary_loss_mlp": 0.01027421,
      "balance_loss_clip": 1.05635715,
      "balance_loss_mlp": 1.01929665,
      "epoch": 0.36385498707388925,
      "flos": 29533941083520.0,
      "grad_norm": 1.9342106898355373,
      "language_loss": 0.75992274,
      "learning_rate": 2.9405170297491395e-06,
      "loss": 0.78262478,
      "num_input_tokens_seen": 64987155,
      "step": 3026,
      "time_per_iteration": 2.75007963180542
    },
    {
      "auxiliary_loss_clip": 0.01438617,
      "auxiliary_loss_mlp": 0.02570457,
      "balance_loss_clip": 1.05392468,
      "balance_loss_mlp": 1.00012934,
      "epoch": 0.36397522996452836,
      "flos": 22236569925120.0,
      "grad_norm": 2.3615952420775694,
      "language_loss": 0.80584311,
      "learning_rate": 2.939829493583353e-06,
      "loss": 0.84593385,
      "num_input_tokens_seen": 65003800,
      "step": 3027,
      "time_per_iteration": 2.7817671298980713
    },
    {
      "auxiliary_loss_clip": 0.01340147,
      "auxiliary_loss_mlp": 0.01029516,
      "balance_loss_clip": 1.04751039,
      "balance_loss_mlp": 1.02114129,
      "epoch": 0.3640954728551674,
      "flos": 21506505995520.0,
      "grad_norm": 2.8841537203748047,
      "language_loss": 0.83205372,
      "learning_rate": 2.939141814846179e-06,
      "loss": 0.85575032,
      "num_input_tokens_seen": 65021215,
      "step": 3028,
      "time_per_iteration": 3.6783313751220703
    },
    {
      "auxiliary_loss_clip": 0.01295916,
      "auxiliary_loss_mlp": 0.01030747,
      "balance_loss_clip": 1.05296624,
      "balance_loss_mlp": 1.02289724,
      "epoch": 0.3642157157458065,
      "flos": 17712867081600.0,
      "grad_norm": 1.9009310647209374,
      "language_loss": 0.8265059,
      "learning_rate": 2.938453993641938e-06,
      "loss": 0.84977257,
      "num_input_tokens_seen": 65039590,
      "step": 3029,
      "time_per_iteration": 2.6706743240356445
    },
    {
      "auxiliary_loss_clip": 0.01294817,
      "auxiliary_loss_mlp": 0.01027196,
      "balance_loss_clip": 1.05951428,
      "balance_loss_mlp": 1.0188756,
      "epoch": 0.36433595863644563,
      "flos": 17639537466240.0,
      "grad_norm": 2.737036002465674,
      "language_loss": 0.70034713,
      "learning_rate": 2.937766030074973e-06,
      "loss": 0.72356725,
      "num_input_tokens_seen": 65056845,
      "step": 3030,
      "time_per_iteration": 2.7120187282562256
    },
    {
      "auxiliary_loss_clip": 0.0125531,
      "auxiliary_loss_mlp": 0.01034789,
      "balance_loss_clip": 1.0537914,
      "balance_loss_mlp": 1.02600312,
      "epoch": 0.3644562015270847,
      "flos": 26833279161600.0,
      "grad_norm": 1.8636186924783458,
      "language_loss": 0.82761604,
      "learning_rate": 2.937077924249646e-06,
      "loss": 0.85051703,
      "num_input_tokens_seen": 65079435,
      "step": 3031,
      "time_per_iteration": 2.715573787689209
    },
    {
      "auxiliary_loss_clip": 0.01209179,
      "auxiliary_loss_mlp": 0.01029439,
      "balance_loss_clip": 1.05471826,
      "balance_loss_mlp": 1.02073073,
      "epoch": 0.3645764444177238,
      "flos": 14282715847680.0,
      "grad_norm": 2.531898145195225,
      "language_loss": 0.76066971,
      "learning_rate": 2.9363896762703443e-06,
      "loss": 0.7830559,
      "num_input_tokens_seen": 65096500,
      "step": 3032,
      "time_per_iteration": 2.6427853107452393
    },
    {
      "auxiliary_loss_clip": 0.01198173,
      "auxiliary_loss_mlp": 0.0102739,
      "balance_loss_clip": 1.05793881,
      "balance_loss_mlp": 1.01898587,
      "epoch": 0.3646966873083629,
      "flos": 20667489137280.0,
      "grad_norm": 1.637874701430685,
      "language_loss": 0.84770143,
      "learning_rate": 2.9357012862414725e-06,
      "loss": 0.86995709,
      "num_input_tokens_seen": 65115860,
      "step": 3033,
      "time_per_iteration": 3.5681889057159424
    },
    {
      "auxiliary_loss_clip": 0.0124855,
      "auxiliary_loss_mlp": 0.01026165,
      "balance_loss_clip": 1.05659258,
      "balance_loss_mlp": 1.01800561,
      "epoch": 0.36481693019900197,
      "flos": 27782613665280.0,
      "grad_norm": 2.207340432909591,
      "language_loss": 0.71892262,
      "learning_rate": 2.9350127542674593e-06,
      "loss": 0.74166977,
      "num_input_tokens_seen": 65138070,
      "step": 3034,
      "time_per_iteration": 2.6805710792541504
    },
    {
      "auxiliary_loss_clip": 0.01209551,
      "auxiliary_loss_mlp": 0.01027644,
      "balance_loss_clip": 1.05845678,
      "balance_loss_mlp": 1.01932955,
      "epoch": 0.3649371730896411,
      "flos": 19712588025600.0,
      "grad_norm": 2.0283670211791045,
      "language_loss": 0.76670092,
      "learning_rate": 2.934324080452755e-06,
      "loss": 0.78907287,
      "num_input_tokens_seen": 65155860,
      "step": 3035,
      "time_per_iteration": 2.752976894378662
    },
    {
      "auxiliary_loss_clip": 0.01341348,
      "auxiliary_loss_mlp": 0.02573531,
      "balance_loss_clip": 1.0506984,
      "balance_loss_mlp": 1.00013387,
      "epoch": 0.3650574159802802,
      "flos": 24750496016640.0,
      "grad_norm": 1.6730990115275195,
      "language_loss": 0.77959442,
      "learning_rate": 2.9336352649018307e-06,
      "loss": 0.81874323,
      "num_input_tokens_seen": 65175930,
      "step": 3036,
      "time_per_iteration": 3.9581828117370605
    },
    {
      "auxiliary_loss_clip": 0.01298966,
      "auxiliary_loss_mlp": 0.01035019,
      "balance_loss_clip": 1.05544519,
      "balance_loss_mlp": 1.02720535,
      "epoch": 0.36517765887091924,
      "flos": 32853487363200.0,
      "grad_norm": 1.758778565828236,
      "language_loss": 0.7019155,
      "learning_rate": 2.9329463077191783e-06,
      "loss": 0.72525537,
      "num_input_tokens_seen": 65199305,
      "step": 3037,
      "time_per_iteration": 2.76631498336792
    },
    {
      "auxiliary_loss_clip": 0.01401117,
      "auxiliary_loss_mlp": 0.01025215,
      "balance_loss_clip": 1.05359769,
      "balance_loss_mlp": 1.0173353,
      "epoch": 0.36529790176155835,
      "flos": 20120318282880.0,
      "grad_norm": 39.03295672251877,
      "language_loss": 0.64477181,
      "learning_rate": 2.9322572090093135e-06,
      "loss": 0.66903508,
      "num_input_tokens_seen": 65218010,
      "step": 3038,
      "time_per_iteration": 2.816297769546509
    },
    {
      "auxiliary_loss_clip": 0.01393553,
      "auxiliary_loss_mlp": 0.01030231,
      "balance_loss_clip": 1.0492053,
      "balance_loss_mlp": 1.02192807,
      "epoch": 0.36541814465219746,
      "flos": 17639573379840.0,
      "grad_norm": 2.4898874646944074,
      "language_loss": 0.76299214,
      "learning_rate": 2.9315679688767713e-06,
      "loss": 0.78722996,
      "num_input_tokens_seen": 65236020,
      "step": 3039,
      "time_per_iteration": 2.7503767013549805
    },
    {
      "auxiliary_loss_clip": 0.01292956,
      "auxiliary_loss_mlp": 0.01031152,
      "balance_loss_clip": 1.04951358,
      "balance_loss_mlp": 1.02314138,
      "epoch": 0.3655383875428365,
      "flos": 22674356887680.0,
      "grad_norm": 1.562543436886404,
      "language_loss": 0.66647226,
      "learning_rate": 2.9308785874261085e-06,
      "loss": 0.68971336,
      "num_input_tokens_seen": 65256210,
      "step": 3040,
      "time_per_iteration": 2.893484592437744
    },
    {
      "auxiliary_loss_clip": 0.0119974,
      "auxiliary_loss_mlp": 0.01030797,
      "balance_loss_clip": 1.0600481,
      "balance_loss_mlp": 1.0228219,
      "epoch": 0.36565863043347563,
      "flos": 21981173247360.0,
      "grad_norm": 1.6414104794618707,
      "language_loss": 0.82039905,
      "learning_rate": 2.9301890647619045e-06,
      "loss": 0.84270442,
      "num_input_tokens_seen": 65275505,
      "step": 3041,
      "time_per_iteration": 2.62060284614563
    },
    {
      "auxiliary_loss_clip": 0.01206564,
      "auxiliary_loss_mlp": 0.01031766,
      "balance_loss_clip": 1.05517852,
      "balance_loss_mlp": 1.02312303,
      "epoch": 0.36577887332411474,
      "flos": 24827632473600.0,
      "grad_norm": 2.7764720708186448,
      "language_loss": 0.80386138,
      "learning_rate": 2.929499400988759e-06,
      "loss": 0.82624471,
      "num_input_tokens_seen": 65296665,
      "step": 3042,
      "time_per_iteration": 2.7128212451934814
    },
    {
      "auxiliary_loss_clip": 0.01245379,
      "auxiliary_loss_mlp": 0.01030202,
      "balance_loss_clip": 1.05594516,
      "balance_loss_mlp": 1.02214932,
      "epoch": 0.3658991162147538,
      "flos": 28293191539200.0,
      "grad_norm": 2.523138678289529,
      "language_loss": 0.65139186,
      "learning_rate": 2.9288095962112927e-06,
      "loss": 0.67414773,
      "num_input_tokens_seen": 65317370,
      "step": 3043,
      "time_per_iteration": 2.7137391567230225
    },
    {
      "auxiliary_loss_clip": 0.01198174,
      "auxiliary_loss_mlp": 0.01026955,
      "balance_loss_clip": 1.05851018,
      "balance_loss_mlp": 1.01901019,
      "epoch": 0.3660193591053929,
      "flos": 17785550252160.0,
      "grad_norm": 2.034722929416078,
      "language_loss": 0.85390007,
      "learning_rate": 2.9281196505341503e-06,
      "loss": 0.87615144,
      "num_input_tokens_seen": 65334540,
      "step": 3044,
      "time_per_iteration": 2.7045884132385254
    },
    {
      "auxiliary_loss_clip": 0.01387306,
      "auxiliary_loss_mlp": 0.02567233,
      "balance_loss_clip": 1.05031824,
      "balance_loss_mlp": 1.00003302,
      "epoch": 0.36613960199603196,
      "flos": 10342776839040.0,
      "grad_norm": 2.271805425234311,
      "language_loss": 0.78598428,
      "learning_rate": 2.9274295640619946e-06,
      "loss": 0.82552963,
      "num_input_tokens_seen": 65351670,
      "step": 3045,
      "time_per_iteration": 2.7838637828826904
    },
    {
      "auxiliary_loss_clip": 0.01341936,
      "auxiliary_loss_mlp": 0.01036378,
      "balance_loss_clip": 1.04975843,
      "balance_loss_mlp": 1.0291661,
      "epoch": 0.36625984488667107,
      "flos": 19755609540480.0,
      "grad_norm": 1.7336001731898243,
      "language_loss": 0.7860083,
      "learning_rate": 2.9267393368995103e-06,
      "loss": 0.80979145,
      "num_input_tokens_seen": 65370900,
      "step": 3046,
      "time_per_iteration": 2.765315294265747
    },
    {
      "auxiliary_loss_clip": 0.01197971,
      "auxiliary_loss_mlp": 0.01031565,
      "balance_loss_clip": 1.05817926,
      "balance_loss_mlp": 1.0235064,
      "epoch": 0.3663800877773102,
      "flos": 17674262939520.0,
      "grad_norm": 3.0780618004346003,
      "language_loss": 0.74052173,
      "learning_rate": 2.926048969151407e-06,
      "loss": 0.76281714,
      "num_input_tokens_seen": 65388185,
      "step": 3047,
      "time_per_iteration": 2.631822109222412
    },
    {
      "auxiliary_loss_clip": 0.01379891,
      "auxiliary_loss_mlp": 0.01025544,
      "balance_loss_clip": 1.05054808,
      "balance_loss_mlp": 1.01734781,
      "epoch": 0.36650033066794924,
      "flos": 20303606407680.0,
      "grad_norm": 1.9790132881766362,
      "language_loss": 0.69057417,
      "learning_rate": 2.92535846092241e-06,
      "loss": 0.71462858,
      "num_input_tokens_seen": 65407200,
      "step": 3048,
      "time_per_iteration": 2.7616312503814697
    },
    {
      "auxiliary_loss_clip": 0.01298824,
      "auxiliary_loss_mlp": 0.01030591,
      "balance_loss_clip": 1.05707884,
      "balance_loss_mlp": 1.02254772,
      "epoch": 0.36662057355858835,
      "flos": 24716237420160.0,
      "grad_norm": 2.2027057997687702,
      "language_loss": 0.82592416,
      "learning_rate": 2.9246678123172704e-06,
      "loss": 0.84921831,
      "num_input_tokens_seen": 65427290,
      "step": 3049,
      "time_per_iteration": 2.739614725112915
    },
    {
      "auxiliary_loss_clip": 0.01200155,
      "auxiliary_loss_mlp": 0.01033247,
      "balance_loss_clip": 1.05880237,
      "balance_loss_mlp": 1.02510238,
      "epoch": 0.36674081644922746,
      "flos": 12385267902720.0,
      "grad_norm": 3.278875414342902,
      "language_loss": 0.74769223,
      "learning_rate": 2.9239770234407596e-06,
      "loss": 0.77002627,
      "num_input_tokens_seen": 65445595,
      "step": 3050,
      "time_per_iteration": 3.5758256912231445
    },
    {
      "auxiliary_loss_clip": 0.01247672,
      "auxiliary_loss_mlp": 0.01031023,
      "balance_loss_clip": 1.05585444,
      "balance_loss_mlp": 1.02288735,
      "epoch": 0.3668610593398665,
      "flos": 21105922544640.0,
      "grad_norm": 1.7091569984452317,
      "language_loss": 0.68896931,
      "learning_rate": 2.9232860943976686e-06,
      "loss": 0.71175623,
      "num_input_tokens_seen": 65466330,
      "step": 3051,
      "time_per_iteration": 2.7167863845825195
    },
    {
      "auxiliary_loss_clip": 0.01292641,
      "auxiliary_loss_mlp": 0.01032164,
      "balance_loss_clip": 1.05263901,
      "balance_loss_mlp": 1.02454066,
      "epoch": 0.3669813022305056,
      "flos": 26758082039040.0,
      "grad_norm": 1.66390243976988,
      "language_loss": 0.84421492,
      "learning_rate": 2.9225950252928115e-06,
      "loss": 0.86746287,
      "num_input_tokens_seen": 65487180,
      "step": 3052,
      "time_per_iteration": 2.8123667240142822
    },
    {
      "auxiliary_loss_clip": 0.01247125,
      "auxiliary_loss_mlp": 0.01029775,
      "balance_loss_clip": 1.05763769,
      "balance_loss_mlp": 1.02168727,
      "epoch": 0.36710154512114473,
      "flos": 19099521671040.0,
      "grad_norm": 2.108705891564849,
      "language_loss": 0.82142246,
      "learning_rate": 2.9219038162310217e-06,
      "loss": 0.84419143,
      "num_input_tokens_seen": 65505380,
      "step": 3053,
      "time_per_iteration": 2.690699815750122
    },
    {
      "auxiliary_loss_clip": 0.01405562,
      "auxiliary_loss_mlp": 0.02575282,
      "balance_loss_clip": 1.0501492,
      "balance_loss_mlp": 1.00006533,
      "epoch": 0.3672217880117838,
      "flos": 20812029465600.0,
      "grad_norm": 2.2419317124070752,
      "language_loss": 0.82721257,
      "learning_rate": 2.921212467317157e-06,
      "loss": 0.86702108,
      "num_input_tokens_seen": 65524825,
      "step": 3054,
      "time_per_iteration": 3.759967088699341
    },
    {
      "auxiliary_loss_clip": 0.01288984,
      "auxiliary_loss_mlp": 0.01028051,
      "balance_loss_clip": 1.05005956,
      "balance_loss_mlp": 1.01940823,
      "epoch": 0.3673420309024229,
      "flos": 13590394133760.0,
      "grad_norm": 2.0158375581342014,
      "language_loss": 0.80451119,
      "learning_rate": 2.920520978656093e-06,
      "loss": 0.82768148,
      "num_input_tokens_seen": 65541790,
      "step": 3055,
      "time_per_iteration": 3.0118439197540283
    },
    {
      "auxiliary_loss_clip": 0.01193541,
      "auxiliary_loss_mlp": 0.02569826,
      "balance_loss_clip": 1.05577803,
      "balance_loss_mlp": 1.00007081,
      "epoch": 0.367462273793062,
      "flos": 28986877969920.0,
      "grad_norm": 2.1605640506405734,
      "language_loss": 0.77369666,
      "learning_rate": 2.919829350352729e-06,
      "loss": 0.81133038,
      "num_input_tokens_seen": 65563395,
      "step": 3056,
      "time_per_iteration": 2.6673061847686768
    },
    {
      "auxiliary_loss_clip": 0.0108272,
      "auxiliary_loss_mlp": 0.01015843,
      "balance_loss_clip": 1.01388383,
      "balance_loss_mlp": 1.01455522,
      "epoch": 0.36758251668370107,
      "flos": 62643148346880.0,
      "grad_norm": 0.7536303211647591,
      "language_loss": 0.59992224,
      "learning_rate": 2.919137582511983e-06,
      "loss": 0.62090784,
      "num_input_tokens_seen": 65619835,
      "step": 3057,
      "time_per_iteration": 3.1133365631103516
    },
    {
      "auxiliary_loss_clip": 0.01359282,
      "auxiliary_loss_mlp": 0.01032925,
      "balance_loss_clip": 1.05938172,
      "balance_loss_mlp": 1.02491438,
      "epoch": 0.3677027595743402,
      "flos": 12713886455040.0,
      "grad_norm": 3.8242454158830395,
      "language_loss": 0.63853329,
      "learning_rate": 2.918445675238797e-06,
      "loss": 0.66245538,
      "num_input_tokens_seen": 65636760,
      "step": 3058,
      "time_per_iteration": 2.705484628677368
    },
    {
      "auxiliary_loss_clip": 0.0119626,
      "auxiliary_loss_mlp": 0.01028132,
      "balance_loss_clip": 1.0554142,
      "balance_loss_mlp": 1.01978087,
      "epoch": 0.36782300246497923,
      "flos": 25046579825280.0,
      "grad_norm": 2.7857906051980263,
      "language_loss": 0.69795287,
      "learning_rate": 2.917753628638132e-06,
      "loss": 0.72019678,
      "num_input_tokens_seen": 65657065,
      "step": 3059,
      "time_per_iteration": 3.5469605922698975
    },
    {
      "auxiliary_loss_clip": 0.01298389,
      "auxiliary_loss_mlp": 0.01026053,
      "balance_loss_clip": 1.05617785,
      "balance_loss_mlp": 1.01755309,
      "epoch": 0.36794324535561834,
      "flos": 17419512706560.0,
      "grad_norm": 9.800223366761692,
      "language_loss": 0.70373672,
      "learning_rate": 2.9170614428149716e-06,
      "loss": 0.72698116,
      "num_input_tokens_seen": 65675400,
      "step": 3060,
      "time_per_iteration": 2.6845271587371826
    },
    {
      "auxiliary_loss_clip": 0.01340386,
      "auxiliary_loss_mlp": 0.01030532,
      "balance_loss_clip": 1.05316305,
      "balance_loss_mlp": 1.02183032,
      "epoch": 0.36806348824625745,
      "flos": 24089128848000.0,
      "grad_norm": 2.91328981138716,
      "language_loss": 0.86779332,
      "learning_rate": 2.9163691178743195e-06,
      "loss": 0.8915025,
      "num_input_tokens_seen": 65694050,
      "step": 3061,
      "time_per_iteration": 3.804323434829712
    },
    {
      "auxiliary_loss_clip": 0.01241967,
      "auxiliary_loss_mlp": 0.01028366,
      "balance_loss_clip": 1.05533087,
      "balance_loss_mlp": 1.01997387,
      "epoch": 0.3681837311368965,
      "flos": 20521871400960.0,
      "grad_norm": 1.7151423420158292,
      "language_loss": 0.77362055,
      "learning_rate": 2.9156766539212006e-06,
      "loss": 0.79632384,
      "num_input_tokens_seen": 65711695,
      "step": 3062,
      "time_per_iteration": 2.6938748359680176
    },
    {
      "auxiliary_loss_clip": 0.01249372,
      "auxiliary_loss_mlp": 0.01029886,
      "balance_loss_clip": 1.05544806,
      "balance_loss_mlp": 1.02217901,
      "epoch": 0.3683039740275356,
      "flos": 21466644877440.0,
      "grad_norm": 2.921041672318301,
      "language_loss": 0.72203708,
      "learning_rate": 2.9149840510606614e-06,
      "loss": 0.74482965,
      "num_input_tokens_seen": 65730350,
      "step": 3063,
      "time_per_iteration": 2.6715407371520996
    },
    {
      "auxiliary_loss_clip": 0.01134381,
      "auxiliary_loss_mlp": 0.02512883,
      "balance_loss_clip": 1.01579177,
      "balance_loss_mlp": 1.00017798,
      "epoch": 0.36842421691817473,
      "flos": 70380999987840.0,
      "grad_norm": 1.0222100209291298,
      "language_loss": 0.64245057,
      "learning_rate": 2.914291309397769e-06,
      "loss": 0.67892325,
      "num_input_tokens_seen": 65787820,
      "step": 3064,
      "time_per_iteration": 3.366791009902954
    },
    {
      "auxiliary_loss_clip": 0.01430584,
      "auxiliary_loss_mlp": 0.01031779,
      "balance_loss_clip": 1.04579973,
      "balance_loss_mlp": 1.02386951,
      "epoch": 0.3685444598088138,
      "flos": 23331378510720.0,
      "grad_norm": 2.5000872486711376,
      "language_loss": 0.78608561,
      "learning_rate": 2.9135984290376117e-06,
      "loss": 0.81070924,
      "num_input_tokens_seen": 65806685,
      "step": 3065,
      "time_per_iteration": 2.85245418548584
    },
    {
      "auxiliary_loss_clip": 0.01439957,
      "auxiliary_loss_mlp": 0.01036502,
      "balance_loss_clip": 1.04713106,
      "balance_loss_mlp": 1.02874804,
      "epoch": 0.3686647026994529,
      "flos": 23070271570560.0,
      "grad_norm": 1.7968244864558323,
      "language_loss": 0.82601631,
      "learning_rate": 2.9129054100853e-06,
      "loss": 0.8507809,
      "num_input_tokens_seen": 65825525,
      "step": 3066,
      "time_per_iteration": 2.839069366455078
    },
    {
      "auxiliary_loss_clip": 0.01297948,
      "auxiliary_loss_mlp": 0.01035383,
      "balance_loss_clip": 1.05537331,
      "balance_loss_mlp": 1.02746153,
      "epoch": 0.368784945590092,
      "flos": 25119909440640.0,
      "grad_norm": 1.7394668448490995,
      "language_loss": 0.76281953,
      "learning_rate": 2.912212252645963e-06,
      "loss": 0.78615284,
      "num_input_tokens_seen": 65848110,
      "step": 3067,
      "time_per_iteration": 2.7729578018188477
    },
    {
      "auxiliary_loss_clip": 0.01251311,
      "auxiliary_loss_mlp": 0.01037839,
      "balance_loss_clip": 1.05575967,
      "balance_loss_mlp": 1.02904749,
      "epoch": 0.36890518848073106,
      "flos": 18442284566400.0,
      "grad_norm": 2.5180062048131346,
      "language_loss": 0.76257062,
      "learning_rate": 2.9115189568247523e-06,
      "loss": 0.78546208,
      "num_input_tokens_seen": 65865670,
      "step": 3068,
      "time_per_iteration": 2.6509242057800293
    },
    {
      "auxiliary_loss_clip": 0.01387978,
      "auxiliary_loss_mlp": 0.0102777,
      "balance_loss_clip": 1.05521011,
      "balance_loss_mlp": 1.02024519,
      "epoch": 0.36902543137137017,
      "flos": 16362446336640.0,
      "grad_norm": 2.181451859723835,
      "language_loss": 0.92351818,
      "learning_rate": 2.910825522726841e-06,
      "loss": 0.94767565,
      "num_input_tokens_seen": 65883195,
      "step": 3069,
      "time_per_iteration": 2.740921974182129
    },
    {
      "auxiliary_loss_clip": 0.01386298,
      "auxiliary_loss_mlp": 0.01027848,
      "balance_loss_clip": 1.04816818,
      "balance_loss_mlp": 1.02051365,
      "epoch": 0.3691456742620093,
      "flos": 12275596702080.0,
      "grad_norm": 2.0935600130105674,
      "language_loss": 0.77569628,
      "learning_rate": 2.9101319504574215e-06,
      "loss": 0.79983771,
      "num_input_tokens_seen": 65899635,
      "step": 3070,
      "time_per_iteration": 2.700392723083496
    },
    {
      "auxiliary_loss_clip": 0.01240022,
      "auxiliary_loss_mlp": 0.0102865,
      "balance_loss_clip": 1.05137086,
      "balance_loss_mlp": 1.02007902,
      "epoch": 0.36926591715264834,
      "flos": 17786412178560.0,
      "grad_norm": 1.6367629664366687,
      "language_loss": 0.76207066,
      "learning_rate": 2.909438240121709e-06,
      "loss": 0.78475744,
      "num_input_tokens_seen": 65919910,
      "step": 3071,
      "time_per_iteration": 2.713472604751587
    },
    {
      "auxiliary_loss_clip": 0.01284994,
      "auxiliary_loss_mlp": 0.01028611,
      "balance_loss_clip": 1.05349553,
      "balance_loss_mlp": 1.0211246,
      "epoch": 0.36938616004328745,
      "flos": 28948309741440.0,
      "grad_norm": 1.8988052153801198,
      "language_loss": 0.70493412,
      "learning_rate": 2.908744391824939e-06,
      "loss": 0.72807014,
      "num_input_tokens_seen": 65940930,
      "step": 3072,
      "time_per_iteration": 2.7147817611694336
    },
    {
      "auxiliary_loss_clip": 0.01450716,
      "auxiliary_loss_mlp": 0.01027652,
      "balance_loss_clip": 1.0487411,
      "balance_loss_mlp": 1.01944435,
      "epoch": 0.36950640293392656,
      "flos": 29205394358400.0,
      "grad_norm": 2.0488632360745003,
      "language_loss": 0.79068029,
      "learning_rate": 2.908050405672367e-06,
      "loss": 0.8154639,
      "num_input_tokens_seen": 65960475,
      "step": 3073,
      "time_per_iteration": 2.866995334625244
    },
    {
      "auxiliary_loss_clip": 0.01303242,
      "auxiliary_loss_mlp": 0.01027044,
      "balance_loss_clip": 1.05066967,
      "balance_loss_mlp": 1.01908076,
      "epoch": 0.3696266458245656,
      "flos": 24827776128000.0,
      "grad_norm": 2.219067358074968,
      "language_loss": 0.79371095,
      "learning_rate": 2.9073562817692703e-06,
      "loss": 0.81701374,
      "num_input_tokens_seen": 65979160,
      "step": 3074,
      "time_per_iteration": 2.6937832832336426
    },
    {
      "auxiliary_loss_clip": 0.01302528,
      "auxiliary_loss_mlp": 0.00999781,
      "balance_loss_clip": 1.01711917,
      "balance_loss_mlp": 0.9984991,
      "epoch": 0.3697468887152047,
      "flos": 59887257264000.0,
      "grad_norm": 0.7231319293404593,
      "language_loss": 0.56512511,
      "learning_rate": 2.9066620202209468e-06,
      "loss": 0.58814812,
      "num_input_tokens_seen": 66041650,
      "step": 3075,
      "time_per_iteration": 3.2697668075561523
    },
    {
      "auxiliary_loss_clip": 0.01337181,
      "auxiliary_loss_mlp": 0.01025533,
      "balance_loss_clip": 1.05146945,
      "balance_loss_mlp": 1.0176177,
      "epoch": 0.3698671316058438,
      "flos": 26137581569280.0,
      "grad_norm": 1.9787390718445712,
      "language_loss": 0.77660656,
      "learning_rate": 2.905967621132716e-06,
      "loss": 0.80023366,
      "num_input_tokens_seen": 66059260,
      "step": 3076,
      "time_per_iteration": 3.8384017944335938
    },
    {
      "auxiliary_loss_clip": 0.01300107,
      "auxiliary_loss_mlp": 0.01035169,
      "balance_loss_clip": 1.055089,
      "balance_loss_mlp": 1.02608526,
      "epoch": 0.3699873744964829,
      "flos": 24607464059520.0,
      "grad_norm": 1.8518153889656865,
      "language_loss": 0.75465935,
      "learning_rate": 2.9052730846099172e-06,
      "loss": 0.77801204,
      "num_input_tokens_seen": 66080605,
      "step": 3077,
      "time_per_iteration": 2.7341861724853516
    },
    {
      "auxiliary_loss_clip": 0.01186528,
      "auxiliary_loss_mlp": 0.01003414,
      "balance_loss_clip": 1.01479363,
      "balance_loss_mlp": 1.00222778,
      "epoch": 0.370107617387122,
      "flos": 64885340050560.0,
      "grad_norm": 0.8743616178666769,
      "language_loss": 0.60926485,
      "learning_rate": 2.9045784107579123e-06,
      "loss": 0.63116431,
      "num_input_tokens_seen": 66140710,
      "step": 3078,
      "time_per_iteration": 3.28112530708313
    },
    {
      "auxiliary_loss_clip": 0.01198032,
      "auxiliary_loss_mlp": 0.01032385,
      "balance_loss_clip": 1.05879128,
      "balance_loss_mlp": 1.02427256,
      "epoch": 0.37022786027776106,
      "flos": 15961683317760.0,
      "grad_norm": 1.9241490880572505,
      "language_loss": 0.67293692,
      "learning_rate": 2.9038835996820807e-06,
      "loss": 0.69524109,
      "num_input_tokens_seen": 66158320,
      "step": 3079,
      "time_per_iteration": 2.6218249797821045
    },
    {
      "auxiliary_loss_clip": 0.0134991,
      "auxiliary_loss_mlp": 0.01028376,
      "balance_loss_clip": 1.04938388,
      "balance_loss_mlp": 1.020872,
      "epoch": 0.37034810316840017,
      "flos": 18546927863040.0,
      "grad_norm": 4.480648710542528,
      "language_loss": 0.7947945,
      "learning_rate": 2.903188651487826e-06,
      "loss": 0.81857729,
      "num_input_tokens_seen": 66176875,
      "step": 3080,
      "time_per_iteration": 3.6267547607421875
    },
    {
      "auxiliary_loss_clip": 0.01155194,
      "auxiliary_loss_mlp": 0.01025831,
      "balance_loss_clip": 1.05592275,
      "balance_loss_mlp": 1.01799285,
      "epoch": 0.3704683460590393,
      "flos": 17821927751040.0,
      "grad_norm": 2.3008562053552364,
      "language_loss": 0.86778522,
      "learning_rate": 2.902493566280571e-06,
      "loss": 0.88959551,
      "num_input_tokens_seen": 66194980,
      "step": 3081,
      "time_per_iteration": 2.651834726333618
    },
    {
      "auxiliary_loss_clip": 0.01292429,
      "auxiliary_loss_mlp": 0.01035523,
      "balance_loss_clip": 1.05344951,
      "balance_loss_mlp": 1.02742839,
      "epoch": 0.37058858894967833,
      "flos": 14134081368960.0,
      "grad_norm": 5.899975652218037,
      "language_loss": 0.81485987,
      "learning_rate": 2.9017983441657595e-06,
      "loss": 0.83813941,
      "num_input_tokens_seen": 66212310,
      "step": 3082,
      "time_per_iteration": 2.6492788791656494
    },
    {
      "auxiliary_loss_clip": 0.0139798,
      "auxiliary_loss_mlp": 0.01027407,
      "balance_loss_clip": 1.04978466,
      "balance_loss_mlp": 1.01965809,
      "epoch": 0.37070883184031744,
      "flos": 13954492344960.0,
      "grad_norm": 2.144956956260684,
      "language_loss": 0.75568271,
      "learning_rate": 2.9011029852488564e-06,
      "loss": 0.77993667,
      "num_input_tokens_seen": 66229545,
      "step": 3083,
      "time_per_iteration": 2.7589683532714844
    },
    {
      "auxiliary_loss_clip": 0.01082258,
      "auxiliary_loss_mlp": 0.01004482,
      "balance_loss_clip": 1.01352,
      "balance_loss_mlp": 1.00324774,
      "epoch": 0.37082907473095655,
      "flos": 52315419306240.0,
      "grad_norm": 1.195926501188446,
      "language_loss": 0.6247679,
      "learning_rate": 2.9004074896353465e-06,
      "loss": 0.64563537,
      "num_input_tokens_seen": 66283545,
      "step": 3084,
      "time_per_iteration": 3.069413423538208
    },
    {
      "auxiliary_loss_clip": 0.01198956,
      "auxiliary_loss_mlp": 0.01026187,
      "balance_loss_clip": 1.06056952,
      "balance_loss_mlp": 1.01868856,
      "epoch": 0.3709493176215956,
      "flos": 15998096730240.0,
      "grad_norm": 3.7836233234078427,
      "language_loss": 0.81544781,
      "learning_rate": 2.8997118574307362e-06,
      "loss": 0.83769923,
      "num_input_tokens_seen": 66300500,
      "step": 3085,
      "time_per_iteration": 3.4709956645965576
    },
    {
      "auxiliary_loss_clip": 0.01259322,
      "auxiliary_loss_mlp": 0.01030927,
      "balance_loss_clip": 1.05531073,
      "balance_loss_mlp": 1.02293968,
      "epoch": 0.3710695605122347,
      "flos": 20959837931520.0,
      "grad_norm": 2.526471752998361,
      "language_loss": 0.74850541,
      "learning_rate": 2.899016088740553e-06,
      "loss": 0.77140784,
      "num_input_tokens_seen": 66318610,
      "step": 3086,
      "time_per_iteration": 2.747607707977295
    },
    {
      "auxiliary_loss_clip": 0.01390198,
      "auxiliary_loss_mlp": 0.01030989,
      "balance_loss_clip": 1.05104637,
      "balance_loss_mlp": 1.02297783,
      "epoch": 0.37118980340287383,
      "flos": 14355578586240.0,
      "grad_norm": 2.6973625649504323,
      "language_loss": 0.79330689,
      "learning_rate": 2.898320183670344e-06,
      "loss": 0.81751877,
      "num_input_tokens_seen": 66336025,
      "step": 3087,
      "time_per_iteration": 3.682981491088867
    },
    {
      "auxiliary_loss_clip": 0.01398417,
      "auxiliary_loss_mlp": 0.01026823,
      "balance_loss_clip": 1.05601668,
      "balance_loss_mlp": 1.01848996,
      "epoch": 0.3713100462935129,
      "flos": 25885381201920.0,
      "grad_norm": 1.7992158890471055,
      "language_loss": 0.89219445,
      "learning_rate": 2.8976241423256767e-06,
      "loss": 0.91644686,
      "num_input_tokens_seen": 66356120,
      "step": 3088,
      "time_per_iteration": 2.821410655975342
    },
    {
      "auxiliary_loss_clip": 0.01296955,
      "auxiliary_loss_mlp": 0.0103079,
      "balance_loss_clip": 1.05380249,
      "balance_loss_mlp": 1.02381957,
      "epoch": 0.371430289184152,
      "flos": 30518934814080.0,
      "grad_norm": 2.289706271751039,
      "language_loss": 0.68592548,
      "learning_rate": 2.896927964812142e-06,
      "loss": 0.70920295,
      "num_input_tokens_seen": 66376685,
      "step": 3089,
      "time_per_iteration": 2.7800779342651367
    },
    {
      "auxiliary_loss_clip": 0.01304061,
      "auxiliary_loss_mlp": 0.01039065,
      "balance_loss_clip": 1.06167912,
      "balance_loss_mlp": 1.03088689,
      "epoch": 0.37155053207479105,
      "flos": 15742233175680.0,
      "grad_norm": 2.9090977564580873,
      "language_loss": 0.75212574,
      "learning_rate": 2.8962316512353465e-06,
      "loss": 0.77555704,
      "num_input_tokens_seen": 66394230,
      "step": 3090,
      "time_per_iteration": 2.6838791370391846
    },
    {
      "auxiliary_loss_clip": 0.01438928,
      "auxiliary_loss_mlp": 0.0102249,
      "balance_loss_clip": 1.04701364,
      "balance_loss_mlp": 1.01462853,
      "epoch": 0.37167077496543016,
      "flos": 23404061681280.0,
      "grad_norm": 1.7279539702096047,
      "language_loss": 0.74792802,
      "learning_rate": 2.8955352017009233e-06,
      "loss": 0.77254224,
      "num_input_tokens_seen": 66413475,
      "step": 3091,
      "time_per_iteration": 2.8148629665374756
    },
    {
      "auxiliary_loss_clip": 0.01294858,
      "auxiliary_loss_mlp": 0.01025031,
      "balance_loss_clip": 1.05614817,
      "balance_loss_mlp": 1.01682627,
      "epoch": 0.3717910178560693,
      "flos": 22088653718400.0,
      "grad_norm": 1.9092540274145933,
      "language_loss": 0.77356136,
      "learning_rate": 2.8948386163145212e-06,
      "loss": 0.7967602,
      "num_input_tokens_seen": 66432685,
      "step": 3092,
      "time_per_iteration": 2.6916961669921875
    },
    {
      "auxiliary_loss_clip": 0.01156802,
      "auxiliary_loss_mlp": 0.01029503,
      "balance_loss_clip": 1.05648839,
      "balance_loss_mlp": 1.02173638,
      "epoch": 0.3719112607467083,
      "flos": 26939969533440.0,
      "grad_norm": 1.8090157883618678,
      "language_loss": 0.79772913,
      "learning_rate": 2.8941418951818135e-06,
      "loss": 0.81959224,
      "num_input_tokens_seen": 66452245,
      "step": 3093,
      "time_per_iteration": 2.666903495788574
    },
    {
      "auxiliary_loss_clip": 0.01346208,
      "auxiliary_loss_mlp": 0.01026458,
      "balance_loss_clip": 1.05155182,
      "balance_loss_mlp": 1.0184834,
      "epoch": 0.37203150363734744,
      "flos": 12166500119040.0,
      "grad_norm": 2.1477107825996744,
      "language_loss": 0.71682936,
      "learning_rate": 2.8934450384084903e-06,
      "loss": 0.740556,
      "num_input_tokens_seen": 66469760,
      "step": 3094,
      "time_per_iteration": 2.699648857116699
    },
    {
      "auxiliary_loss_clip": 0.01290577,
      "auxiliary_loss_mlp": 0.01030069,
      "balance_loss_clip": 1.0495975,
      "balance_loss_mlp": 1.02132463,
      "epoch": 0.37215174652798655,
      "flos": 23697595624320.0,
      "grad_norm": 2.0367096308028727,
      "language_loss": 0.70473707,
      "learning_rate": 2.8927480461002653e-06,
      "loss": 0.72794348,
      "num_input_tokens_seen": 66489730,
      "step": 3095,
      "time_per_iteration": 2.7301511764526367
    },
    {
      "auxiliary_loss_clip": 0.01298781,
      "auxiliary_loss_mlp": 0.01029303,
      "balance_loss_clip": 1.0519942,
      "balance_loss_mlp": 1.01987362,
      "epoch": 0.3722719894186256,
      "flos": 17887751424000.0,
      "grad_norm": 2.5379591238931,
      "language_loss": 0.8609944,
      "learning_rate": 2.892050918362872e-06,
      "loss": 0.8842752,
      "num_input_tokens_seen": 66504785,
      "step": 3096,
      "time_per_iteration": 2.6489176750183105
    },
    {
      "auxiliary_loss_clip": 0.01396371,
      "auxiliary_loss_mlp": 0.0100313,
      "balance_loss_clip": 1.01463532,
      "balance_loss_mlp": 1.00176549,
      "epoch": 0.3723922323092647,
      "flos": 62419891363200.0,
      "grad_norm": 0.8528029580203966,
      "language_loss": 0.55895519,
      "learning_rate": 2.8913536553020626e-06,
      "loss": 0.58295017,
      "num_input_tokens_seen": 66558840,
      "step": 3097,
      "time_per_iteration": 3.5380163192749023
    },
    {
      "auxiliary_loss_clip": 0.01388713,
      "auxiliary_loss_mlp": 0.01030082,
      "balance_loss_clip": 1.04728186,
      "balance_loss_mlp": 1.02188659,
      "epoch": 0.3725124751999038,
      "flos": 23039747988480.0,
      "grad_norm": 1.914132469785722,
      "language_loss": 0.84810507,
      "learning_rate": 2.8906562570236137e-06,
      "loss": 0.87229311,
      "num_input_tokens_seen": 66576750,
      "step": 3098,
      "time_per_iteration": 3.1141517162323
    },
    {
      "auxiliary_loss_clip": 0.01439876,
      "auxiliary_loss_mlp": 0.01034449,
      "balance_loss_clip": 1.04362273,
      "balance_loss_mlp": 1.0265038,
      "epoch": 0.3726327180905429,
      "flos": 20920551431040.0,
      "grad_norm": 4.7718567569279715,
      "language_loss": 0.76615345,
      "learning_rate": 2.889958723633318e-06,
      "loss": 0.79089671,
      "num_input_tokens_seen": 66595690,
      "step": 3099,
      "time_per_iteration": 2.755072832107544
    },
    {
      "auxiliary_loss_clip": 0.01341618,
      "auxiliary_loss_mlp": 0.01033716,
      "balance_loss_clip": 1.05243802,
      "balance_loss_mlp": 1.02605736,
      "epoch": 0.372752960981182,
      "flos": 30592156688640.0,
      "grad_norm": 1.7699201148811547,
      "language_loss": 0.73807693,
      "learning_rate": 2.889261055236992e-06,
      "loss": 0.76183033,
      "num_input_tokens_seen": 66617905,
      "step": 3100,
      "time_per_iteration": 2.795851469039917
    },
    {
      "auxiliary_loss_clip": 0.01291952,
      "auxiliary_loss_mlp": 0.01031177,
      "balance_loss_clip": 1.05464101,
      "balance_loss_mlp": 1.02289832,
      "epoch": 0.3728732038718211,
      "flos": 25116749043840.0,
      "grad_norm": 1.7734284478473232,
      "language_loss": 0.82616132,
      "learning_rate": 2.8885632519404704e-06,
      "loss": 0.84939259,
      "num_input_tokens_seen": 66638175,
      "step": 3101,
      "time_per_iteration": 2.679410457611084
    },
    {
      "auxiliary_loss_clip": 0.01289231,
      "auxiliary_loss_mlp": 0.01031533,
      "balance_loss_clip": 1.05328918,
      "balance_loss_mlp": 1.02306318,
      "epoch": 0.37299344676246016,
      "flos": 25302048330240.0,
      "grad_norm": 2.2528523885958163,
      "language_loss": 0.76134902,
      "learning_rate": 2.8878653138496107e-06,
      "loss": 0.78455669,
      "num_input_tokens_seen": 66658670,
      "step": 3102,
      "time_per_iteration": 3.60315203666687
    },
    {
      "auxiliary_loss_clip": 0.01349964,
      "auxiliary_loss_mlp": 0.01029419,
      "balance_loss_clip": 1.04523492,
      "balance_loss_mlp": 1.02113676,
      "epoch": 0.37311368965309927,
      "flos": 23842531002240.0,
      "grad_norm": 2.7499521402208984,
      "language_loss": 0.76064837,
      "learning_rate": 2.8871672410702878e-06,
      "loss": 0.78444219,
      "num_input_tokens_seen": 66676030,
      "step": 3103,
      "time_per_iteration": 2.844857931137085
    },
    {
      "auxiliary_loss_clip": 0.01352721,
      "auxiliary_loss_mlp": 0.01030336,
      "balance_loss_clip": 1.05215454,
      "balance_loss_mlp": 1.02156234,
      "epoch": 0.3732339325437384,
      "flos": 25811943845760.0,
      "grad_norm": 1.6937147857160744,
      "language_loss": 0.81905711,
      "learning_rate": 2.8864690337084008e-06,
      "loss": 0.8428877,
      "num_input_tokens_seen": 66695305,
      "step": 3104,
      "time_per_iteration": 2.762549638748169
    },
    {
      "auxiliary_loss_clip": 0.01244234,
      "auxiliary_loss_mlp": 0.01029069,
      "balance_loss_clip": 1.05246103,
      "balance_loss_mlp": 1.0205965,
      "epoch": 0.37335417543437743,
      "flos": 26208433146240.0,
      "grad_norm": 1.7719854952291259,
      "language_loss": 0.78188145,
      "learning_rate": 2.885770691869866e-06,
      "loss": 0.80461454,
      "num_input_tokens_seen": 66716185,
      "step": 3105,
      "time_per_iteration": 2.6865594387054443
    },
    {
      "auxiliary_loss_clip": 0.0124102,
      "auxiliary_loss_mlp": 0.01030409,
      "balance_loss_clip": 1.05482113,
      "balance_loss_mlp": 1.02247226,
      "epoch": 0.37347441832501654,
      "flos": 24023879792640.0,
      "grad_norm": 2.3916393848429314,
      "language_loss": 0.75099933,
      "learning_rate": 2.8850722156606207e-06,
      "loss": 0.77371359,
      "num_input_tokens_seen": 66734575,
      "step": 3106,
      "time_per_iteration": 3.9893603324890137
    },
    {
      "auxiliary_loss_clip": 0.01238584,
      "auxiliary_loss_mlp": 0.01030229,
      "balance_loss_clip": 1.05217052,
      "balance_loss_mlp": 1.02233779,
      "epoch": 0.3735946612156556,
      "flos": 19714922409600.0,
      "grad_norm": 1.67742768756399,
      "language_loss": 0.67408931,
      "learning_rate": 2.8843736051866252e-06,
      "loss": 0.69677746,
      "num_input_tokens_seen": 66753500,
      "step": 3107,
      "time_per_iteration": 2.6693367958068848
    },
    {
      "auxiliary_loss_clip": 0.01389417,
      "auxiliary_loss_mlp": 0.02566978,
      "balance_loss_clip": 1.04869378,
      "balance_loss_mlp": 0.99999666,
      "epoch": 0.3737149041062947,
      "flos": 23039604334080.0,
      "grad_norm": 1.7957007859532768,
      "language_loss": 0.69501257,
      "learning_rate": 2.8836748605538557e-06,
      "loss": 0.73457646,
      "num_input_tokens_seen": 66775140,
      "step": 3108,
      "time_per_iteration": 2.9054551124572754
    },
    {
      "auxiliary_loss_clip": 0.01296032,
      "auxiliary_loss_mlp": 0.01029348,
      "balance_loss_clip": 1.05068398,
      "balance_loss_mlp": 1.02095544,
      "epoch": 0.3738351469969338,
      "flos": 34678108483200.0,
      "grad_norm": 3.5132068858955714,
      "language_loss": 0.63786203,
      "learning_rate": 2.882975981868313e-06,
      "loss": 0.66111577,
      "num_input_tokens_seen": 66795525,
      "step": 3109,
      "time_per_iteration": 2.8151938915252686
    },
    {
      "auxiliary_loss_clip": 0.01245718,
      "auxiliary_loss_mlp": 0.01033621,
      "balance_loss_clip": 1.05658841,
      "balance_loss_mlp": 1.02536607,
      "epoch": 0.3739553898875729,
      "flos": 43507967448960.0,
      "grad_norm": 2.3392718075339065,
      "language_loss": 0.68746674,
      "learning_rate": 2.882276969236016e-06,
      "loss": 0.71026015,
      "num_input_tokens_seen": 66816885,
      "step": 3110,
      "time_per_iteration": 2.8545894622802734
    },
    {
      "auxiliary_loss_clip": 0.0129079,
      "auxiliary_loss_mlp": 0.01035183,
      "balance_loss_clip": 1.05236673,
      "balance_loss_mlp": 1.02660584,
      "epoch": 0.374075632778212,
      "flos": 12856487448960.0,
      "grad_norm": 2.077524333029217,
      "language_loss": 0.76532888,
      "learning_rate": 2.881577822763005e-06,
      "loss": 0.78858864,
      "num_input_tokens_seen": 66834835,
      "step": 3111,
      "time_per_iteration": 3.7685089111328125
    },
    {
      "auxiliary_loss_clip": 0.01243097,
      "auxiliary_loss_mlp": 0.01025701,
      "balance_loss_clip": 1.05419338,
      "balance_loss_mlp": 1.01819086,
      "epoch": 0.3741958756688511,
      "flos": 26024031699840.0,
      "grad_norm": 2.040081417978842,
      "language_loss": 0.87721491,
      "learning_rate": 2.880878542555338e-06,
      "loss": 0.89990294,
      "num_input_tokens_seen": 66852600,
      "step": 3112,
      "time_per_iteration": 3.6105687618255615
    },
    {
      "auxiliary_loss_clip": 0.01197089,
      "auxiliary_loss_mlp": 0.01035102,
      "balance_loss_clip": 1.05690122,
      "balance_loss_mlp": 1.02688193,
      "epoch": 0.37431611855949015,
      "flos": 21433894652160.0,
      "grad_norm": 1.992005447087093,
      "language_loss": 0.79920161,
      "learning_rate": 2.8801791287190976e-06,
      "loss": 0.82152355,
      "num_input_tokens_seen": 66870595,
      "step": 3113,
      "time_per_iteration": 2.6409547328948975
    },
    {
      "auxiliary_loss_clip": 0.01247387,
      "auxiliary_loss_mlp": 0.01030107,
      "balance_loss_clip": 1.05537987,
      "balance_loss_mlp": 1.02176237,
      "epoch": 0.37443636145012926,
      "flos": 24207096090240.0,
      "grad_norm": 2.62560756454543,
      "language_loss": 0.8585099,
      "learning_rate": 2.8794795813603817e-06,
      "loss": 0.88128489,
      "num_input_tokens_seen": 66886060,
      "step": 3114,
      "time_per_iteration": 2.659003734588623
    },
    {
      "auxiliary_loss_clip": 0.01154738,
      "auxiliary_loss_mlp": 0.01029889,
      "balance_loss_clip": 1.05449486,
      "balance_loss_mlp": 1.02178299,
      "epoch": 0.3745566043407684,
      "flos": 15378601841280.0,
      "grad_norm": 2.7349768325367254,
      "language_loss": 0.81646311,
      "learning_rate": 2.878779900585314e-06,
      "loss": 0.83830941,
      "num_input_tokens_seen": 66903900,
      "step": 3115,
      "time_per_iteration": 2.6319336891174316
    },
    {
      "auxiliary_loss_clip": 0.01303197,
      "auxiliary_loss_mlp": 0.01028438,
      "balance_loss_clip": 1.05706215,
      "balance_loss_mlp": 1.02076054,
      "epoch": 0.37467684723140743,
      "flos": 24608218245120.0,
      "grad_norm": 1.495998963800741,
      "language_loss": 0.75345874,
      "learning_rate": 2.8780800865000336e-06,
      "loss": 0.77677506,
      "num_input_tokens_seen": 66925210,
      "step": 3116,
      "time_per_iteration": 2.6904497146606445
    },
    {
      "auxiliary_loss_clip": 0.01138138,
      "auxiliary_loss_mlp": 0.01014216,
      "balance_loss_clip": 1.01648426,
      "balance_loss_mlp": 1.0128926,
      "epoch": 0.37479709012204654,
      "flos": 64377491610240.0,
      "grad_norm": 0.9824208484623803,
      "language_loss": 0.59165329,
      "learning_rate": 2.877380139210702e-06,
      "loss": 0.61317682,
      "num_input_tokens_seen": 66983880,
      "step": 3117,
      "time_per_iteration": 3.1853835582733154
    },
    {
      "auxiliary_loss_clip": 0.01251228,
      "auxiliary_loss_mlp": 0.01026337,
      "balance_loss_clip": 1.05288625,
      "balance_loss_mlp": 1.01813555,
      "epoch": 0.37491733301268565,
      "flos": 23803962773760.0,
      "grad_norm": 1.8029654920250584,
      "language_loss": 0.76583827,
      "learning_rate": 2.876680058823501e-06,
      "loss": 0.78861392,
      "num_input_tokens_seen": 67004280,
      "step": 3118,
      "time_per_iteration": 2.790881633758545
    },
    {
      "auxiliary_loss_clip": 0.01286962,
      "auxiliary_loss_mlp": 0.01026448,
      "balance_loss_clip": 1.05060422,
      "balance_loss_mlp": 1.01822233,
      "epoch": 0.3750375759033247,
      "flos": 32160950167680.0,
      "grad_norm": 2.0939301429068666,
      "language_loss": 0.6567471,
      "learning_rate": 2.8759798454446314e-06,
      "loss": 0.6798811,
      "num_input_tokens_seen": 67027445,
      "step": 3119,
      "time_per_iteration": 2.7498703002929688
    },
    {
      "auxiliary_loss_clip": 0.01246445,
      "auxiliary_loss_mlp": 0.01027214,
      "balance_loss_clip": 1.05435133,
      "balance_loss_mlp": 1.02020168,
      "epoch": 0.3751578187939638,
      "flos": 23367791923200.0,
      "grad_norm": 2.2192393752751434,
      "language_loss": 0.81658167,
      "learning_rate": 2.8752794991803173e-06,
      "loss": 0.83931828,
      "num_input_tokens_seen": 67045130,
      "step": 3120,
      "time_per_iteration": 2.7061307430267334
    },
    {
      "auxiliary_loss_clip": 0.01292204,
      "auxiliary_loss_mlp": 0.01024919,
      "balance_loss_clip": 1.05344963,
      "balance_loss_mlp": 1.01771259,
      "epoch": 0.37527806168460287,
      "flos": 14605731878400.0,
      "grad_norm": 2.0237174505027493,
      "language_loss": 0.75258696,
      "learning_rate": 2.8745790201367976e-06,
      "loss": 0.77575815,
      "num_input_tokens_seen": 67060885,
      "step": 3121,
      "time_per_iteration": 2.6293084621429443
    },
    {
      "auxiliary_loss_clip": 0.01196582,
      "auxiliary_loss_mlp": 0.01033841,
      "balance_loss_clip": 1.05515063,
      "balance_loss_mlp": 1.02533531,
      "epoch": 0.375398304575242,
      "flos": 26390823431040.0,
      "grad_norm": 2.718271538856479,
      "language_loss": 0.84327686,
      "learning_rate": 2.8738784084203373e-06,
      "loss": 0.8655811,
      "num_input_tokens_seen": 67080960,
      "step": 3122,
      "time_per_iteration": 2.660172462463379
    },
    {
      "auxiliary_loss_clip": 0.01286623,
      "auxiliary_loss_mlp": 0.01024802,
      "balance_loss_clip": 1.04661965,
      "balance_loss_mlp": 1.0174942,
      "epoch": 0.3755185474658811,
      "flos": 22236605838720.0,
      "grad_norm": 1.8275815639332953,
      "language_loss": 0.78874111,
      "learning_rate": 2.873177664137216e-06,
      "loss": 0.81185538,
      "num_input_tokens_seen": 67101890,
      "step": 3123,
      "time_per_iteration": 2.6609041690826416
    },
    {
      "auxiliary_loss_clip": 0.01335757,
      "auxiliary_loss_mlp": 0.01027148,
      "balance_loss_clip": 1.05184495,
      "balance_loss_mlp": 1.01953065,
      "epoch": 0.37563879035652015,
      "flos": 30812935633920.0,
      "grad_norm": 2.1523622686703865,
      "language_loss": 0.6914109,
      "learning_rate": 2.8724767873937384e-06,
      "loss": 0.71503991,
      "num_input_tokens_seen": 67126010,
      "step": 3124,
      "time_per_iteration": 2.814842462539673
    },
    {
      "auxiliary_loss_clip": 0.01198816,
      "auxiliary_loss_mlp": 0.01022119,
      "balance_loss_clip": 1.0533185,
      "balance_loss_mlp": 1.01488328,
      "epoch": 0.37575903324715926,
      "flos": 20773533064320.0,
      "grad_norm": 2.1283314274975895,
      "language_loss": 0.8745296,
      "learning_rate": 2.871775778296225e-06,
      "loss": 0.89673901,
      "num_input_tokens_seen": 67143100,
      "step": 3125,
      "time_per_iteration": 2.6530778408050537
    },
    {
      "auxiliary_loss_clip": 0.01252174,
      "auxiliary_loss_mlp": 0.01035372,
      "balance_loss_clip": 1.06002307,
      "balance_loss_mlp": 1.02680135,
      "epoch": 0.37587927613779837,
      "flos": 18697681244160.0,
      "grad_norm": 2.919054888179233,
      "language_loss": 0.78600943,
      "learning_rate": 2.8710746369510196e-06,
      "loss": 0.80888498,
      "num_input_tokens_seen": 67161085,
      "step": 3126,
      "time_per_iteration": 2.670923948287964
    },
    {
      "auxiliary_loss_clip": 0.01293505,
      "auxiliary_loss_mlp": 0.0102755,
      "balance_loss_clip": 1.05259824,
      "balance_loss_mlp": 1.01932502,
      "epoch": 0.3759995190284374,
      "flos": 13624796384640.0,
      "grad_norm": 2.5161274112388274,
      "language_loss": 0.83141422,
      "learning_rate": 2.8703733634644846e-06,
      "loss": 0.85462475,
      "num_input_tokens_seen": 67175840,
      "step": 3127,
      "time_per_iteration": 2.622692108154297
    },
    {
      "auxiliary_loss_clip": 0.01194211,
      "auxiliary_loss_mlp": 0.01027829,
      "balance_loss_clip": 1.05673993,
      "balance_loss_mlp": 1.02056909,
      "epoch": 0.37611976191907653,
      "flos": 20484847457280.0,
      "grad_norm": 1.6684415983593122,
      "language_loss": 0.7956351,
      "learning_rate": 2.869671957943002e-06,
      "loss": 0.81785548,
      "num_input_tokens_seen": 67194995,
      "step": 3128,
      "time_per_iteration": 3.59295654296875
    },
    {
      "auxiliary_loss_clip": 0.01296719,
      "auxiliary_loss_mlp": 0.01028124,
      "balance_loss_clip": 1.06042218,
      "balance_loss_mlp": 1.02008343,
      "epoch": 0.37624000480971564,
      "flos": 21141797253120.0,
      "grad_norm": 3.2419200957551055,
      "language_loss": 0.74479777,
      "learning_rate": 2.8689704204929747e-06,
      "loss": 0.7680462,
      "num_input_tokens_seen": 67214175,
      "step": 3129,
      "time_per_iteration": 2.682708740234375
    },
    {
      "auxiliary_loss_clip": 0.01195503,
      "auxiliary_loss_mlp": 0.0102328,
      "balance_loss_clip": 1.05721784,
      "balance_loss_mlp": 1.01538837,
      "epoch": 0.3763602477003547,
      "flos": 22564470205440.0,
      "grad_norm": 2.4302835105834197,
      "language_loss": 0.81308985,
      "learning_rate": 2.8682687512208253e-06,
      "loss": 0.83527762,
      "num_input_tokens_seen": 67233185,
      "step": 3130,
      "time_per_iteration": 2.611602306365967
    },
    {
      "auxiliary_loss_clip": 0.01249164,
      "auxiliary_loss_mlp": 0.01027808,
      "balance_loss_clip": 1.05345941,
      "balance_loss_mlp": 1.02016711,
      "epoch": 0.3764804905909938,
      "flos": 27526857851520.0,
      "grad_norm": 1.9192060572584764,
      "language_loss": 0.80653882,
      "learning_rate": 2.8675669502329972e-06,
      "loss": 0.82930851,
      "num_input_tokens_seen": 67254715,
      "step": 3131,
      "time_per_iteration": 2.7191195487976074
    },
    {
      "auxiliary_loss_clip": 0.01240999,
      "auxiliary_loss_mlp": 0.0256667,
      "balance_loss_clip": 1.05388892,
      "balance_loss_mlp": 1.00002372,
      "epoch": 0.3766007334816329,
      "flos": 22528092706560.0,
      "grad_norm": 2.3338040436484913,
      "language_loss": 0.85942608,
      "learning_rate": 2.866865017635952e-06,
      "loss": 0.89750278,
      "num_input_tokens_seen": 67272535,
      "step": 3132,
      "time_per_iteration": 2.643761157989502
    },
    {
      "auxiliary_loss_clip": 0.01347154,
      "auxiliary_loss_mlp": 0.01023275,
      "balance_loss_clip": 1.05744028,
      "balance_loss_mlp": 1.01581895,
      "epoch": 0.376720976372272,
      "flos": 25957166532480.0,
      "grad_norm": 1.891223283197314,
      "language_loss": 0.7935999,
      "learning_rate": 2.866162953536174e-06,
      "loss": 0.81730419,
      "num_input_tokens_seen": 67293505,
      "step": 3133,
      "time_per_iteration": 3.6602272987365723
    },
    {
      "auxiliary_loss_clip": 0.01294435,
      "auxiliary_loss_mlp": 0.02566697,
      "balance_loss_clip": 1.05204868,
      "balance_loss_mlp": 0.99994934,
      "epoch": 0.3768412192629111,
      "flos": 18041162411520.0,
      "grad_norm": 1.7917135653399359,
      "language_loss": 0.75303805,
      "learning_rate": 2.8654607580401634e-06,
      "loss": 0.79164934,
      "num_input_tokens_seen": 67313240,
      "step": 3134,
      "time_per_iteration": 2.7213101387023926
    },
    {
      "auxiliary_loss_clip": 0.01136867,
      "auxiliary_loss_mlp": 0.01005014,
      "balance_loss_clip": 1.01990485,
      "balance_loss_mlp": 1.0037322,
      "epoch": 0.3769614621535502,
      "flos": 62989472304000.0,
      "grad_norm": 0.882991127663658,
      "language_loss": 0.65172124,
      "learning_rate": 2.8647584312544446e-06,
      "loss": 0.67314005,
      "num_input_tokens_seen": 67378445,
      "step": 3135,
      "time_per_iteration": 3.2497687339782715
    },
    {
      "auxiliary_loss_clip": 0.01245077,
      "auxiliary_loss_mlp": 0.02566613,
      "balance_loss_clip": 1.05020785,
      "balance_loss_mlp": 1.00002694,
      "epoch": 0.37708170504418925,
      "flos": 23661685002240.0,
      "grad_norm": 1.7114961906062856,
      "language_loss": 0.8569665,
      "learning_rate": 2.864055973285559e-06,
      "loss": 0.89508343,
      "num_input_tokens_seen": 67400445,
      "step": 3136,
      "time_per_iteration": 2.8739869594573975
    },
    {
      "auxiliary_loss_clip": 0.01284257,
      "auxiliary_loss_mlp": 0.01024324,
      "balance_loss_clip": 1.05091476,
      "balance_loss_mlp": 1.01713896,
      "epoch": 0.37720194793482836,
      "flos": 24423170353920.0,
      "grad_norm": 1.9322845059485376,
      "language_loss": 0.86773467,
      "learning_rate": 2.8633533842400698e-06,
      "loss": 0.8908205,
      "num_input_tokens_seen": 67420645,
      "step": 3137,
      "time_per_iteration": 3.632336139678955
    },
    {
      "auxiliary_loss_clip": 0.01247968,
      "auxiliary_loss_mlp": 0.02573441,
      "balance_loss_clip": 1.05722773,
      "balance_loss_mlp": 1.00001621,
      "epoch": 0.3773221908254674,
      "flos": 20996502739200.0,
      "grad_norm": 1.8070396840476057,
      "language_loss": 0.77437145,
      "learning_rate": 2.862650664224558e-06,
      "loss": 0.81258559,
      "num_input_tokens_seen": 67439495,
      "step": 3138,
      "time_per_iteration": 3.566664457321167
    },
    {
      "auxiliary_loss_clip": 0.0124159,
      "auxiliary_loss_mlp": 0.01030329,
      "balance_loss_clip": 1.05780482,
      "balance_loss_mlp": 1.02297688,
      "epoch": 0.37744243371610653,
      "flos": 37631724958080.0,
      "grad_norm": 1.5839133227275146,
      "language_loss": 0.70062917,
      "learning_rate": 2.861947813345627e-06,
      "loss": 0.72334838,
      "num_input_tokens_seen": 67462195,
      "step": 3139,
      "time_per_iteration": 2.8184397220611572
    },
    {
      "auxiliary_loss_clip": 0.01196272,
      "auxiliary_loss_mlp": 0.02569363,
      "balance_loss_clip": 1.05697012,
      "balance_loss_mlp": 0.9999373,
      "epoch": 0.37756267660674564,
      "flos": 26140526484480.0,
      "grad_norm": 1.8436772044479168,
      "language_loss": 0.72758913,
      "learning_rate": 2.8612448317098974e-06,
      "loss": 0.7652455,
      "num_input_tokens_seen": 67482530,
      "step": 3140,
      "time_per_iteration": 2.655693292617798
    },
    {
      "auxiliary_loss_clip": 0.01340797,
      "auxiliary_loss_mlp": 0.0257037,
      "balance_loss_clip": 1.05040908,
      "balance_loss_mlp": 1.00009668,
      "epoch": 0.3776829194973847,
      "flos": 19427888828160.0,
      "grad_norm": 2.1037717597282524,
      "language_loss": 0.82816637,
      "learning_rate": 2.8605417194240114e-06,
      "loss": 0.86727804,
      "num_input_tokens_seen": 67500890,
      "step": 3141,
      "time_per_iteration": 2.8008902072906494
    },
    {
      "auxiliary_loss_clip": 0.01237236,
      "auxiliary_loss_mlp": 0.01029636,
      "balance_loss_clip": 1.05374956,
      "balance_loss_mlp": 1.0227282,
      "epoch": 0.3778031623880238,
      "flos": 17382309194880.0,
      "grad_norm": 2.030626725126595,
      "language_loss": 0.791673,
      "learning_rate": 2.8598384765946315e-06,
      "loss": 0.81434178,
      "num_input_tokens_seen": 67519545,
      "step": 3142,
      "time_per_iteration": 2.5639336109161377
    },
    {
      "auxiliary_loss_clip": 0.0119288,
      "auxiliary_loss_mlp": 0.01022158,
      "balance_loss_clip": 1.05391479,
      "balance_loss_mlp": 1.01538372,
      "epoch": 0.3779234052786629,
      "flos": 27125843437440.0,
      "grad_norm": 2.0577253569409284,
      "language_loss": 0.71877658,
      "learning_rate": 2.8591351033284377e-06,
      "loss": 0.74092698,
      "num_input_tokens_seen": 67539275,
      "step": 3143,
      "time_per_iteration": 2.647655487060547
    },
    {
      "auxiliary_loss_clip": 0.01150138,
      "auxiliary_loss_mlp": 0.01030346,
      "balance_loss_clip": 1.05262876,
      "balance_loss_mlp": 1.02271914,
      "epoch": 0.37804364816930197,
      "flos": 19682639061120.0,
      "grad_norm": 2.3815263613026723,
      "language_loss": 0.83949804,
      "learning_rate": 2.8584315997321325e-06,
      "loss": 0.86130291,
      "num_input_tokens_seen": 67558280,
      "step": 3144,
      "time_per_iteration": 2.629415273666382
    },
    {
      "auxiliary_loss_clip": 0.01194894,
      "auxiliary_loss_mlp": 0.01026905,
      "balance_loss_clip": 1.05504751,
      "balance_loss_mlp": 1.01924539,
      "epoch": 0.3781638910599411,
      "flos": 22702905221760.0,
      "grad_norm": 2.3906731748634713,
      "language_loss": 0.78226721,
      "learning_rate": 2.8577279659124356e-06,
      "loss": 0.8044852,
      "num_input_tokens_seen": 67575955,
      "step": 3145,
      "time_per_iteration": 2.585972547531128
    },
    {
      "auxiliary_loss_clip": 0.01241322,
      "auxiliary_loss_mlp": 0.01027107,
      "balance_loss_clip": 1.05213594,
      "balance_loss_mlp": 1.0201211,
      "epoch": 0.3782841339505802,
      "flos": 14647604158080.0,
      "grad_norm": 2.0346928515105276,
      "language_loss": 0.83364546,
      "learning_rate": 2.857024201976089e-06,
      "loss": 0.85632974,
      "num_input_tokens_seen": 67593515,
      "step": 3146,
      "time_per_iteration": 2.6354966163635254
    },
    {
      "auxiliary_loss_clip": 0.01290461,
      "auxiliary_loss_mlp": 0.01035953,
      "balance_loss_clip": 1.05247402,
      "balance_loss_mlp": 1.02793574,
      "epoch": 0.37840437684121925,
      "flos": 32818223185920.0,
      "grad_norm": 1.9085778929231576,
      "language_loss": 0.73371357,
      "learning_rate": 2.8563203080298516e-06,
      "loss": 0.75697768,
      "num_input_tokens_seen": 67614290,
      "step": 3147,
      "time_per_iteration": 2.7575082778930664
    },
    {
      "auxiliary_loss_clip": 0.01288771,
      "auxiliary_loss_mlp": 0.02567581,
      "balance_loss_clip": 1.05383587,
      "balance_loss_mlp": 1.00003767,
      "epoch": 0.37852461973185836,
      "flos": 18369206346240.0,
      "grad_norm": 2.156857685442955,
      "language_loss": 0.89474595,
      "learning_rate": 2.855616284180505e-06,
      "loss": 0.93330956,
      "num_input_tokens_seen": 67631340,
      "step": 3148,
      "time_per_iteration": 2.652336359024048
    },
    {
      "auxiliary_loss_clip": 0.01140252,
      "auxiliary_loss_mlp": 0.01002529,
      "balance_loss_clip": 1.01747847,
      "balance_loss_mlp": 1.0013907,
      "epoch": 0.37864486262249747,
      "flos": 59500680117120.0,
      "grad_norm": 0.8713301673635935,
      "language_loss": 0.6606859,
      "learning_rate": 2.8549121305348477e-06,
      "loss": 0.68211365,
      "num_input_tokens_seen": 67691125,
      "step": 3149,
      "time_per_iteration": 3.1922800540924072
    },
    {
      "auxiliary_loss_clip": 0.01245059,
      "auxiliary_loss_mlp": 0.01027958,
      "balance_loss_clip": 1.05466402,
      "balance_loss_mlp": 1.0207603,
      "epoch": 0.3787651055131365,
      "flos": 23363015414400.0,
      "grad_norm": 2.846549855291197,
      "language_loss": 0.83688384,
      "learning_rate": 2.8542078471997006e-06,
      "loss": 0.85961401,
      "num_input_tokens_seen": 67708740,
      "step": 3150,
      "time_per_iteration": 2.6453568935394287
    },
    {
      "auxiliary_loss_clip": 0.01242432,
      "auxiliary_loss_mlp": 0.01029806,
      "balance_loss_clip": 1.0529964,
      "balance_loss_mlp": 1.02259052,
      "epoch": 0.37888534840377563,
      "flos": 24601394661120.0,
      "grad_norm": 2.389372158834539,
      "language_loss": 0.7569443,
      "learning_rate": 2.8535034342819013e-06,
      "loss": 0.77966666,
      "num_input_tokens_seen": 67726150,
      "step": 3151,
      "time_per_iteration": 2.667666435241699
    },
    {
      "auxiliary_loss_clip": 0.01190215,
      "auxiliary_loss_mlp": 0.01029907,
      "balance_loss_clip": 1.05421507,
      "balance_loss_mlp": 1.02199745,
      "epoch": 0.37900559129441475,
      "flos": 23986891762560.0,
      "grad_norm": 1.5571050421978772,
      "language_loss": 0.72537053,
      "learning_rate": 2.85279889188831e-06,
      "loss": 0.74757171,
      "num_input_tokens_seen": 67746525,
      "step": 3152,
      "time_per_iteration": 2.6578457355499268
    },
    {
      "auxiliary_loss_clip": 0.01343089,
      "auxiliary_loss_mlp": 0.01026891,
      "balance_loss_clip": 1.0480082,
      "balance_loss_mlp": 1.01942253,
      "epoch": 0.3791258341850538,
      "flos": 24644667571200.0,
      "grad_norm": 2.155583056941778,
      "language_loss": 0.81233299,
      "learning_rate": 2.852094220125805e-06,
      "loss": 0.83603269,
      "num_input_tokens_seen": 67766035,
      "step": 3153,
      "time_per_iteration": 2.7316648960113525
    },
    {
      "auxiliary_loss_clip": 0.01241093,
      "auxiliary_loss_mlp": 0.01027548,
      "balance_loss_clip": 1.05580151,
      "balance_loss_mlp": 1.02024066,
      "epoch": 0.3792460770756929,
      "flos": 17420841509760.0,
      "grad_norm": 2.6476659752461997,
      "language_loss": 0.71359175,
      "learning_rate": 2.8513894191012846e-06,
      "loss": 0.73627818,
      "num_input_tokens_seen": 67785015,
      "step": 3154,
      "time_per_iteration": 3.544900894165039
    },
    {
      "auxiliary_loss_clip": 0.01194793,
      "auxiliary_loss_mlp": 0.01020546,
      "balance_loss_clip": 1.05603731,
      "balance_loss_mlp": 1.0133338,
      "epoch": 0.37936631996633197,
      "flos": 24206557386240.0,
      "grad_norm": 1.5302160615405733,
      "language_loss": 0.78800189,
      "learning_rate": 2.8506844889216664e-06,
      "loss": 0.81015533,
      "num_input_tokens_seen": 67804400,
      "step": 3155,
      "time_per_iteration": 2.6266186237335205
    },
    {
      "auxiliary_loss_clip": 0.01139036,
      "auxiliary_loss_mlp": 0.00999699,
      "balance_loss_clip": 1.02291369,
      "balance_loss_mlp": 0.99854296,
      "epoch": 0.3794865628569711,
      "flos": 70297114752000.0,
      "grad_norm": 0.8891434934520204,
      "language_loss": 0.62769395,
      "learning_rate": 2.849979429693887e-06,
      "loss": 0.64908135,
      "num_input_tokens_seen": 67865385,
      "step": 3156,
      "time_per_iteration": 3.250614881515503
    },
    {
      "auxiliary_loss_clip": 0.01191101,
      "auxiliary_loss_mlp": 0.01033087,
      "balance_loss_clip": 1.05354822,
      "balance_loss_mlp": 1.02493334,
      "epoch": 0.3796068057476102,
      "flos": 15779364860160.0,
      "grad_norm": 1.906592097760744,
      "language_loss": 0.74356413,
      "learning_rate": 2.8492742415249042e-06,
      "loss": 0.76580596,
      "num_input_tokens_seen": 67883030,
      "step": 3157,
      "time_per_iteration": 2.593364953994751
    },
    {
      "auxiliary_loss_clip": 0.01189002,
      "auxiliary_loss_mlp": 0.01025919,
      "balance_loss_clip": 1.05279267,
      "balance_loss_mlp": 1.01873112,
      "epoch": 0.37972704863824924,
      "flos": 25191694771200.0,
      "grad_norm": 1.8750817318375423,
      "language_loss": 0.76647824,
      "learning_rate": 2.848568924521694e-06,
      "loss": 0.78862745,
      "num_input_tokens_seen": 67903810,
      "step": 3158,
      "time_per_iteration": 3.5111873149871826
    },
    {
      "auxiliary_loss_clip": 0.01235449,
      "auxiliary_loss_mlp": 0.01028221,
      "balance_loss_clip": 1.04941595,
      "balance_loss_mlp": 1.02066946,
      "epoch": 0.37984729152888835,
      "flos": 26210372480640.0,
      "grad_norm": 1.8801025857005682,
      "language_loss": 0.73495674,
      "learning_rate": 2.8478634787912526e-06,
      "loss": 0.75759345,
      "num_input_tokens_seen": 67921865,
      "step": 3159,
      "time_per_iteration": 2.657728910446167
    },
    {
      "auxiliary_loss_clip": 0.01246022,
      "auxiliary_loss_mlp": 0.01031429,
      "balance_loss_clip": 1.05455279,
      "balance_loss_mlp": 1.02362061,
      "epoch": 0.37996753441952746,
      "flos": 25629302165760.0,
      "grad_norm": 2.1007527549646112,
      "language_loss": 0.76299155,
      "learning_rate": 2.847157904440596e-06,
      "loss": 0.78576601,
      "num_input_tokens_seen": 67941595,
      "step": 3160,
      "time_per_iteration": 2.695172071456909
    },
    {
      "auxiliary_loss_clip": 0.01237163,
      "auxiliary_loss_mlp": 0.01024381,
      "balance_loss_clip": 1.05264187,
      "balance_loss_mlp": 1.01703167,
      "epoch": 0.3800877773101665,
      "flos": 20118414862080.0,
      "grad_norm": 1.6507716018862728,
      "language_loss": 0.73661309,
      "learning_rate": 2.846452201576759e-06,
      "loss": 0.75922853,
      "num_input_tokens_seen": 67960970,
      "step": 3161,
      "time_per_iteration": 2.5962162017822266
    },
    {
      "auxiliary_loss_clip": 0.01192515,
      "auxiliary_loss_mlp": 0.01005517,
      "balance_loss_clip": 1.01657391,
      "balance_loss_mlp": 1.00434291,
      "epoch": 0.38020802020080563,
      "flos": 63053608037760.0,
      "grad_norm": 0.9253613530839787,
      "language_loss": 0.62794882,
      "learning_rate": 2.845746370306795e-06,
      "loss": 0.64992917,
      "num_input_tokens_seen": 68026160,
      "step": 3162,
      "time_per_iteration": 3.3010714054107666
    },
    {
      "auxiliary_loss_clip": 0.01246312,
      "auxiliary_loss_mlp": 0.0102956,
      "balance_loss_clip": 1.05513322,
      "balance_loss_mlp": 1.02210641,
      "epoch": 0.38032826309144474,
      "flos": 21288420570240.0,
      "grad_norm": 3.0571342481077943,
      "language_loss": 0.78638422,
      "learning_rate": 2.84504041073778e-06,
      "loss": 0.80914295,
      "num_input_tokens_seen": 68044575,
      "step": 3163,
      "time_per_iteration": 3.5671546459198
    },
    {
      "auxiliary_loss_clip": 0.01286925,
      "auxiliary_loss_mlp": 0.01029437,
      "balance_loss_clip": 1.05010498,
      "balance_loss_mlp": 1.02189398,
      "epoch": 0.3804485059820838,
      "flos": 18954119416320.0,
      "grad_norm": 1.9625682482062767,
      "language_loss": 0.79253131,
      "learning_rate": 2.844334322976806e-06,
      "loss": 0.81569493,
      "num_input_tokens_seen": 68064790,
      "step": 3164,
      "time_per_iteration": 2.680466413497925
    },
    {
      "auxiliary_loss_clip": 0.0139497,
      "auxiliary_loss_mlp": 0.01022351,
      "balance_loss_clip": 1.04862821,
      "balance_loss_mlp": 1.01416183,
      "epoch": 0.3805687488727229,
      "flos": 21833759831040.0,
      "grad_norm": 1.7688213979720362,
      "language_loss": 0.83296978,
      "learning_rate": 2.8436281071309866e-06,
      "loss": 0.85714293,
      "num_input_tokens_seen": 68083330,
      "step": 3165,
      "time_per_iteration": 3.620453357696533
    },
    {
      "auxiliary_loss_clip": 0.01305444,
      "auxiliary_loss_mlp": 0.01000433,
      "balance_loss_clip": 1.01402974,
      "balance_loss_mlp": 0.99925917,
      "epoch": 0.380688991763362,
      "flos": 58546209968640.0,
      "grad_norm": 0.7280143145629807,
      "language_loss": 0.52977526,
      "learning_rate": 2.842921763307455e-06,
      "loss": 0.55283403,
      "num_input_tokens_seen": 68146140,
      "step": 3166,
      "time_per_iteration": 3.2630279064178467
    },
    {
      "auxiliary_loss_clip": 0.01289525,
      "auxiliary_loss_mlp": 0.01030154,
      "balance_loss_clip": 1.04938269,
      "balance_loss_mlp": 1.02252471,
      "epoch": 0.38080923465400107,
      "flos": 23799509487360.0,
      "grad_norm": 2.181580046708282,
      "language_loss": 0.8268044,
      "learning_rate": 2.842215291613361e-06,
      "loss": 0.85000122,
      "num_input_tokens_seen": 68164520,
      "step": 3167,
      "time_per_iteration": 2.700155019760132
    },
    {
      "auxiliary_loss_clip": 0.01392354,
      "auxiliary_loss_mlp": 0.01016399,
      "balance_loss_clip": 1.01641619,
      "balance_loss_mlp": 1.01520085,
      "epoch": 0.3809294775446402,
      "flos": 54969866380800.0,
      "grad_norm": 0.7948504083475938,
      "language_loss": 0.59228337,
      "learning_rate": 2.8415086921558774e-06,
      "loss": 0.61637086,
      "num_input_tokens_seen": 68227945,
      "step": 3168,
      "time_per_iteration": 3.562152624130249
    },
    {
      "auxiliary_loss_clip": 0.01282717,
      "auxiliary_loss_mlp": 0.01025213,
      "balance_loss_clip": 1.04609036,
      "balance_loss_mlp": 1.01791692,
      "epoch": 0.38104972043527924,
      "flos": 24643697904000.0,
      "grad_norm": 1.5653632446888952,
      "language_loss": 0.78756958,
      "learning_rate": 2.840801965042194e-06,
      "loss": 0.81064892,
      "num_input_tokens_seen": 68247405,
      "step": 3169,
      "time_per_iteration": 2.8943099975585938
    },
    {
      "auxiliary_loss_clip": 0.01286161,
      "auxiliary_loss_mlp": 0.01026806,
      "balance_loss_clip": 1.04767799,
      "balance_loss_mlp": 1.0193491,
      "epoch": 0.38116996332591835,
      "flos": 22856783086080.0,
      "grad_norm": 1.8089321043139683,
      "language_loss": 0.83736598,
      "learning_rate": 2.840095110379521e-06,
      "loss": 0.86049563,
      "num_input_tokens_seen": 68266925,
      "step": 3170,
      "time_per_iteration": 2.7105891704559326
    },
    {
      "auxiliary_loss_clip": 0.01282124,
      "auxiliary_loss_mlp": 0.01001914,
      "balance_loss_clip": 1.01257181,
      "balance_loss_mlp": 1.00088882,
      "epoch": 0.38129020621655746,
      "flos": 60836160804480.0,
      "grad_norm": 0.7369701736921025,
      "language_loss": 0.53813159,
      "learning_rate": 2.8393881282750884e-06,
      "loss": 0.56097198,
      "num_input_tokens_seen": 68329755,
      "step": 3171,
      "time_per_iteration": 3.202564239501953
    },
    {
      "auxiliary_loss_clip": 0.01289374,
      "auxiliary_loss_mlp": 0.01027571,
      "balance_loss_clip": 1.05140042,
      "balance_loss_mlp": 1.01919639,
      "epoch": 0.3814104491071965,
      "flos": 21648101408640.0,
      "grad_norm": 2.435873721968084,
      "language_loss": 0.78688079,
      "learning_rate": 2.838681018836144e-06,
      "loss": 0.81005019,
      "num_input_tokens_seen": 68347075,
      "step": 3172,
      "time_per_iteration": 2.70814847946167
    },
    {
      "auxiliary_loss_clip": 0.01342616,
      "auxiliary_loss_mlp": 0.0256389,
      "balance_loss_clip": 1.04894412,
      "balance_loss_mlp": 1.000072,
      "epoch": 0.3815306919978356,
      "flos": 19099090707840.0,
      "grad_norm": 2.387154467329958,
      "language_loss": 0.78927106,
      "learning_rate": 2.837973782169955e-06,
      "loss": 0.82833612,
      "num_input_tokens_seen": 68365450,
      "step": 3173,
      "time_per_iteration": 2.700079917907715
    },
    {
      "auxiliary_loss_clip": 0.01080288,
      "auxiliary_loss_mlp": 0.01004168,
      "balance_loss_clip": 1.01343918,
      "balance_loss_mlp": 1.00310135,
      "epoch": 0.38165093488847474,
      "flos": 67067918156160.0,
      "grad_norm": 0.8076850725381319,
      "language_loss": 0.59133506,
      "learning_rate": 2.8372664183838096e-06,
      "loss": 0.61217964,
      "num_input_tokens_seen": 68428470,
      "step": 3174,
      "time_per_iteration": 3.2572011947631836
    },
    {
      "auxiliary_loss_clip": 0.01188643,
      "auxiliary_loss_mlp": 0.01031497,
      "balance_loss_clip": 1.053684,
      "balance_loss_mlp": 1.02419591,
      "epoch": 0.3817711777791138,
      "flos": 22341105480960.0,
      "grad_norm": 2.3499340145079692,
      "language_loss": 0.69040054,
      "learning_rate": 2.836558927585015e-06,
      "loss": 0.7126019,
      "num_input_tokens_seen": 68445440,
      "step": 3175,
      "time_per_iteration": 2.6211469173431396
    },
    {
      "auxiliary_loss_clip": 0.01243052,
      "auxiliary_loss_mlp": 0.0103268,
      "balance_loss_clip": 1.05240822,
      "balance_loss_mlp": 1.02500236,
      "epoch": 0.3818914206697529,
      "flos": 22820621068800.0,
      "grad_norm": 2.017119681113495,
      "language_loss": 0.82648933,
      "learning_rate": 2.8358513098808957e-06,
      "loss": 0.84924656,
      "num_input_tokens_seen": 68465755,
      "step": 3176,
      "time_per_iteration": 2.6547610759735107
    },
    {
      "auxiliary_loss_clip": 0.01375568,
      "auxiliary_loss_mlp": 0.01036461,
      "balance_loss_clip": 1.045856,
      "balance_loss_mlp": 1.02876043,
      "epoch": 0.382011663560392,
      "flos": 24386074583040.0,
      "grad_norm": 2.0056659887467285,
      "language_loss": 0.77119225,
      "learning_rate": 2.835143565378798e-06,
      "loss": 0.79531252,
      "num_input_tokens_seen": 68486220,
      "step": 3177,
      "time_per_iteration": 2.7676126956939697
    },
    {
      "auxiliary_loss_clip": 0.0142722,
      "auxiliary_loss_mlp": 0.01025163,
      "balance_loss_clip": 1.04347134,
      "balance_loss_mlp": 1.01691389,
      "epoch": 0.38213190645103107,
      "flos": 21981568296960.0,
      "grad_norm": 3.2109017802435416,
      "language_loss": 0.7795769,
      "learning_rate": 2.8344356941860847e-06,
      "loss": 0.80410075,
      "num_input_tokens_seen": 68505850,
      "step": 3178,
      "time_per_iteration": 2.815690517425537
    },
    {
      "auxiliary_loss_clip": 0.01339383,
      "auxiliary_loss_mlp": 0.01026191,
      "balance_loss_clip": 1.04915679,
      "balance_loss_mlp": 1.0187968,
      "epoch": 0.3822521493416702,
      "flos": 35516945773440.0,
      "grad_norm": 2.3163389089422575,
      "language_loss": 0.66367561,
      "learning_rate": 2.8337276964101403e-06,
      "loss": 0.68733138,
      "num_input_tokens_seen": 68526290,
      "step": 3179,
      "time_per_iteration": 4.167910575866699
    },
    {
      "auxiliary_loss_clip": 0.01243175,
      "auxiliary_loss_mlp": 0.01024548,
      "balance_loss_clip": 1.05286872,
      "balance_loss_mlp": 1.01700199,
      "epoch": 0.3823723922323093,
      "flos": 21069904181760.0,
      "grad_norm": 2.218069588101566,
      "language_loss": 0.76351023,
      "learning_rate": 2.833019572158367e-06,
      "loss": 0.78618753,
      "num_input_tokens_seen": 68544725,
      "step": 3180,
      "time_per_iteration": 2.72501540184021
    },
    {
      "auxiliary_loss_clip": 0.012872,
      "auxiliary_loss_mlp": 0.0102965,
      "balance_loss_clip": 1.05295956,
      "balance_loss_mlp": 1.02226543,
      "epoch": 0.38249263512294834,
      "flos": 19789149864960.0,
      "grad_norm": 1.8676258820512213,
      "language_loss": 0.80120307,
      "learning_rate": 2.8323113215381872e-06,
      "loss": 0.82437158,
      "num_input_tokens_seen": 68563070,
      "step": 3181,
      "time_per_iteration": 2.6754629611968994
    },
    {
      "auxiliary_loss_clip": 0.01336588,
      "auxiliary_loss_mlp": 0.0102881,
      "balance_loss_clip": 1.04925621,
      "balance_loss_mlp": 1.02079916,
      "epoch": 0.38261287801358745,
      "flos": 21433930565760.0,
      "grad_norm": 2.1046325976184925,
      "language_loss": 0.76229143,
      "learning_rate": 2.831602944657042e-06,
      "loss": 0.78594542,
      "num_input_tokens_seen": 68581150,
      "step": 3182,
      "time_per_iteration": 2.750411033630371
    },
    {
      "auxiliary_loss_clip": 0.01199357,
      "auxiliary_loss_mlp": 0.01025543,
      "balance_loss_clip": 1.04874194,
      "balance_loss_mlp": 1.01834893,
      "epoch": 0.38273312090422656,
      "flos": 21981568296960.0,
      "grad_norm": 2.444612505503567,
      "language_loss": 0.74689561,
      "learning_rate": 2.830894441622391e-06,
      "loss": 0.76914465,
      "num_input_tokens_seen": 68597800,
      "step": 3183,
      "time_per_iteration": 2.670667886734009
    },
    {
      "auxiliary_loss_clip": 0.01342723,
      "auxiliary_loss_mlp": 0.02566361,
      "balance_loss_clip": 1.04719985,
      "balance_loss_mlp": 1.00003874,
      "epoch": 0.3828533637948656,
      "flos": 24790895838720.0,
      "grad_norm": 2.1048293625481858,
      "language_loss": 0.79946673,
      "learning_rate": 2.8301858125417134e-06,
      "loss": 0.83855754,
      "num_input_tokens_seen": 68617640,
      "step": 3184,
      "time_per_iteration": 3.7010602951049805
    },
    {
      "auxiliary_loss_clip": 0.01296825,
      "auxiliary_loss_mlp": 0.01030239,
      "balance_loss_clip": 1.05452895,
      "balance_loss_mlp": 1.02311897,
      "epoch": 0.38297360668550473,
      "flos": 22455445449600.0,
      "grad_norm": 1.8896965525497378,
      "language_loss": 0.73990285,
      "learning_rate": 2.8294770575225082e-06,
      "loss": 0.76317346,
      "num_input_tokens_seen": 68637770,
      "step": 3185,
      "time_per_iteration": 2.729907989501953
    },
    {
      "auxiliary_loss_clip": 0.01240217,
      "auxiliary_loss_mlp": 0.01026129,
      "balance_loss_clip": 1.05448651,
      "balance_loss_mlp": 1.01877367,
      "epoch": 0.3830938495761438,
      "flos": 24896903852160.0,
      "grad_norm": 1.9077463357339228,
      "language_loss": 0.84123486,
      "learning_rate": 2.828768176672293e-06,
      "loss": 0.86389834,
      "num_input_tokens_seen": 68656885,
      "step": 3186,
      "time_per_iteration": 2.658665657043457
    },
    {
      "auxiliary_loss_clip": 0.01340684,
      "auxiliary_loss_mlp": 0.01025426,
      "balance_loss_clip": 1.04673958,
      "balance_loss_mlp": 1.01709306,
      "epoch": 0.3832140924667829,
      "flos": 33036236784000.0,
      "grad_norm": 2.3699117653438653,
      "language_loss": 0.71721745,
      "learning_rate": 2.8280591700986044e-06,
      "loss": 0.74087858,
      "num_input_tokens_seen": 68678750,
      "step": 3187,
      "time_per_iteration": 2.838984966278076
    },
    {
      "auxiliary_loss_clip": 0.01294083,
      "auxiliary_loss_mlp": 0.01022482,
      "balance_loss_clip": 1.05011737,
      "balance_loss_mlp": 1.01508546,
      "epoch": 0.383334335357422,
      "flos": 31903721896320.0,
      "grad_norm": 2.007625815735119,
      "language_loss": 0.75242102,
      "learning_rate": 2.827350037908999e-06,
      "loss": 0.77558672,
      "num_input_tokens_seen": 68698190,
      "step": 3188,
      "time_per_iteration": 2.7596468925476074
    },
    {
      "auxiliary_loss_clip": 0.01345365,
      "auxiliary_loss_mlp": 0.01026995,
      "balance_loss_clip": 1.04950881,
      "balance_loss_mlp": 1.01879978,
      "epoch": 0.38345457824806106,
      "flos": 19791915212160.0,
      "grad_norm": 2.2665438358439736,
      "language_loss": 0.79246759,
      "learning_rate": 2.8266407802110496e-06,
      "loss": 0.81619114,
      "num_input_tokens_seen": 68716445,
      "step": 3189,
      "time_per_iteration": 3.555495262145996
    },
    {
      "auxiliary_loss_clip": 0.01493872,
      "auxiliary_loss_mlp": 0.0102888,
      "balance_loss_clip": 1.04388404,
      "balance_loss_mlp": 1.02104747,
      "epoch": 0.3835748211387002,
      "flos": 22419391173120.0,
      "grad_norm": 2.087356685251463,
      "language_loss": 0.76125646,
      "learning_rate": 2.8259313971123515e-06,
      "loss": 0.78648394,
      "num_input_tokens_seen": 68737565,
      "step": 3190,
      "time_per_iteration": 2.9187867641448975
    },
    {
      "auxiliary_loss_clip": 0.01236427,
      "auxiliary_loss_mlp": 0.01027855,
      "balance_loss_clip": 1.05364084,
      "balance_loss_mlp": 1.02070856,
      "epoch": 0.3836950640293393,
      "flos": 25118436983040.0,
      "grad_norm": 1.5921029842738508,
      "language_loss": 0.78417611,
      "learning_rate": 2.8252218887205166e-06,
      "loss": 0.8068189,
      "num_input_tokens_seen": 68758255,
      "step": 3191,
      "time_per_iteration": 3.6375017166137695
    },
    {
      "auxiliary_loss_clip": 0.01439828,
      "auxiliary_loss_mlp": 0.01028221,
      "balance_loss_clip": 1.04692531,
      "balance_loss_mlp": 1.02045441,
      "epoch": 0.38381530691997834,
      "flos": 21799213925760.0,
      "grad_norm": 1.7134412647851787,
      "language_loss": 0.80389321,
      "learning_rate": 2.824512255143178e-06,
      "loss": 0.82857364,
      "num_input_tokens_seen": 68777490,
      "step": 3192,
      "time_per_iteration": 2.841970205307007
    },
    {
      "auxiliary_loss_clip": 0.01343962,
      "auxiliary_loss_mlp": 0.01030918,
      "balance_loss_clip": 1.0502336,
      "balance_loss_mlp": 1.02369988,
      "epoch": 0.38393554981061745,
      "flos": 21252689516160.0,
      "grad_norm": 1.7669549508463924,
      "language_loss": 0.79345584,
      "learning_rate": 2.8238024964879855e-06,
      "loss": 0.81720459,
      "num_input_tokens_seen": 68798385,
      "step": 3193,
      "time_per_iteration": 2.725078821182251
    },
    {
      "auxiliary_loss_clip": 0.01196757,
      "auxiliary_loss_mlp": 0.0103506,
      "balance_loss_clip": 1.05695057,
      "balance_loss_mlp": 1.02680147,
      "epoch": 0.38405579270125656,
      "flos": 17019360218880.0,
      "grad_norm": 2.2659316431188934,
      "language_loss": 0.77323675,
      "learning_rate": 2.8230926128626095e-06,
      "loss": 0.795555,
      "num_input_tokens_seen": 68816880,
      "step": 3194,
      "time_per_iteration": 2.6614410877227783
    },
    {
      "auxiliary_loss_clip": 0.01284694,
      "auxiliary_loss_mlp": 0.01031034,
      "balance_loss_clip": 1.04812205,
      "balance_loss_mlp": 1.02357101,
      "epoch": 0.3841760355918956,
      "flos": 21835375943040.0,
      "grad_norm": 1.885627670617992,
      "language_loss": 0.79964149,
      "learning_rate": 2.822382604374738e-06,
      "loss": 0.82279867,
      "num_input_tokens_seen": 68835805,
      "step": 3195,
      "time_per_iteration": 2.7607107162475586
    },
    {
      "auxiliary_loss_clip": 0.01290601,
      "auxiliary_loss_mlp": 0.01031694,
      "balance_loss_clip": 1.05413294,
      "balance_loss_mlp": 1.02466619,
      "epoch": 0.3842962784825347,
      "flos": 25915114684800.0,
      "grad_norm": 2.0179969391048136,
      "language_loss": 0.66190314,
      "learning_rate": 2.8216724711320793e-06,
      "loss": 0.68512613,
      "num_input_tokens_seen": 68854930,
      "step": 3196,
      "time_per_iteration": 2.758011817932129
    },
    {
      "auxiliary_loss_clip": 0.01187945,
      "auxiliary_loss_mlp": 0.02560195,
      "balance_loss_clip": 1.05247188,
      "balance_loss_mlp": 1.0001148,
      "epoch": 0.38441652137317384,
      "flos": 25337492075520.0,
      "grad_norm": 1.5366082694822143,
      "language_loss": 0.79592776,
      "learning_rate": 2.820962213242361e-06,
      "loss": 0.83340919,
      "num_input_tokens_seen": 68874260,
      "step": 3197,
      "time_per_iteration": 2.697155714035034
    },
    {
      "auxiliary_loss_clip": 0.01246076,
      "auxiliary_loss_mlp": 0.01032285,
      "balance_loss_clip": 1.0582273,
      "balance_loss_mlp": 1.02448273,
      "epoch": 0.3845367642638129,
      "flos": 18113486446080.0,
      "grad_norm": 2.2253312947397816,
      "language_loss": 0.83891201,
      "learning_rate": 2.8202518308133264e-06,
      "loss": 0.86169565,
      "num_input_tokens_seen": 68891535,
      "step": 3198,
      "time_per_iteration": 2.6350111961364746
    },
    {
      "auxiliary_loss_clip": 0.01194055,
      "auxiliary_loss_mlp": 0.01028879,
      "balance_loss_clip": 1.05469728,
      "balance_loss_mlp": 1.02119029,
      "epoch": 0.384657007154452,
      "flos": 25228395492480.0,
      "grad_norm": 2.0348568967475336,
      "language_loss": 0.7346198,
      "learning_rate": 2.8195413239527426e-06,
      "loss": 0.75684911,
      "num_input_tokens_seen": 68911275,
      "step": 3199,
      "time_per_iteration": 2.6504735946655273
    },
    {
      "auxiliary_loss_clip": 0.01238759,
      "auxiliary_loss_mlp": 0.0103213,
      "balance_loss_clip": 1.05097151,
      "balance_loss_mlp": 1.02464974,
      "epoch": 0.38477725004509106,
      "flos": 19865855358720.0,
      "grad_norm": 1.9502649758273667,
      "language_loss": 0.80824465,
      "learning_rate": 2.8188306927683906e-06,
      "loss": 0.83095354,
      "num_input_tokens_seen": 68930745,
      "step": 3200,
      "time_per_iteration": 2.681459665298462
    },
    {
      "auxiliary_loss_clip": 0.01292597,
      "auxiliary_loss_mlp": 0.01027288,
      "balance_loss_clip": 1.05309165,
      "balance_loss_mlp": 1.02041233,
      "epoch": 0.38489749293573017,
      "flos": 18259391491200.0,
      "grad_norm": 2.333515940643704,
      "language_loss": 0.74375331,
      "learning_rate": 2.818119937368074e-06,
      "loss": 0.76695216,
      "num_input_tokens_seen": 68949380,
      "step": 3201,
      "time_per_iteration": 2.7107772827148438
    },
    {
      "auxiliary_loss_clip": 0.01246799,
      "auxiliary_loss_mlp": 0.01028707,
      "balance_loss_clip": 1.05289292,
      "balance_loss_mlp": 1.02066064,
      "epoch": 0.3850177358263693,
      "flos": 24389163152640.0,
      "grad_norm": 3.35376078311206,
      "language_loss": 0.65412688,
      "learning_rate": 2.817409057859613e-06,
      "loss": 0.67688197,
      "num_input_tokens_seen": 68968370,
      "step": 3202,
      "time_per_iteration": 2.6618170738220215
    },
    {
      "auxiliary_loss_clip": 0.01381393,
      "auxiliary_loss_mlp": 0.01025443,
      "balance_loss_clip": 1.04793668,
      "balance_loss_mlp": 1.01803458,
      "epoch": 0.38513797871700833,
      "flos": 17671533505920.0,
      "grad_norm": 2.8123818358850743,
      "language_loss": 0.79368377,
      "learning_rate": 2.8166980543508482e-06,
      "loss": 0.81775212,
      "num_input_tokens_seen": 68984260,
      "step": 3203,
      "time_per_iteration": 2.7937402725219727
    },
    {
      "auxiliary_loss_clip": 0.0119563,
      "auxiliary_loss_mlp": 0.01025068,
      "balance_loss_clip": 1.05710769,
      "balance_loss_mlp": 1.01743841,
      "epoch": 0.38525822160764744,
      "flos": 25739583897600.0,
      "grad_norm": 2.6610465774360588,
      "language_loss": 0.79906356,
      "learning_rate": 2.815986926949638e-06,
      "loss": 0.82127047,
      "num_input_tokens_seen": 69002760,
      "step": 3204,
      "time_per_iteration": 2.6518466472625732
    },
    {
      "auxiliary_loss_clip": 0.01235144,
      "auxiliary_loss_mlp": 0.01026684,
      "balance_loss_clip": 1.05290103,
      "balance_loss_mlp": 1.01945055,
      "epoch": 0.38537846449828655,
      "flos": 20193647898240.0,
      "grad_norm": 1.9014123710653381,
      "language_loss": 0.80580133,
      "learning_rate": 2.8152756757638597e-06,
      "loss": 0.82841963,
      "num_input_tokens_seen": 69021260,
      "step": 3205,
      "time_per_iteration": 3.7779510021209717
    },
    {
      "auxiliary_loss_clip": 0.01238323,
      "auxiliary_loss_mlp": 0.01027797,
      "balance_loss_clip": 1.05438137,
      "balance_loss_mlp": 1.02071261,
      "epoch": 0.3854987073889256,
      "flos": 23039352938880.0,
      "grad_norm": 2.2425325722910605,
      "language_loss": 0.84511995,
      "learning_rate": 2.8145643009014093e-06,
      "loss": 0.8677811,
      "num_input_tokens_seen": 69039755,
      "step": 3206,
      "time_per_iteration": 2.7388932704925537
    },
    {
      "auxiliary_loss_clip": 0.01241187,
      "auxiliary_loss_mlp": 0.01025975,
      "balance_loss_clip": 1.05197215,
      "balance_loss_mlp": 1.01931143,
      "epoch": 0.3856189502795647,
      "flos": 20190631155840.0,
      "grad_norm": 1.9277421805842203,
      "language_loss": 0.78926289,
      "learning_rate": 2.813852802470202e-06,
      "loss": 0.81193453,
      "num_input_tokens_seen": 69057650,
      "step": 3207,
      "time_per_iteration": 2.6069629192352295
    },
    {
      "auxiliary_loss_clip": 0.01288228,
      "auxiliary_loss_mlp": 0.01028695,
      "balance_loss_clip": 1.05199194,
      "balance_loss_mlp": 1.02111292,
      "epoch": 0.38573919317020383,
      "flos": 25702631781120.0,
      "grad_norm": 2.368439226755666,
      "language_loss": 0.72794861,
      "learning_rate": 2.8131411805781717e-06,
      "loss": 0.75111783,
      "num_input_tokens_seen": 69077775,
      "step": 3208,
      "time_per_iteration": 2.7361230850219727
    },
    {
      "auxiliary_loss_clip": 0.01292226,
      "auxiliary_loss_mlp": 0.01028807,
      "balance_loss_clip": 1.05504155,
      "balance_loss_mlp": 1.02155352,
      "epoch": 0.3858594360608429,
      "flos": 29821405628160.0,
      "grad_norm": 3.066041251493373,
      "language_loss": 0.6449573,
      "learning_rate": 2.8124294353332707e-06,
      "loss": 0.66816759,
      "num_input_tokens_seen": 69096450,
      "step": 3209,
      "time_per_iteration": 2.7139575481414795
    },
    {
      "auxiliary_loss_clip": 0.0134421,
      "auxiliary_loss_mlp": 0.01023845,
      "balance_loss_clip": 1.05145121,
      "balance_loss_mlp": 1.01647162,
      "epoch": 0.385979678951482,
      "flos": 24790428961920.0,
      "grad_norm": 2.0186482145999753,
      "language_loss": 0.77494013,
      "learning_rate": 2.8117175668434713e-06,
      "loss": 0.7986207,
      "num_input_tokens_seen": 69116110,
      "step": 3210,
      "time_per_iteration": 3.651909351348877
    },
    {
      "auxiliary_loss_clip": 0.01194626,
      "auxiliary_loss_mlp": 0.01026257,
      "balance_loss_clip": 1.05617881,
      "balance_loss_mlp": 1.0187645,
      "epoch": 0.3860999218421211,
      "flos": 21287881866240.0,
      "grad_norm": 2.661388110076216,
      "language_loss": 0.70292091,
      "learning_rate": 2.811005575216762e-06,
      "loss": 0.72512978,
      "num_input_tokens_seen": 69134825,
      "step": 3211,
      "time_per_iteration": 2.639404535293579
    },
    {
      "auxiliary_loss_clip": 0.01343484,
      "auxiliary_loss_mlp": 0.01029298,
      "balance_loss_clip": 1.05228531,
      "balance_loss_mlp": 1.02223229,
      "epoch": 0.38622016473276016,
      "flos": 24536720223360.0,
      "grad_norm": 1.514788989628463,
      "language_loss": 0.79061633,
      "learning_rate": 2.8102934605611513e-06,
      "loss": 0.81434417,
      "num_input_tokens_seen": 69156460,
      "step": 3212,
      "time_per_iteration": 2.7515461444854736
    },
    {
      "auxiliary_loss_clip": 0.01295561,
      "auxiliary_loss_mlp": 0.0102938,
      "balance_loss_clip": 1.05495358,
      "balance_loss_mlp": 1.02256763,
      "epoch": 0.3863404076233993,
      "flos": 20558212986240.0,
      "grad_norm": 2.382460382579082,
      "language_loss": 0.67431402,
      "learning_rate": 2.8095812229846665e-06,
      "loss": 0.69756347,
      "num_input_tokens_seen": 69176420,
      "step": 3213,
      "time_per_iteration": 2.7034125328063965
    },
    {
      "auxiliary_loss_clip": 0.01291638,
      "auxiliary_loss_mlp": 0.01027512,
      "balance_loss_clip": 1.05217886,
      "balance_loss_mlp": 1.01962638,
      "epoch": 0.3864606505140384,
      "flos": 22346277039360.0,
      "grad_norm": 2.8929859322960447,
      "language_loss": 0.691365,
      "learning_rate": 2.808868862595355e-06,
      "loss": 0.71455657,
      "num_input_tokens_seen": 69196665,
      "step": 3214,
      "time_per_iteration": 2.7086334228515625
    },
    {
      "auxiliary_loss_clip": 0.01244961,
      "auxiliary_loss_mlp": 0.01029703,
      "balance_loss_clip": 1.05416954,
      "balance_loss_mlp": 1.02228808,
      "epoch": 0.38658089340467744,
      "flos": 25703601448320.0,
      "grad_norm": 2.0045113991927286,
      "language_loss": 0.79764915,
      "learning_rate": 2.8081563795012795e-06,
      "loss": 0.82039571,
      "num_input_tokens_seen": 69216290,
      "step": 3215,
      "time_per_iteration": 3.599130153656006
    },
    {
      "auxiliary_loss_clip": 0.01299274,
      "auxiliary_loss_mlp": 0.01024729,
      "balance_loss_clip": 1.05186558,
      "balance_loss_mlp": 1.01747477,
      "epoch": 0.38670113629531655,
      "flos": 33802534558080.0,
      "grad_norm": 1.9700562060465656,
      "language_loss": 0.73885095,
      "learning_rate": 2.807443773810524e-06,
      "loss": 0.76209098,
      "num_input_tokens_seen": 69237550,
      "step": 3216,
      "time_per_iteration": 3.763050079345703
    },
    {
      "auxiliary_loss_clip": 0.01342339,
      "auxiliary_loss_mlp": 0.01030195,
      "balance_loss_clip": 1.05544972,
      "balance_loss_mlp": 1.0224551,
      "epoch": 0.3868213791859556,
      "flos": 23331522165120.0,
      "grad_norm": 1.890494527481991,
      "language_loss": 0.89468515,
      "learning_rate": 2.80673104563119e-06,
      "loss": 0.91841054,
      "num_input_tokens_seen": 69258175,
      "step": 3217,
      "time_per_iteration": 2.7416720390319824
    },
    {
      "auxiliary_loss_clip": 0.01236831,
      "auxiliary_loss_mlp": 0.01028427,
      "balance_loss_clip": 1.05370641,
      "balance_loss_mlp": 1.02168012,
      "epoch": 0.3869416220765947,
      "flos": 18441530380800.0,
      "grad_norm": 2.219466106222124,
      "language_loss": 0.79071593,
      "learning_rate": 2.8060181950713976e-06,
      "loss": 0.81336856,
      "num_input_tokens_seen": 69274965,
      "step": 3218,
      "time_per_iteration": 2.666572332382202
    },
    {
      "auxiliary_loss_clip": 0.01343013,
      "auxiliary_loss_mlp": 0.01031907,
      "balance_loss_clip": 1.0496999,
      "balance_loss_mlp": 1.02387881,
      "epoch": 0.3870618649672338,
      "flos": 15632992938240.0,
      "grad_norm": 1.959692364203572,
      "language_loss": 0.80981839,
      "learning_rate": 2.805305222239286e-06,
      "loss": 0.83356762,
      "num_input_tokens_seen": 69292220,
      "step": 3219,
      "time_per_iteration": 2.6673171520233154
    },
    {
      "auxiliary_loss_clip": 0.01288732,
      "auxiliary_loss_mlp": 0.01036498,
      "balance_loss_clip": 1.05323219,
      "balance_loss_mlp": 1.02911329,
      "epoch": 0.3871821078578729,
      "flos": 23513804709120.0,
      "grad_norm": 1.7857305138758812,
      "language_loss": 0.73715317,
      "learning_rate": 2.8045921272430118e-06,
      "loss": 0.76040554,
      "num_input_tokens_seen": 69311900,
      "step": 3220,
      "time_per_iteration": 2.724794387817383
    },
    {
      "auxiliary_loss_clip": 0.0124987,
      "auxiliary_loss_mlp": 0.01028425,
      "balance_loss_clip": 1.05542636,
      "balance_loss_mlp": 1.02035427,
      "epoch": 0.387302350748512,
      "flos": 17778259791360.0,
      "grad_norm": 2.258988464885024,
      "language_loss": 0.76675713,
      "learning_rate": 2.803878910190753e-06,
      "loss": 0.78954005,
      "num_input_tokens_seen": 69328820,
      "step": 3221,
      "time_per_iteration": 2.6279170513153076
    },
    {
      "auxiliary_loss_clip": 0.01245741,
      "auxiliary_loss_mlp": 0.01030459,
      "balance_loss_clip": 1.05298054,
      "balance_loss_mlp": 1.02315712,
      "epoch": 0.3874225936391511,
      "flos": 11503409097600.0,
      "grad_norm": 3.4873468683967017,
      "language_loss": 0.82270277,
      "learning_rate": 2.8031655711907017e-06,
      "loss": 0.84546477,
      "num_input_tokens_seen": 69342525,
      "step": 3222,
      "time_per_iteration": 2.6165103912353516
    },
    {
      "auxiliary_loss_clip": 0.01241672,
      "auxiliary_loss_mlp": 0.01027482,
      "balance_loss_clip": 1.0554235,
      "balance_loss_mlp": 1.02042747,
      "epoch": 0.38754283652979016,
      "flos": 21945154884480.0,
      "grad_norm": 2.0727043793815376,
      "language_loss": 0.80855596,
      "learning_rate": 2.8024521103510723e-06,
      "loss": 0.83124751,
      "num_input_tokens_seen": 69359295,
      "step": 3223,
      "time_per_iteration": 2.614391565322876
    },
    {
      "auxiliary_loss_clip": 0.01240246,
      "auxiliary_loss_mlp": 0.0102612,
      "balance_loss_clip": 1.05012906,
      "balance_loss_mlp": 1.01841283,
      "epoch": 0.38766307942042927,
      "flos": 21175984022400.0,
      "grad_norm": 1.7969597932684584,
      "language_loss": 0.75279438,
      "learning_rate": 2.8017385277800952e-06,
      "loss": 0.7754581,
      "num_input_tokens_seen": 69377650,
      "step": 3224,
      "time_per_iteration": 2.6493000984191895
    },
    {
      "auxiliary_loss_clip": 0.01344113,
      "auxiliary_loss_mlp": 0.01029551,
      "balance_loss_clip": 1.05318201,
      "balance_loss_mlp": 1.02177238,
      "epoch": 0.3877833223110684,
      "flos": 27417294391680.0,
      "grad_norm": 1.8299109062145495,
      "language_loss": 0.74913079,
      "learning_rate": 2.8010248235860213e-06,
      "loss": 0.77286744,
      "num_input_tokens_seen": 69397765,
      "step": 3225,
      "time_per_iteration": 2.7388575077056885
    },
    {
      "auxiliary_loss_clip": 0.01195545,
      "auxiliary_loss_mlp": 0.02508441,
      "balance_loss_clip": 1.01666796,
      "balance_loss_mlp": 1.00019217,
      "epoch": 0.38790356520170743,
      "flos": 64500019879680.0,
      "grad_norm": 0.8262551152289839,
      "language_loss": 0.62800068,
      "learning_rate": 2.8003109978771192e-06,
      "loss": 0.66504049,
      "num_input_tokens_seen": 69458930,
      "step": 3226,
      "time_per_iteration": 3.363452911376953
    },
    {
      "auxiliary_loss_clip": 0.01332464,
      "auxiliary_loss_mlp": 0.01023534,
      "balance_loss_clip": 1.04586148,
      "balance_loss_mlp": 1.01580334,
      "epoch": 0.38802380809234654,
      "flos": 22345415112960.0,
      "grad_norm": 2.3450719125577115,
      "language_loss": 0.7876904,
      "learning_rate": 2.799597050761674e-06,
      "loss": 0.81125039,
      "num_input_tokens_seen": 69475135,
      "step": 3227,
      "time_per_iteration": 2.770724058151245
    },
    {
      "auxiliary_loss_clip": 0.01195355,
      "auxiliary_loss_mlp": 0.01024126,
      "balance_loss_clip": 1.05622387,
      "balance_loss_mlp": 1.01680601,
      "epoch": 0.38814405098298566,
      "flos": 25261361199360.0,
      "grad_norm": 2.182797083423832,
      "language_loss": 0.79204786,
      "learning_rate": 2.7988829823479924e-06,
      "loss": 0.81424266,
      "num_input_tokens_seen": 69493525,
      "step": 3228,
      "time_per_iteration": 2.685530185699463
    },
    {
      "auxiliary_loss_clip": 0.0128642,
      "auxiliary_loss_mlp": 0.01024112,
      "balance_loss_clip": 1.05045068,
      "balance_loss_mlp": 1.0166111,
      "epoch": 0.3882642938736247,
      "flos": 18841180078080.0,
      "grad_norm": 2.0392095307584537,
      "language_loss": 0.64158261,
      "learning_rate": 2.7981687927443976e-06,
      "loss": 0.66468793,
      "num_input_tokens_seen": 69510325,
      "step": 3229,
      "time_per_iteration": 2.659245252609253
    },
    {
      "auxiliary_loss_clip": 0.01243529,
      "auxiliary_loss_mlp": 0.01032346,
      "balance_loss_clip": 1.05129993,
      "balance_loss_mlp": 1.02545834,
      "epoch": 0.3883845367642638,
      "flos": 21652806090240.0,
      "grad_norm": 2.6123069843899978,
      "language_loss": 0.85512352,
      "learning_rate": 2.797454482059231e-06,
      "loss": 0.87788236,
      "num_input_tokens_seen": 69530480,
      "step": 3230,
      "time_per_iteration": 2.627046585083008
    },
    {
      "auxiliary_loss_clip": 0.01197288,
      "auxiliary_loss_mlp": 0.01032262,
      "balance_loss_clip": 1.05824518,
      "balance_loss_mlp": 1.02448344,
      "epoch": 0.3885047796549029,
      "flos": 20557530627840.0,
      "grad_norm": 1.6440536467845568,
      "language_loss": 0.84521759,
      "learning_rate": 2.7967400504008537e-06,
      "loss": 0.86751312,
      "num_input_tokens_seen": 69549780,
      "step": 3231,
      "time_per_iteration": 3.52471923828125
    },
    {
      "auxiliary_loss_clip": 0.01295635,
      "auxiliary_loss_mlp": 0.01010229,
      "balance_loss_clip": 1.01599503,
      "balance_loss_mlp": 1.00901306,
      "epoch": 0.388625022545542,
      "flos": 64325491695360.0,
      "grad_norm": 0.7883924369784058,
      "language_loss": 0.57427609,
      "learning_rate": 2.7960254978776456e-06,
      "loss": 0.5973348,
      "num_input_tokens_seen": 69611870,
      "step": 3232,
      "time_per_iteration": 3.3015406131744385
    },
    {
      "auxiliary_loss_clip": 0.01197189,
      "auxiliary_loss_mlp": 0.01037807,
      "balance_loss_clip": 1.05738091,
      "balance_loss_mlp": 1.03017211,
      "epoch": 0.3887452654361811,
      "flos": 18113881495680.0,
      "grad_norm": 6.132917576270898,
      "language_loss": 0.81145716,
      "learning_rate": 2.7953108245980006e-06,
      "loss": 0.83380705,
      "num_input_tokens_seen": 69630385,
      "step": 3233,
      "time_per_iteration": 2.5764636993408203
    },
    {
      "auxiliary_loss_clip": 0.01288793,
      "auxiliary_loss_mlp": 0.0103201,
      "balance_loss_clip": 1.05521965,
      "balance_loss_mlp": 1.0246073,
      "epoch": 0.38886550832682015,
      "flos": 24975261371520.0,
      "grad_norm": 1.820929193586301,
      "language_loss": 0.73662913,
      "learning_rate": 2.7945960306703365e-06,
      "loss": 0.75983715,
      "num_input_tokens_seen": 69653370,
      "step": 3234,
      "time_per_iteration": 2.7453856468200684
    },
    {
      "auxiliary_loss_clip": 0.01244806,
      "auxiliary_loss_mlp": 0.01036278,
      "balance_loss_clip": 1.05298483,
      "balance_loss_mlp": 1.02861249,
      "epoch": 0.38898575121745926,
      "flos": 27199496275200.0,
      "grad_norm": 2.496728839163769,
      "language_loss": 0.65918499,
      "learning_rate": 2.7938811162030865e-06,
      "loss": 0.68199581,
      "num_input_tokens_seen": 69673635,
      "step": 3235,
      "time_per_iteration": 2.6843411922454834
    },
    {
      "auxiliary_loss_clip": 0.01237584,
      "auxiliary_loss_mlp": 0.01030193,
      "balance_loss_clip": 1.05346656,
      "balance_loss_mlp": 1.02338052,
      "epoch": 0.3891059941080984,
      "flos": 28763728727040.0,
      "grad_norm": 2.0002836117780696,
      "language_loss": 0.82325041,
      "learning_rate": 2.793166081304702e-06,
      "loss": 0.84592819,
      "num_input_tokens_seen": 69694130,
      "step": 3236,
      "time_per_iteration": 3.685251235961914
    },
    {
      "auxiliary_loss_clip": 0.0134442,
      "auxiliary_loss_mlp": 0.01025385,
      "balance_loss_clip": 1.04986787,
      "balance_loss_mlp": 1.01713014,
      "epoch": 0.38922623699873743,
      "flos": 22893447893760.0,
      "grad_norm": 1.7917907903666261,
      "language_loss": 0.82167423,
      "learning_rate": 2.7924509260836543e-06,
      "loss": 0.84537232,
      "num_input_tokens_seen": 69713255,
      "step": 3237,
      "time_per_iteration": 2.710801839828491
    },
    {
      "auxiliary_loss_clip": 0.01337431,
      "auxiliary_loss_mlp": 0.01027428,
      "balance_loss_clip": 1.04999185,
      "balance_loss_mlp": 1.02017438,
      "epoch": 0.38934647988937654,
      "flos": 19792418002560.0,
      "grad_norm": 1.6132480515168308,
      "language_loss": 0.68574744,
      "learning_rate": 2.791735650648431e-06,
      "loss": 0.70939606,
      "num_input_tokens_seen": 69732375,
      "step": 3238,
      "time_per_iteration": 2.7006843090057373
    },
    {
      "auxiliary_loss_clip": 0.01287716,
      "auxiliary_loss_mlp": 0.01031253,
      "balance_loss_clip": 1.05230761,
      "balance_loss_mlp": 1.02370143,
      "epoch": 0.38946672278001565,
      "flos": 19202081978880.0,
      "grad_norm": 7.616729826500581,
      "language_loss": 0.74479604,
      "learning_rate": 2.791020255107538e-06,
      "loss": 0.7679857,
      "num_input_tokens_seen": 69749745,
      "step": 3239,
      "time_per_iteration": 2.718636989593506
    },
    {
      "auxiliary_loss_clip": 0.0133828,
      "auxiliary_loss_mlp": 0.01025583,
      "balance_loss_clip": 1.047364,
      "balance_loss_mlp": 1.01800466,
      "epoch": 0.3895869656706547,
      "flos": 24936477661440.0,
      "grad_norm": 1.6151597629561283,
      "language_loss": 0.80834603,
      "learning_rate": 2.7903047395695023e-06,
      "loss": 0.83198464,
      "num_input_tokens_seen": 69769645,
      "step": 3240,
      "time_per_iteration": 2.769615888595581
    },
    {
      "auxiliary_loss_clip": 0.01240215,
      "auxiliary_loss_mlp": 0.02565014,
      "balance_loss_clip": 1.05616164,
      "balance_loss_mlp": 0.9999997,
      "epoch": 0.3897072085612938,
      "flos": 24133622820480.0,
      "grad_norm": 2.3811642799204447,
      "language_loss": 0.90666485,
      "learning_rate": 2.789589104142865e-06,
      "loss": 0.94471717,
      "num_input_tokens_seen": 69787270,
      "step": 3241,
      "time_per_iteration": 4.541808843612671
    },
    {
      "auxiliary_loss_clip": 0.0134909,
      "auxiliary_loss_mlp": 0.01030808,
      "balance_loss_clip": 1.05340004,
      "balance_loss_mlp": 1.02362561,
      "epoch": 0.3898274514519329,
      "flos": 17166342672000.0,
      "grad_norm": 6.37984128761456,
      "language_loss": 0.76654387,
      "learning_rate": 2.7888733489361895e-06,
      "loss": 0.79034281,
      "num_input_tokens_seen": 69805685,
      "step": 3242,
      "time_per_iteration": 2.8276987075805664
    },
    {
      "auxiliary_loss_clip": 0.01081052,
      "auxiliary_loss_mlp": 0.010031,
      "balance_loss_clip": 1.01441097,
      "balance_loss_mlp": 1.00215864,
      "epoch": 0.389947694342572,
      "flos": 66074807952000.0,
      "grad_norm": 0.7270605568843346,
      "language_loss": 0.58661026,
      "learning_rate": 2.788157474058054e-06,
      "loss": 0.6074518,
      "num_input_tokens_seen": 69867960,
      "step": 3243,
      "time_per_iteration": 3.394801378250122
    },
    {
      "auxiliary_loss_clip": 0.01189755,
      "auxiliary_loss_mlp": 0.01020571,
      "balance_loss_clip": 1.05400586,
      "balance_loss_mlp": 1.01364493,
      "epoch": 0.3900679372332111,
      "flos": 25740912700800.0,
      "grad_norm": 1.4868352608196058,
      "language_loss": 0.69914275,
      "learning_rate": 2.7874414796170555e-06,
      "loss": 0.721246,
      "num_input_tokens_seen": 69889450,
      "step": 3244,
      "time_per_iteration": 2.694288730621338
    },
    {
      "auxiliary_loss_clip": 0.01241961,
      "auxiliary_loss_mlp": 0.01025049,
      "balance_loss_clip": 1.05308616,
      "balance_loss_mlp": 1.01715112,
      "epoch": 0.3901881801238502,
      "flos": 11801611808640.0,
      "grad_norm": 3.3135797931948385,
      "language_loss": 0.83994007,
      "learning_rate": 2.7867253657218113e-06,
      "loss": 0.86261022,
      "num_input_tokens_seen": 69903340,
      "step": 3245,
      "time_per_iteration": 2.602369785308838
    },
    {
      "auxiliary_loss_clip": 0.01291273,
      "auxiliary_loss_mlp": 0.02566271,
      "balance_loss_clip": 1.05107164,
      "balance_loss_mlp": 1.00003672,
      "epoch": 0.39030842301448926,
      "flos": 27308951994240.0,
      "grad_norm": 1.8192435983661714,
      "language_loss": 0.73010677,
      "learning_rate": 2.7860091324809544e-06,
      "loss": 0.76868224,
      "num_input_tokens_seen": 69924400,
      "step": 3246,
      "time_per_iteration": 2.7641847133636475
    },
    {
      "auxiliary_loss_clip": 0.01237941,
      "auxiliary_loss_mlp": 0.01022254,
      "balance_loss_clip": 1.05500281,
      "balance_loss_mlp": 1.01494658,
      "epoch": 0.39042866590512837,
      "flos": 27163334257920.0,
      "grad_norm": 2.082126861995118,
      "language_loss": 0.81072557,
      "learning_rate": 2.7852927800031377e-06,
      "loss": 0.83332753,
      "num_input_tokens_seen": 69944565,
      "step": 3247,
      "time_per_iteration": 2.677333116531372
    },
    {
      "auxiliary_loss_clip": 0.01290247,
      "auxiliary_loss_mlp": 0.01027478,
      "balance_loss_clip": 1.05221939,
      "balance_loss_mlp": 1.02022123,
      "epoch": 0.3905489087957674,
      "flos": 29716115886720.0,
      "grad_norm": 1.8970322703627414,
      "language_loss": 0.83323359,
      "learning_rate": 2.7845763083970298e-06,
      "loss": 0.85641092,
      "num_input_tokens_seen": 69964965,
      "step": 3248,
      "time_per_iteration": 2.748837947845459
    },
    {
      "auxiliary_loss_clip": 0.01237732,
      "auxiliary_loss_mlp": 0.01028134,
      "balance_loss_clip": 1.05175149,
      "balance_loss_mlp": 1.02041554,
      "epoch": 0.39066915168640653,
      "flos": 24498618871680.0,
      "grad_norm": 1.9789146678909455,
      "language_loss": 0.82298541,
      "learning_rate": 2.7838597177713205e-06,
      "loss": 0.84564412,
      "num_input_tokens_seen": 69986055,
      "step": 3249,
      "time_per_iteration": 2.6625468730926514
    },
    {
      "auxiliary_loss_clip": 0.01441665,
      "auxiliary_loss_mlp": 0.01030491,
      "balance_loss_clip": 1.05465901,
      "balance_loss_mlp": 1.0231241,
      "epoch": 0.39078939457704565,
      "flos": 20558572122240.0,
      "grad_norm": 18.839958374051903,
      "language_loss": 0.73455989,
      "learning_rate": 2.7831430082347143e-06,
      "loss": 0.7592814,
      "num_input_tokens_seen": 70005260,
      "step": 3250,
      "time_per_iteration": 2.7900984287261963
    },
    {
      "auxiliary_loss_clip": 0.01242125,
      "auxiliary_loss_mlp": 0.0256167,
      "balance_loss_clip": 1.05367684,
      "balance_loss_mlp": 0.99998468,
      "epoch": 0.3909096374676847,
      "flos": 22783417557120.0,
      "grad_norm": 2.3685143466154495,
      "language_loss": 0.82234645,
      "learning_rate": 2.7824261798959373e-06,
      "loss": 0.8603844,
      "num_input_tokens_seen": 70023440,
      "step": 3251,
      "time_per_iteration": 2.7184128761291504
    },
    {
      "auxiliary_loss_clip": 0.01295221,
      "auxiliary_loss_mlp": 0.01034065,
      "balance_loss_clip": 1.04947734,
      "balance_loss_mlp": 1.02669168,
      "epoch": 0.3910298803583238,
      "flos": 23003119094400.0,
      "grad_norm": 1.960404090479542,
      "language_loss": 0.79358715,
      "learning_rate": 2.78170923286373e-06,
      "loss": 0.81687999,
      "num_input_tokens_seen": 70043040,
      "step": 3252,
      "time_per_iteration": 2.672183036804199
    },
    {
      "auxiliary_loss_clip": 0.01493884,
      "auxiliary_loss_mlp": 0.01030794,
      "balance_loss_clip": 1.05182815,
      "balance_loss_mlp": 1.02308106,
      "epoch": 0.3911501232489629,
      "flos": 24316264500480.0,
      "grad_norm": 2.186389661726091,
      "language_loss": 0.84016603,
      "learning_rate": 2.780992167246854e-06,
      "loss": 0.86541271,
      "num_input_tokens_seen": 70060565,
      "step": 3253,
      "time_per_iteration": 2.940718412399292
    },
    {
      "auxiliary_loss_clip": 0.01190805,
      "auxiliary_loss_mlp": 0.01001541,
      "balance_loss_clip": 1.01396894,
      "balance_loss_mlp": 1.00067043,
      "epoch": 0.391270366139602,
      "flos": 60869054684160.0,
      "grad_norm": 0.9709122347991077,
      "language_loss": 0.72097194,
      "learning_rate": 2.7802749831540883e-06,
      "loss": 0.74289536,
      "num_input_tokens_seen": 70119465,
      "step": 3254,
      "time_per_iteration": 3.3678650856018066
    },
    {
      "auxiliary_loss_clip": 0.01300972,
      "auxiliary_loss_mlp": 0.01029409,
      "balance_loss_clip": 1.05169451,
      "balance_loss_mlp": 1.02283418,
      "epoch": 0.3913906090302411,
      "flos": 21543494025600.0,
      "grad_norm": 2.0420650401118325,
      "language_loss": 0.82135695,
      "learning_rate": 2.7795576806942268e-06,
      "loss": 0.84466076,
      "num_input_tokens_seen": 70138270,
      "step": 3255,
      "time_per_iteration": 2.7661383152008057
    },
    {
      "auxiliary_loss_clip": 0.01204503,
      "auxiliary_loss_mlp": 0.01008693,
      "balance_loss_clip": 1.0334785,
      "balance_loss_mlp": 1.00750065,
      "epoch": 0.3915108519208802,
      "flos": 49839953702400.0,
      "grad_norm": 0.764785976622441,
      "language_loss": 0.54875255,
      "learning_rate": 2.778840259976085e-06,
      "loss": 0.57088447,
      "num_input_tokens_seen": 70193500,
      "step": 3256,
      "time_per_iteration": 4.109271764755249
    },
    {
      "auxiliary_loss_clip": 0.01242206,
      "auxiliary_loss_mlp": 0.01035236,
      "balance_loss_clip": 1.054268,
      "balance_loss_mlp": 1.0272969,
      "epoch": 0.39163109481151925,
      "flos": 16506447960960.0,
      "grad_norm": 3.6955667299468873,
      "language_loss": 0.76653987,
      "learning_rate": 2.778122721108495e-06,
      "loss": 0.78931427,
      "num_input_tokens_seen": 70211730,
      "step": 3257,
      "time_per_iteration": 2.645414113998413
    },
    {
      "auxiliary_loss_clip": 0.0124141,
      "auxiliary_loss_mlp": 0.01026124,
      "balance_loss_clip": 1.05464017,
      "balance_loss_mlp": 1.01838756,
      "epoch": 0.39175133770215836,
      "flos": 26067484177920.0,
      "grad_norm": 2.155989091014263,
      "language_loss": 0.88188154,
      "learning_rate": 2.7774050642003076e-06,
      "loss": 0.90455687,
      "num_input_tokens_seen": 70232540,
      "step": 3258,
      "time_per_iteration": 2.7484490871429443
    },
    {
      "auxiliary_loss_clip": 0.01196007,
      "auxiliary_loss_mlp": 0.01034211,
      "balance_loss_clip": 1.05627954,
      "balance_loss_mlp": 1.02617073,
      "epoch": 0.3918715805927975,
      "flos": 21872076664320.0,
      "grad_norm": 1.9706893219130057,
      "language_loss": 0.93434799,
      "learning_rate": 2.7766872893603896e-06,
      "loss": 0.95665014,
      "num_input_tokens_seen": 70252515,
      "step": 3259,
      "time_per_iteration": 2.6565232276916504
    },
    {
      "auxiliary_loss_clip": 0.01243895,
      "auxiliary_loss_mlp": 0.01030399,
      "balance_loss_clip": 1.05423534,
      "balance_loss_mlp": 1.02330053,
      "epoch": 0.39199182348343653,
      "flos": 20376181837440.0,
      "grad_norm": 1.7464051114237589,
      "language_loss": 0.72926539,
      "learning_rate": 2.7759693966976275e-06,
      "loss": 0.75200832,
      "num_input_tokens_seen": 70271020,
      "step": 3260,
      "time_per_iteration": 2.6310641765594482
    },
    {
      "auxiliary_loss_clip": 0.01337985,
      "auxiliary_loss_mlp": 0.01036766,
      "balance_loss_clip": 1.04877615,
      "balance_loss_mlp": 1.02863026,
      "epoch": 0.39211206637407564,
      "flos": 21683545153920.0,
      "grad_norm": 2.047792780952871,
      "language_loss": 0.85259795,
      "learning_rate": 2.7752513863209242e-06,
      "loss": 0.8763454,
      "num_input_tokens_seen": 70289600,
      "step": 3261,
      "time_per_iteration": 2.7698426246643066
    },
    {
      "auxiliary_loss_clip": 0.01285602,
      "auxiliary_loss_mlp": 0.02561809,
      "balance_loss_clip": 1.05588913,
      "balance_loss_mlp": 1.00002038,
      "epoch": 0.39223230926471475,
      "flos": 21066276908160.0,
      "grad_norm": 1.7862208627292369,
      "language_loss": 0.84573597,
      "learning_rate": 2.774533258339203e-06,
      "loss": 0.88421011,
      "num_input_tokens_seen": 70307060,
      "step": 3262,
      "time_per_iteration": 2.7084884643554688
    },
    {
      "auxiliary_loss_clip": 0.01303517,
      "auxiliary_loss_mlp": 0.01028677,
      "balance_loss_clip": 1.04819632,
      "balance_loss_mlp": 1.02036202,
      "epoch": 0.3923525521553538,
      "flos": 17603016312960.0,
      "grad_norm": 3.112495679070274,
      "language_loss": 0.80334008,
      "learning_rate": 2.7738150128614014e-06,
      "loss": 0.82666206,
      "num_input_tokens_seen": 70324465,
      "step": 3263,
      "time_per_iteration": 3.616485357284546
    },
    {
      "auxiliary_loss_clip": 0.01335217,
      "auxiliary_loss_mlp": 0.01033808,
      "balance_loss_clip": 1.05119491,
      "balance_loss_mlp": 1.02638757,
      "epoch": 0.3924727950459929,
      "flos": 20558284813440.0,
      "grad_norm": 2.0206633196874244,
      "language_loss": 0.89773262,
      "learning_rate": 2.7730966499964777e-06,
      "loss": 0.92142284,
      "num_input_tokens_seen": 70341415,
      "step": 3264,
      "time_per_iteration": 2.678500175476074
    },
    {
      "auxiliary_loss_clip": 0.01193061,
      "auxiliary_loss_mlp": 0.01028885,
      "balance_loss_clip": 1.05463481,
      "balance_loss_mlp": 1.02047443,
      "epoch": 0.39259303793663197,
      "flos": 16216110328320.0,
      "grad_norm": 3.109206614820065,
      "language_loss": 0.80980968,
      "learning_rate": 2.772378169853408e-06,
      "loss": 0.83202916,
      "num_input_tokens_seen": 70358985,
      "step": 3265,
      "time_per_iteration": 2.664966344833374
    },
    {
      "auxiliary_loss_clip": 0.01338433,
      "auxiliary_loss_mlp": 0.01033042,
      "balance_loss_clip": 1.05239952,
      "balance_loss_mlp": 1.02553523,
      "epoch": 0.3927132808272711,
      "flos": 16797001075200.0,
      "grad_norm": 2.1571245791492584,
      "language_loss": 0.73979998,
      "learning_rate": 2.771659572541183e-06,
      "loss": 0.76351476,
      "num_input_tokens_seen": 70376915,
      "step": 3266,
      "time_per_iteration": 2.6380715370178223
    },
    {
      "auxiliary_loss_clip": 0.01247782,
      "auxiliary_loss_mlp": 0.01026689,
      "balance_loss_clip": 1.05721474,
      "balance_loss_mlp": 1.01943457,
      "epoch": 0.3928335237179102,
      "flos": 20267228908800.0,
      "grad_norm": 2.1545356422842903,
      "language_loss": 0.86889136,
      "learning_rate": 2.7709408581688143e-06,
      "loss": 0.89163613,
      "num_input_tokens_seen": 70396900,
      "step": 3267,
      "time_per_iteration": 4.510973691940308
    },
    {
      "auxiliary_loss_clip": 0.0125283,
      "auxiliary_loss_mlp": 0.01030914,
      "balance_loss_clip": 1.05335593,
      "balance_loss_mlp": 1.02360082,
      "epoch": 0.39295376660854925,
      "flos": 24973250209920.0,
      "grad_norm": 1.639623453251414,
      "language_loss": 0.88102746,
      "learning_rate": 2.7702220268453307e-06,
      "loss": 0.90386498,
      "num_input_tokens_seen": 70417260,
      "step": 3268,
      "time_per_iteration": 2.806018829345703
    },
    {
      "auxiliary_loss_clip": 0.01294633,
      "auxiliary_loss_mlp": 0.01030995,
      "balance_loss_clip": 1.0546509,
      "balance_loss_mlp": 1.02283525,
      "epoch": 0.39307400949918836,
      "flos": 18697788984960.0,
      "grad_norm": 1.9915248051040308,
      "language_loss": 0.84905672,
      "learning_rate": 2.7695030786797785e-06,
      "loss": 0.87231302,
      "num_input_tokens_seen": 70433155,
      "step": 3269,
      "time_per_iteration": 2.81109881401062
    },
    {
      "auxiliary_loss_clip": 0.01389542,
      "auxiliary_loss_mlp": 0.01029531,
      "balance_loss_clip": 1.04889536,
      "balance_loss_mlp": 1.02126384,
      "epoch": 0.39319425238982747,
      "flos": 22415476590720.0,
      "grad_norm": 3.384549475696944,
      "language_loss": 0.7450344,
      "learning_rate": 2.7687840137812206e-06,
      "loss": 0.76922512,
      "num_input_tokens_seen": 70451240,
      "step": 3270,
      "time_per_iteration": 2.8113348484039307
    },
    {
      "auxiliary_loss_clip": 0.01127355,
      "auxiliary_loss_mlp": 0.01012359,
      "balance_loss_clip": 1.01374555,
      "balance_loss_mlp": 1.01150632,
      "epoch": 0.3933144952804665,
      "flos": 66192954762240.0,
      "grad_norm": 0.797242445133077,
      "language_loss": 0.62038964,
      "learning_rate": 2.7680648322587395e-06,
      "loss": 0.64178681,
      "num_input_tokens_seen": 70516115,
      "step": 3271,
      "time_per_iteration": 3.3045899868011475
    },
    {
      "auxiliary_loss_clip": 0.01190673,
      "auxiliary_loss_mlp": 0.01026108,
      "balance_loss_clip": 1.05452871,
      "balance_loss_mlp": 1.01812065,
      "epoch": 0.39343473817110564,
      "flos": 15487159720320.0,
      "grad_norm": 4.900614238782518,
      "language_loss": 0.80788267,
      "learning_rate": 2.7673455342214334e-06,
      "loss": 0.83005047,
      "num_input_tokens_seen": 70533105,
      "step": 3272,
      "time_per_iteration": 2.7263402938842773
    },
    {
      "auxiliary_loss_clip": 0.01241852,
      "auxiliary_loss_mlp": 0.01027843,
      "balance_loss_clip": 1.0547415,
      "balance_loss_mlp": 1.01980233,
      "epoch": 0.39355498106174475,
      "flos": 21324905809920.0,
      "grad_norm": 6.923640798923827,
      "language_loss": 0.7631588,
      "learning_rate": 2.7666261197784198e-06,
      "loss": 0.78585577,
      "num_input_tokens_seen": 70551920,
      "step": 3273,
      "time_per_iteration": 2.6707022190093994
    },
    {
      "auxiliary_loss_clip": 0.01289705,
      "auxiliary_loss_mlp": 0.0102917,
      "balance_loss_clip": 1.05685616,
      "balance_loss_mlp": 1.02167201,
      "epoch": 0.3936752239523838,
      "flos": 13296357400320.0,
      "grad_norm": 2.0474826876181136,
      "language_loss": 0.76558286,
      "learning_rate": 2.7659065890388336e-06,
      "loss": 0.78877163,
      "num_input_tokens_seen": 70567920,
      "step": 3274,
      "time_per_iteration": 2.6100752353668213
    },
    {
      "auxiliary_loss_clip": 0.01291491,
      "auxiliary_loss_mlp": 0.01027672,
      "balance_loss_clip": 1.05062079,
      "balance_loss_mlp": 1.01993513,
      "epoch": 0.3937954668430229,
      "flos": 16800161472000.0,
      "grad_norm": 1.9178533308482817,
      "language_loss": 0.85251069,
      "learning_rate": 2.7651869421118266e-06,
      "loss": 0.87570238,
      "num_input_tokens_seen": 70584530,
      "step": 3275,
      "time_per_iteration": 2.700011730194092
    },
    {
      "auxiliary_loss_clip": 0.01153894,
      "auxiliary_loss_mlp": 0.01026228,
      "balance_loss_clip": 1.05792212,
      "balance_loss_mlp": 1.01857531,
      "epoch": 0.393915709733662,
      "flos": 21064229832960.0,
      "grad_norm": 2.121178767510199,
      "language_loss": 0.83139509,
      "learning_rate": 2.76446717910657e-06,
      "loss": 0.85319626,
      "num_input_tokens_seen": 70605235,
      "step": 3276,
      "time_per_iteration": 2.685286521911621
    },
    {
      "auxiliary_loss_clip": 0.01237856,
      "auxiliary_loss_mlp": 0.01031313,
      "balance_loss_clip": 1.05418444,
      "balance_loss_mlp": 1.02391064,
      "epoch": 0.3940359526243011,
      "flos": 17165265264000.0,
      "grad_norm": 2.659093262561644,
      "language_loss": 0.77011096,
      "learning_rate": 2.763747300132249e-06,
      "loss": 0.79280269,
      "num_input_tokens_seen": 70622675,
      "step": 3277,
      "time_per_iteration": 2.660062551498413
    },
    {
      "auxiliary_loss_clip": 0.01194832,
      "auxiliary_loss_mlp": 0.01027421,
      "balance_loss_clip": 1.05718136,
      "balance_loss_mlp": 1.01950002,
      "epoch": 0.3941561955149402,
      "flos": 20995856294400.0,
      "grad_norm": 1.7477201903478525,
      "language_loss": 0.86770666,
      "learning_rate": 2.7630273052980704e-06,
      "loss": 0.88992918,
      "num_input_tokens_seen": 70643265,
      "step": 3278,
      "time_per_iteration": 2.6150243282318115
    },
    {
      "auxiliary_loss_clip": 0.01286425,
      "auxiliary_loss_mlp": 0.01030844,
      "balance_loss_clip": 1.05197024,
      "balance_loss_mlp": 1.02359009,
      "epoch": 0.39427643840557924,
      "flos": 18843406721280.0,
      "grad_norm": 2.9884349369459478,
      "language_loss": 0.66784751,
      "learning_rate": 2.7623071947132554e-06,
      "loss": 0.69102025,
      "num_input_tokens_seen": 70660295,
      "step": 3279,
      "time_per_iteration": 2.702446699142456
    },
    {
      "auxiliary_loss_clip": 0.0129933,
      "auxiliary_loss_mlp": 0.01027558,
      "balance_loss_clip": 1.05270195,
      "balance_loss_mlp": 1.02017283,
      "epoch": 0.39439668129621835,
      "flos": 23258659426560.0,
      "grad_norm": 2.812149234159523,
      "language_loss": 0.78947353,
      "learning_rate": 2.7615869684870458e-06,
      "loss": 0.81274247,
      "num_input_tokens_seen": 70679605,
      "step": 3280,
      "time_per_iteration": 2.7098050117492676
    },
    {
      "auxiliary_loss_clip": 0.0124255,
      "auxiliary_loss_mlp": 0.01028007,
      "balance_loss_clip": 1.05634022,
      "balance_loss_mlp": 1.02063966,
      "epoch": 0.39451692418685746,
      "flos": 26652289507200.0,
      "grad_norm": 1.7765846246657737,
      "language_loss": 0.8424561,
      "learning_rate": 2.7608666267286986e-06,
      "loss": 0.86516166,
      "num_input_tokens_seen": 70699835,
      "step": 3281,
      "time_per_iteration": 2.7536628246307373
    },
    {
      "auxiliary_loss_clip": 0.01431081,
      "auxiliary_loss_mlp": 0.01025459,
      "balance_loss_clip": 1.04351711,
      "balance_loss_mlp": 1.01702189,
      "epoch": 0.3946371670774965,
      "flos": 18258709132800.0,
      "grad_norm": 2.0709055788254673,
      "language_loss": 0.86614782,
      "learning_rate": 2.760146169547489e-06,
      "loss": 0.89071321,
      "num_input_tokens_seen": 70716600,
      "step": 3282,
      "time_per_iteration": 2.823744773864746
    },
    {
      "auxiliary_loss_clip": 0.01294725,
      "auxiliary_loss_mlp": 0.01031983,
      "balance_loss_clip": 1.05772877,
      "balance_loss_mlp": 1.0245595,
      "epoch": 0.39475740996813563,
      "flos": 24206126423040.0,
      "grad_norm": 1.4928448021832808,
      "language_loss": 0.76665246,
      "learning_rate": 2.75942559705271e-06,
      "loss": 0.78991961,
      "num_input_tokens_seen": 70736335,
      "step": 3283,
      "time_per_iteration": 3.6473917961120605
    },
    {
      "auxiliary_loss_clip": 0.01240648,
      "auxiliary_loss_mlp": 0.01028734,
      "balance_loss_clip": 1.05413222,
      "balance_loss_mlp": 1.02124727,
      "epoch": 0.39487765285877474,
      "flos": 19317858491520.0,
      "grad_norm": 1.8505414587863103,
      "language_loss": 0.89268589,
      "learning_rate": 2.7587049093536713e-06,
      "loss": 0.91537964,
      "num_input_tokens_seen": 70752665,
      "step": 3284,
      "time_per_iteration": 2.6279995441436768
    },
    {
      "auxiliary_loss_clip": 0.01152365,
      "auxiliary_loss_mlp": 0.01026878,
      "balance_loss_clip": 1.0547688,
      "balance_loss_mlp": 1.0194397,
      "epoch": 0.3949978957494138,
      "flos": 17311744926720.0,
      "grad_norm": 3.038054704846765,
      "language_loss": 0.80735308,
      "learning_rate": 2.757984106559701e-06,
      "loss": 0.82914555,
      "num_input_tokens_seen": 70771650,
      "step": 3285,
      "time_per_iteration": 2.5860536098480225
    },
    {
      "auxiliary_loss_clip": 0.01286712,
      "auxiliary_loss_mlp": 0.01029478,
      "balance_loss_clip": 1.0531292,
      "balance_loss_mlp": 1.02177453,
      "epoch": 0.3951181386400529,
      "flos": 36317861280000.0,
      "grad_norm": 5.200109994164949,
      "language_loss": 0.7171396,
      "learning_rate": 2.7572631887801446e-06,
      "loss": 0.74030149,
      "num_input_tokens_seen": 70793275,
      "step": 3286,
      "time_per_iteration": 2.8399264812469482
    },
    {
      "auxiliary_loss_clip": 0.0124379,
      "auxiliary_loss_mlp": 0.01035301,
      "balance_loss_clip": 1.05547214,
      "balance_loss_mlp": 1.02749324,
      "epoch": 0.395238381530692,
      "flos": 23110348170240.0,
      "grad_norm": 1.8940907944578396,
      "language_loss": 0.77038461,
      "learning_rate": 2.7565421561243654e-06,
      "loss": 0.79317552,
      "num_input_tokens_seen": 70811440,
      "step": 3287,
      "time_per_iteration": 2.6005754470825195
    },
    {
      "auxiliary_loss_clip": 0.01337567,
      "auxiliary_loss_mlp": 0.01023191,
      "balance_loss_clip": 1.0489862,
      "balance_loss_mlp": 1.01615167,
      "epoch": 0.3953586244213311,
      "flos": 24347614095360.0,
      "grad_norm": 2.14467114322596,
      "language_loss": 0.82117373,
      "learning_rate": 2.7558210087017413e-06,
      "loss": 0.8447814,
      "num_input_tokens_seen": 70831375,
      "step": 3288,
      "time_per_iteration": 3.5893282890319824
    },
    {
      "auxiliary_loss_clip": 0.01343547,
      "auxiliary_loss_mlp": 0.01036953,
      "balance_loss_clip": 1.05726743,
      "balance_loss_mlp": 1.02910352,
      "epoch": 0.3954788673119702,
      "flos": 23440080044160.0,
      "grad_norm": 3.700661605450855,
      "language_loss": 0.73529458,
      "learning_rate": 2.7550997466216724e-06,
      "loss": 0.75909948,
      "num_input_tokens_seen": 70849170,
      "step": 3289,
      "time_per_iteration": 2.620497703552246
    },
    {
      "auxiliary_loss_clip": 0.01296668,
      "auxiliary_loss_mlp": 0.01031391,
      "balance_loss_clip": 1.06039512,
      "balance_loss_mlp": 1.02348113,
      "epoch": 0.3955991102026093,
      "flos": 17494063384320.0,
      "grad_norm": 2.5382870670290485,
      "language_loss": 0.81261182,
      "learning_rate": 2.7543783699935714e-06,
      "loss": 0.83589244,
      "num_input_tokens_seen": 70867200,
      "step": 3290,
      "time_per_iteration": 2.559481620788574
    },
    {
      "auxiliary_loss_clip": 0.0124439,
      "auxiliary_loss_mlp": 0.01027771,
      "balance_loss_clip": 1.05920482,
      "balance_loss_mlp": 1.01968241,
      "epoch": 0.39571935309324835,
      "flos": 18221326053120.0,
      "grad_norm": 2.41417311045862,
      "language_loss": 0.86179733,
      "learning_rate": 2.753656878926872e-06,
      "loss": 0.88451892,
      "num_input_tokens_seen": 70883080,
      "step": 3291,
      "time_per_iteration": 2.5224215984344482
    },
    {
      "auxiliary_loss_clip": 0.01285027,
      "auxiliary_loss_mlp": 0.01024072,
      "balance_loss_clip": 1.05221963,
      "balance_loss_mlp": 1.01704454,
      "epoch": 0.39583959598388746,
      "flos": 17748813617280.0,
      "grad_norm": 1.8663559904964206,
      "language_loss": 0.74312085,
      "learning_rate": 2.752935273531023e-06,
      "loss": 0.76621187,
      "num_input_tokens_seen": 70901230,
      "step": 3292,
      "time_per_iteration": 2.571831464767456
    },
    {
      "auxiliary_loss_clip": 0.0124853,
      "auxiliary_loss_mlp": 0.01023351,
      "balance_loss_clip": 1.05939722,
      "balance_loss_mlp": 1.01537287,
      "epoch": 0.39595983887452657,
      "flos": 19352368483200.0,
      "grad_norm": 2.7353929437720215,
      "language_loss": 0.78323334,
      "learning_rate": 2.752213553915492e-06,
      "loss": 0.80595219,
      "num_input_tokens_seen": 70919585,
      "step": 3293,
      "time_per_iteration": 3.50278902053833
    },
    {
      "auxiliary_loss_clip": 0.01182349,
      "auxiliary_loss_mlp": 0.01009323,
      "balance_loss_clip": 1.01585054,
      "balance_loss_mlp": 1.00832164,
      "epoch": 0.3960800817651656,
      "flos": 60682282940160.0,
      "grad_norm": 0.8169458486840196,
      "language_loss": 0.66059256,
      "learning_rate": 2.751491720189762e-06,
      "loss": 0.6825093,
      "num_input_tokens_seen": 70977695,
      "step": 3294,
      "time_per_iteration": 4.016615390777588
    },
    {
      "auxiliary_loss_clip": 0.01292256,
      "auxiliary_loss_mlp": 0.02566887,
      "balance_loss_clip": 1.05558741,
      "balance_loss_mlp": 1.00011516,
      "epoch": 0.39620032465580474,
      "flos": 16836718538880.0,
      "grad_norm": 2.980766852725907,
      "language_loss": 0.9179132,
      "learning_rate": 2.7507697724633364e-06,
      "loss": 0.95650464,
      "num_input_tokens_seen": 70994455,
      "step": 3295,
      "time_per_iteration": 2.7509605884552
    },
    {
      "auxiliary_loss_clip": 0.01250553,
      "auxiliary_loss_mlp": 0.01004187,
      "balance_loss_clip": 1.0248785,
      "balance_loss_mlp": 1.00304878,
      "epoch": 0.3963205675464438,
      "flos": 69071445941760.0,
      "grad_norm": 0.7751601435089982,
      "language_loss": 0.54652143,
      "learning_rate": 2.7500477108457327e-06,
      "loss": 0.56906891,
      "num_input_tokens_seen": 71046465,
      "step": 3296,
      "time_per_iteration": 3.0880935192108154
    },
    {
      "auxiliary_loss_clip": 0.01242113,
      "auxiliary_loss_mlp": 0.01027426,
      "balance_loss_clip": 1.05393052,
      "balance_loss_mlp": 1.01984715,
      "epoch": 0.3964408104370829,
      "flos": 25667439431040.0,
      "grad_norm": 2.5022442228868713,
      "language_loss": 0.81166101,
      "learning_rate": 2.7493255354464877e-06,
      "loss": 0.83435631,
      "num_input_tokens_seen": 71064275,
      "step": 3297,
      "time_per_iteration": 2.6488380432128906
    },
    {
      "auxiliary_loss_clip": 0.01578364,
      "auxiliary_loss_mlp": 0.01029128,
      "balance_loss_clip": 1.03987646,
      "balance_loss_mlp": 1.0220176,
      "epoch": 0.396561053327722,
      "flos": 24277480790400.0,
      "grad_norm": 1.9623778353569905,
      "language_loss": 0.76508796,
      "learning_rate": 2.748603246375156e-06,
      "loss": 0.79116285,
      "num_input_tokens_seen": 71082290,
      "step": 3298,
      "time_per_iteration": 3.0068135261535645
    },
    {
      "auxiliary_loss_clip": 0.01193887,
      "auxiliary_loss_mlp": 0.01026152,
      "balance_loss_clip": 1.05782127,
      "balance_loss_mlp": 1.01807547,
      "epoch": 0.39668129621836107,
      "flos": 20522302364160.0,
      "grad_norm": 2.33582031516074,
      "language_loss": 0.69684386,
      "learning_rate": 2.7478808437413055e-06,
      "loss": 0.71904421,
      "num_input_tokens_seen": 71101700,
      "step": 3299,
      "time_per_iteration": 3.4688849449157715
    },
    {
      "auxiliary_loss_clip": 0.01397135,
      "auxiliary_loss_mlp": 0.01031163,
      "balance_loss_clip": 1.0578351,
      "balance_loss_mlp": 1.02346218,
      "epoch": 0.3968015391090002,
      "flos": 27052585649280.0,
      "grad_norm": 1.6802374410180145,
      "language_loss": 0.66052914,
      "learning_rate": 2.7471583276545263e-06,
      "loss": 0.68481207,
      "num_input_tokens_seen": 71122360,
      "step": 3300,
      "time_per_iteration": 2.8056371212005615
    },
    {
      "auxiliary_loss_clip": 0.01294092,
      "auxiliary_loss_mlp": 0.01027548,
      "balance_loss_clip": 1.05432653,
      "balance_loss_mlp": 1.01979351,
      "epoch": 0.3969217819996393,
      "flos": 12531819392640.0,
      "grad_norm": 1.866462683453211,
      "language_loss": 0.7108041,
      "learning_rate": 2.7464356982244224e-06,
      "loss": 0.73402047,
      "num_input_tokens_seen": 71140360,
      "step": 3301,
      "time_per_iteration": 2.670027017593384
    },
    {
      "auxiliary_loss_clip": 0.01152369,
      "auxiliary_loss_mlp": 0.01001869,
      "balance_loss_clip": 1.03078127,
      "balance_loss_mlp": 1.00060582,
      "epoch": 0.39704202489027834,
      "flos": 66241399230720.0,
      "grad_norm": 0.7774977409858631,
      "language_loss": 0.61660099,
      "learning_rate": 2.745712955560617e-06,
      "loss": 0.63814342,
      "num_input_tokens_seen": 71196565,
      "step": 3302,
      "time_per_iteration": 3.158172607421875
    },
    {
      "auxiliary_loss_clip": 0.01440585,
      "auxiliary_loss_mlp": 0.01032553,
      "balance_loss_clip": 1.04796684,
      "balance_loss_mlp": 1.02445245,
      "epoch": 0.39716226778091746,
      "flos": 16982982720000.0,
      "grad_norm": 2.381136700060021,
      "language_loss": 0.76963788,
      "learning_rate": 2.7449900997727496e-06,
      "loss": 0.79436922,
      "num_input_tokens_seen": 71214675,
      "step": 3303,
      "time_per_iteration": 2.759643316268921
    },
    {
      "auxiliary_loss_clip": 0.01297463,
      "auxiliary_loss_mlp": 0.01028572,
      "balance_loss_clip": 1.05946565,
      "balance_loss_mlp": 1.0217886,
      "epoch": 0.39728251067155657,
      "flos": 23477139901440.0,
      "grad_norm": 1.7039873590567383,
      "language_loss": 0.84027076,
      "learning_rate": 2.744267130970476e-06,
      "loss": 0.86353111,
      "num_input_tokens_seen": 71234400,
      "step": 3304,
      "time_per_iteration": 2.6617252826690674
    },
    {
      "auxiliary_loss_clip": 0.01287243,
      "auxiliary_loss_mlp": 0.01027761,
      "balance_loss_clip": 1.05351675,
      "balance_loss_mlp": 1.01995289,
      "epoch": 0.3974027535621956,
      "flos": 20704441253760.0,
      "grad_norm": 1.8572114685057362,
      "language_loss": 0.77188563,
      "learning_rate": 2.7435440492634697e-06,
      "loss": 0.79503572,
      "num_input_tokens_seen": 71253725,
      "step": 3305,
      "time_per_iteration": 2.7156810760498047
    },
    {
      "auxiliary_loss_clip": 0.01293839,
      "auxiliary_loss_mlp": 0.01028962,
      "balance_loss_clip": 1.05407023,
      "balance_loss_mlp": 1.0200628,
      "epoch": 0.39752299645283473,
      "flos": 21543278544000.0,
      "grad_norm": 2.1471967114302473,
      "language_loss": 0.67093199,
      "learning_rate": 2.7428208547614228e-06,
      "loss": 0.69415998,
      "num_input_tokens_seen": 71273220,
      "step": 3306,
      "time_per_iteration": 2.720916986465454
    },
    {
      "auxiliary_loss_clip": 0.01241571,
      "auxiliary_loss_mlp": 0.01029682,
      "balance_loss_clip": 1.05591679,
      "balance_loss_mlp": 1.02085161,
      "epoch": 0.39764323934347384,
      "flos": 19208295031680.0,
      "grad_norm": 2.762325696126657,
      "language_loss": 0.77447516,
      "learning_rate": 2.742097547574043e-06,
      "loss": 0.79718769,
      "num_input_tokens_seen": 71291445,
      "step": 3307,
      "time_per_iteration": 2.609881639480591
    },
    {
      "auxiliary_loss_clip": 0.01300534,
      "auxiliary_loss_mlp": 0.02569987,
      "balance_loss_clip": 1.05481827,
      "balance_loss_mlp": 1.00009942,
      "epoch": 0.3977634822341129,
      "flos": 20850202644480.0,
      "grad_norm": 2.171103436077037,
      "language_loss": 0.78094876,
      "learning_rate": 2.7413741278110544e-06,
      "loss": 0.81965399,
      "num_input_tokens_seen": 71310135,
      "step": 3308,
      "time_per_iteration": 3.714444875717163
    },
    {
      "auxiliary_loss_clip": 0.01300129,
      "auxiliary_loss_mlp": 0.01027888,
      "balance_loss_clip": 1.05639458,
      "balance_loss_mlp": 1.0197227,
      "epoch": 0.397883725124752,
      "flos": 39786042038400.0,
      "grad_norm": 2.5496839536406064,
      "language_loss": 0.69372523,
      "learning_rate": 2.7406505955822016e-06,
      "loss": 0.71700537,
      "num_input_tokens_seen": 71331160,
      "step": 3309,
      "time_per_iteration": 2.843712329864502
    },
    {
      "auxiliary_loss_clip": 0.01295521,
      "auxiliary_loss_mlp": 0.01025275,
      "balance_loss_clip": 1.05176926,
      "balance_loss_mlp": 1.01779497,
      "epoch": 0.39800396801539106,
      "flos": 17379507934080.0,
      "grad_norm": 2.188972429276266,
      "language_loss": 0.66598129,
      "learning_rate": 2.7399269509972415e-06,
      "loss": 0.68918926,
      "num_input_tokens_seen": 71345315,
      "step": 3310,
      "time_per_iteration": 2.663363456726074
    },
    {
      "auxiliary_loss_clip": 0.01291598,
      "auxiliary_loss_mlp": 0.01034364,
      "balance_loss_clip": 1.05004406,
      "balance_loss_mlp": 1.02539313,
      "epoch": 0.3981242109060302,
      "flos": 19202764337280.0,
      "grad_norm": 2.6169275175173516,
      "language_loss": 0.84624302,
      "learning_rate": 2.7392031941659514e-06,
      "loss": 0.86950266,
      "num_input_tokens_seen": 71363160,
      "step": 3311,
      "time_per_iteration": 2.6275110244750977
    },
    {
      "auxiliary_loss_clip": 0.01294518,
      "auxiliary_loss_mlp": 0.01034957,
      "balance_loss_clip": 1.06009364,
      "balance_loss_mlp": 1.02717829,
      "epoch": 0.3982444537966693,
      "flos": 24565124903040.0,
      "grad_norm": 1.7576248625471849,
      "language_loss": 0.86000884,
      "learning_rate": 2.7384793251981244e-06,
      "loss": 0.88330352,
      "num_input_tokens_seen": 71382145,
      "step": 3312,
      "time_per_iteration": 2.7231054306030273
    },
    {
      "auxiliary_loss_clip": 0.01248262,
      "auxiliary_loss_mlp": 0.01028664,
      "balance_loss_clip": 1.0545156,
      "balance_loss_mlp": 1.02088571,
      "epoch": 0.39836469668730834,
      "flos": 26213856099840.0,
      "grad_norm": 3.455614947568465,
      "language_loss": 0.80935663,
      "learning_rate": 2.737755344203571e-06,
      "loss": 0.8321259,
      "num_input_tokens_seen": 71402095,
      "step": 3313,
      "time_per_iteration": 2.709806203842163
    },
    {
      "auxiliary_loss_clip": 0.01244761,
      "auxiliary_loss_mlp": 0.01030349,
      "balance_loss_clip": 1.05656004,
      "balance_loss_mlp": 1.02279401,
      "epoch": 0.39848493957794745,
      "flos": 27636134002560.0,
      "grad_norm": 1.888267103019629,
      "language_loss": 0.79903817,
      "learning_rate": 2.7370312512921186e-06,
      "loss": 0.82178926,
      "num_input_tokens_seen": 71423875,
      "step": 3314,
      "time_per_iteration": 3.5513343811035156
    },
    {
      "auxiliary_loss_clip": 0.01292919,
      "auxiliary_loss_mlp": 0.01029424,
      "balance_loss_clip": 1.05162716,
      "balance_loss_mlp": 1.021276,
      "epoch": 0.39860518246858656,
      "flos": 12239326944000.0,
      "grad_norm": 2.5364508638443377,
      "language_loss": 0.7703253,
      "learning_rate": 2.736307046573611e-06,
      "loss": 0.7935487,
      "num_input_tokens_seen": 71439745,
      "step": 3315,
      "time_per_iteration": 2.6591134071350098
    },
    {
      "auxiliary_loss_clip": 0.01191677,
      "auxiliary_loss_mlp": 0.01028263,
      "balance_loss_clip": 1.05461586,
      "balance_loss_mlp": 1.0212028,
      "epoch": 0.3987254253592256,
      "flos": 22379135005440.0,
      "grad_norm": 1.6131799018559925,
      "language_loss": 0.82000637,
      "learning_rate": 2.73558273015791e-06,
      "loss": 0.84220576,
      "num_input_tokens_seen": 71459575,
      "step": 3316,
      "time_per_iteration": 2.6924021244049072
    },
    {
      "auxiliary_loss_clip": 0.01196008,
      "auxiliary_loss_mlp": 0.01034136,
      "balance_loss_clip": 1.05547428,
      "balance_loss_mlp": 1.0254513,
      "epoch": 0.3988456682498647,
      "flos": 23514020190720.0,
      "grad_norm": 2.1492199888616015,
      "language_loss": 0.70531988,
      "learning_rate": 2.734858302154894e-06,
      "loss": 0.72762132,
      "num_input_tokens_seen": 71481075,
      "step": 3317,
      "time_per_iteration": 2.6544349193573
    },
    {
      "auxiliary_loss_clip": 0.01287206,
      "auxiliary_loss_mlp": 0.01028715,
      "balance_loss_clip": 1.05366027,
      "balance_loss_mlp": 1.02037609,
      "epoch": 0.39896591114050384,
      "flos": 19208761908480.0,
      "grad_norm": 2.0817342952221716,
      "language_loss": 0.7675792,
      "learning_rate": 2.734133762674457e-06,
      "loss": 0.79073834,
      "num_input_tokens_seen": 71500665,
      "step": 3318,
      "time_per_iteration": 3.6260833740234375
    },
    {
      "auxiliary_loss_clip": 0.01291196,
      "auxiliary_loss_mlp": 0.01026032,
      "balance_loss_clip": 1.0535295,
      "balance_loss_mlp": 1.01763928,
      "epoch": 0.3990861540311429,
      "flos": 28401031146240.0,
      "grad_norm": 1.9149781791039362,
      "language_loss": 0.70507121,
      "learning_rate": 2.7334091118265124e-06,
      "loss": 0.72824347,
      "num_input_tokens_seen": 71522560,
      "step": 3319,
      "time_per_iteration": 2.704987049102783
    },
    {
      "auxiliary_loss_clip": 0.01139555,
      "auxiliary_loss_mlp": 0.01004624,
      "balance_loss_clip": 1.01956773,
      "balance_loss_mlp": 1.00352716,
      "epoch": 0.399206396921782,
      "flos": 61758563086080.0,
      "grad_norm": 0.6701135211042666,
      "language_loss": 0.57765323,
      "learning_rate": 2.732684349720989e-06,
      "loss": 0.59909505,
      "num_input_tokens_seen": 71590520,
      "step": 3320,
      "time_per_iteration": 4.116127252578735
    },
    {
      "auxiliary_loss_clip": 0.01257973,
      "auxiliary_loss_mlp": 0.01031908,
      "balance_loss_clip": 1.05572653,
      "balance_loss_mlp": 1.02263927,
      "epoch": 0.3993266398124211,
      "flos": 28074567409920.0,
      "grad_norm": 1.560645316632205,
      "language_loss": 0.75292623,
      "learning_rate": 2.7319594764678318e-06,
      "loss": 0.77582508,
      "num_input_tokens_seen": 71612620,
      "step": 3321,
      "time_per_iteration": 2.779829502105713
    },
    {
      "auxiliary_loss_clip": 0.0129509,
      "auxiliary_loss_mlp": 0.01034601,
      "balance_loss_clip": 1.05303991,
      "balance_loss_mlp": 1.02625048,
      "epoch": 0.39944688270306017,
      "flos": 23225083188480.0,
      "grad_norm": 1.7755111190030972,
      "language_loss": 0.83402681,
      "learning_rate": 2.7312344921770044e-06,
      "loss": 0.85732365,
      "num_input_tokens_seen": 71634320,
      "step": 3322,
      "time_per_iteration": 2.7767059803009033
    },
    {
      "auxiliary_loss_clip": 0.01293893,
      "auxiliary_loss_mlp": 0.01033883,
      "balance_loss_clip": 1.05098557,
      "balance_loss_mlp": 1.02590823,
      "epoch": 0.3995671255936993,
      "flos": 19390433921280.0,
      "grad_norm": 1.930147914256232,
      "language_loss": 0.7852931,
      "learning_rate": 2.7305093969584857e-06,
      "loss": 0.80857086,
      "num_input_tokens_seen": 71653145,
      "step": 3323,
      "time_per_iteration": 2.69221830368042
    },
    {
      "auxiliary_loss_clip": 0.01239571,
      "auxiliary_loss_mlp": 0.01029662,
      "balance_loss_clip": 1.0536797,
      "balance_loss_mlp": 1.02123976,
      "epoch": 0.3996873684843384,
      "flos": 23842638743040.0,
      "grad_norm": 2.2185610159361615,
      "language_loss": 0.79949725,
      "learning_rate": 2.729784190922272e-06,
      "loss": 0.82218957,
      "num_input_tokens_seen": 71674580,
      "step": 3324,
      "time_per_iteration": 2.6421306133270264
    },
    {
      "auxiliary_loss_clip": 0.01089623,
      "auxiliary_loss_mlp": 0.01005672,
      "balance_loss_clip": 1.0167737,
      "balance_loss_mlp": 1.00468874,
      "epoch": 0.39980761137497745,
      "flos": 66576877280640.0,
      "grad_norm": 0.9554523356177217,
      "language_loss": 0.57175618,
      "learning_rate": 2.729058874178378e-06,
      "loss": 0.59270912,
      "num_input_tokens_seen": 71745260,
      "step": 3325,
      "time_per_iteration": 3.3300294876098633
    },
    {
      "auxiliary_loss_clip": 0.01297965,
      "auxiliary_loss_mlp": 0.01030029,
      "balance_loss_clip": 1.05582643,
      "balance_loss_mlp": 1.0215354,
      "epoch": 0.39992785426561656,
      "flos": 28549162834560.0,
      "grad_norm": 2.236547358510773,
      "language_loss": 0.69094986,
      "learning_rate": 2.7283334468368315e-06,
      "loss": 0.71422982,
      "num_input_tokens_seen": 71766540,
      "step": 3326,
      "time_per_iteration": 2.8591043949127197
    },
    {
      "auxiliary_loss_clip": 0.01445364,
      "auxiliary_loss_mlp": 0.01034996,
      "balance_loss_clip": 1.03933978,
      "balance_loss_mlp": 1.02632332,
      "epoch": 0.4000480971562556,
      "flos": 15049408671360.0,
      "grad_norm": 10.342615399466746,
      "language_loss": 0.72934306,
      "learning_rate": 2.72760790900768e-06,
      "loss": 0.7541467,
      "num_input_tokens_seen": 71783125,
      "step": 3327,
      "time_per_iteration": 3.047367572784424
    },
    {
      "auxiliary_loss_clip": 0.01199276,
      "auxiliary_loss_mlp": 0.01035924,
      "balance_loss_clip": 1.05997562,
      "balance_loss_mlp": 1.02759671,
      "epoch": 0.4001683400468947,
      "flos": 23915609222400.0,
      "grad_norm": 1.7776318647815952,
      "language_loss": 0.78804386,
      "learning_rate": 2.7268822608009875e-06,
      "loss": 0.81039584,
      "num_input_tokens_seen": 71802500,
      "step": 3328,
      "time_per_iteration": 2.8163461685180664
    },
    {
      "auxiliary_loss_clip": 0.0134888,
      "auxiliary_loss_mlp": 0.010338,
      "balance_loss_clip": 1.05400598,
      "balance_loss_mlp": 1.02510941,
      "epoch": 0.40028858293753383,
      "flos": 24352677912960.0,
      "grad_norm": 1.8246560634670606,
      "language_loss": 0.78543776,
      "learning_rate": 2.726156502326834e-06,
      "loss": 0.80926454,
      "num_input_tokens_seen": 71823800,
      "step": 3329,
      "time_per_iteration": 2.7694475650787354
    },
    {
      "auxiliary_loss_clip": 0.01271435,
      "auxiliary_loss_mlp": 0.01003421,
      "balance_loss_clip": 1.02671182,
      "balance_loss_mlp": 1.00221086,
      "epoch": 0.4004088258281729,
      "flos": 66787025800320.0,
      "grad_norm": 0.7082745624370083,
      "language_loss": 0.60276765,
      "learning_rate": 2.725430633695316e-06,
      "loss": 0.62551618,
      "num_input_tokens_seen": 71886880,
      "step": 3330,
      "time_per_iteration": 3.4696404933929443
    },
    {
      "auxiliary_loss_clip": 0.01080256,
      "auxiliary_loss_mlp": 0.01002513,
      "balance_loss_clip": 1.01435566,
      "balance_loss_mlp": 1.00150001,
      "epoch": 0.400529068718812,
      "flos": 58598386473600.0,
      "grad_norm": 0.8779240877470704,
      "language_loss": 0.57930148,
      "learning_rate": 2.7247046550165485e-06,
      "loss": 0.60012925,
      "num_input_tokens_seen": 71939005,
      "step": 3331,
      "time_per_iteration": 3.7733991146087646
    },
    {
      "auxiliary_loss_clip": 0.01195906,
      "auxiliary_loss_mlp": 0.01031367,
      "balance_loss_clip": 1.05769598,
      "balance_loss_mlp": 1.02321851,
      "epoch": 0.4006493116094511,
      "flos": 25377460934400.0,
      "grad_norm": 1.7075775088813077,
      "language_loss": 0.75934601,
      "learning_rate": 2.7239785664006606e-06,
      "loss": 0.78161871,
      "num_input_tokens_seen": 71962545,
      "step": 3332,
      "time_per_iteration": 2.7241997718811035
    },
    {
      "auxiliary_loss_clip": 0.01134083,
      "auxiliary_loss_mlp": 0.01003338,
      "balance_loss_clip": 1.01340628,
      "balance_loss_mlp": 1.00232506,
      "epoch": 0.40076955450009016,
      "flos": 60280729822080.0,
      "grad_norm": 0.8807791959000575,
      "language_loss": 0.61749184,
      "learning_rate": 2.7232523679578002e-06,
      "loss": 0.63886607,
      "num_input_tokens_seen": 72025625,
      "step": 3333,
      "time_per_iteration": 3.3043599128723145
    },
    {
      "auxiliary_loss_clip": 0.01241597,
      "auxiliary_loss_mlp": 0.01033463,
      "balance_loss_clip": 1.0550611,
      "balance_loss_mlp": 1.02570248,
      "epoch": 0.4008897973907293,
      "flos": 16617268396800.0,
      "grad_norm": 2.2836800894736977,
      "language_loss": 0.79478645,
      "learning_rate": 2.7225260597981295e-06,
      "loss": 0.81753707,
      "num_input_tokens_seen": 72043330,
      "step": 3334,
      "time_per_iteration": 3.8916537761688232
    },
    {
      "auxiliary_loss_clip": 0.01348393,
      "auxiliary_loss_mlp": 0.02577439,
      "balance_loss_clip": 1.05716002,
      "balance_loss_mlp": 1.00016308,
      "epoch": 0.4010100402813684,
      "flos": 15377344865280.0,
      "grad_norm": 2.4312013847600005,
      "language_loss": 0.78860152,
      "learning_rate": 2.721799642031831e-06,
      "loss": 0.82785982,
      "num_input_tokens_seen": 72059500,
      "step": 3335,
      "time_per_iteration": 2.8428988456726074
    },
    {
      "auxiliary_loss_clip": 0.01298223,
      "auxiliary_loss_mlp": 0.01027148,
      "balance_loss_clip": 1.05105066,
      "balance_loss_mlp": 1.01895249,
      "epoch": 0.40113028317200744,
      "flos": 13298835438720.0,
      "grad_norm": 2.17940531073706,
      "language_loss": 0.77303654,
      "learning_rate": 2.721073114769101e-06,
      "loss": 0.79629028,
      "num_input_tokens_seen": 72077175,
      "step": 3336,
      "time_per_iteration": 2.647428035736084
    },
    {
      "auxiliary_loss_clip": 0.01334206,
      "auxiliary_loss_mlp": 0.01028069,
      "balance_loss_clip": 1.05082154,
      "balance_loss_mlp": 1.02039146,
      "epoch": 0.40125052606264655,
      "flos": 20668027841280.0,
      "grad_norm": 2.03578206710025,
      "language_loss": 0.75359064,
      "learning_rate": 2.7203464781201523e-06,
      "loss": 0.77721339,
      "num_input_tokens_seen": 72096490,
      "step": 3337,
      "time_per_iteration": 2.7662100791931152
    },
    {
      "auxiliary_loss_clip": 0.01195564,
      "auxiliary_loss_mlp": 0.01026148,
      "balance_loss_clip": 1.05802536,
      "balance_loss_mlp": 1.01788092,
      "epoch": 0.40137076895328566,
      "flos": 24607679541120.0,
      "grad_norm": 2.3598880145077437,
      "language_loss": 0.78219706,
      "learning_rate": 2.719619732195215e-06,
      "loss": 0.80441421,
      "num_input_tokens_seen": 72118130,
      "step": 3338,
      "time_per_iteration": 2.6748733520507812
    },
    {
      "auxiliary_loss_clip": 0.01340151,
      "auxiliary_loss_mlp": 0.01029111,
      "balance_loss_clip": 1.05125356,
      "balance_loss_mlp": 1.02091503,
      "epoch": 0.4014910118439247,
      "flos": 24206593299840.0,
      "grad_norm": 1.4607547529717289,
      "language_loss": 0.72577107,
      "learning_rate": 2.7188928771045377e-06,
      "loss": 0.74946374,
      "num_input_tokens_seen": 72139450,
      "step": 3339,
      "time_per_iteration": 3.6455564498901367
    },
    {
      "auxiliary_loss_clip": 0.01334814,
      "auxiliary_loss_mlp": 0.01028427,
      "balance_loss_clip": 1.04884338,
      "balance_loss_mlp": 1.02020121,
      "epoch": 0.4016112547345638,
      "flos": 26725080418560.0,
      "grad_norm": 1.9052680503651094,
      "language_loss": 0.80047989,
      "learning_rate": 2.7181659129583815e-06,
      "loss": 0.8241123,
      "num_input_tokens_seen": 72159040,
      "step": 3340,
      "time_per_iteration": 2.754040241241455
    },
    {
      "auxiliary_loss_clip": 0.01286295,
      "auxiliary_loss_mlp": 0.01030464,
      "balance_loss_clip": 1.04707241,
      "balance_loss_mlp": 1.02145195,
      "epoch": 0.4017314976252029,
      "flos": 21288025520640.0,
      "grad_norm": 1.7510604692317788,
      "language_loss": 0.7581104,
      "learning_rate": 2.7174388398670276e-06,
      "loss": 0.78127801,
      "num_input_tokens_seen": 72178220,
      "step": 3341,
      "time_per_iteration": 2.7242462635040283
    },
    {
      "auxiliary_loss_clip": 0.01196002,
      "auxiliary_loss_mlp": 0.01028087,
      "balance_loss_clip": 1.05425644,
      "balance_loss_mlp": 1.02008176,
      "epoch": 0.401851740515842,
      "flos": 25484690010240.0,
      "grad_norm": 2.1431139884821224,
      "language_loss": 0.92168176,
      "learning_rate": 2.716711657940773e-06,
      "loss": 0.94392264,
      "num_input_tokens_seen": 72199230,
      "step": 3342,
      "time_per_iteration": 2.6808359622955322
    },
    {
      "auxiliary_loss_clip": 0.01143051,
      "auxiliary_loss_mlp": 0.01002512,
      "balance_loss_clip": 1.01526237,
      "balance_loss_mlp": 1.00154614,
      "epoch": 0.4019719834064811,
      "flos": 55395334978560.0,
      "grad_norm": 0.8169332649592138,
      "language_loss": 0.564776,
      "learning_rate": 2.7159843672899284e-06,
      "loss": 0.58623165,
      "num_input_tokens_seen": 72263430,
      "step": 3343,
      "time_per_iteration": 4.272258520126343
    },
    {
      "auxiliary_loss_clip": 0.01243858,
      "auxiliary_loss_mlp": 0.01028179,
      "balance_loss_clip": 1.05728781,
      "balance_loss_mlp": 1.01932812,
      "epoch": 0.40209222629712016,
      "flos": 18180100218240.0,
      "grad_norm": 2.2458966713511175,
      "language_loss": 0.81015992,
      "learning_rate": 2.715256968024825e-06,
      "loss": 0.83288026,
      "num_input_tokens_seen": 72280505,
      "step": 3344,
      "time_per_iteration": 2.7222230434417725
    },
    {
      "auxiliary_loss_clip": 0.01204498,
      "auxiliary_loss_mlp": 0.01026283,
      "balance_loss_clip": 1.053087,
      "balance_loss_mlp": 1.01764596,
      "epoch": 0.40221246918775927,
      "flos": 25961009287680.0,
      "grad_norm": 1.7987043981480022,
      "language_loss": 0.81985545,
      "learning_rate": 2.7145294602558083e-06,
      "loss": 0.84216332,
      "num_input_tokens_seen": 72301215,
      "step": 3345,
      "time_per_iteration": 3.687080144882202
    },
    {
      "auxiliary_loss_clip": 0.01244522,
      "auxiliary_loss_mlp": 0.01032441,
      "balance_loss_clip": 1.05507612,
      "balance_loss_mlp": 1.02412581,
      "epoch": 0.4023327120783984,
      "flos": 33838912056960.0,
      "grad_norm": 1.7836952798666454,
      "language_loss": 0.7108376,
      "learning_rate": 2.713801844093241e-06,
      "loss": 0.73360723,
      "num_input_tokens_seen": 72322365,
      "step": 3346,
      "time_per_iteration": 2.801084518432617
    },
    {
      "auxiliary_loss_clip": 0.0124411,
      "auxiliary_loss_mlp": 0.01028516,
      "balance_loss_clip": 1.0543611,
      "balance_loss_mlp": 1.02104163,
      "epoch": 0.40245295496903744,
      "flos": 26900252069760.0,
      "grad_norm": 2.4872325920099847,
      "language_loss": 0.88656473,
      "learning_rate": 2.7130741196475014e-06,
      "loss": 0.90929091,
      "num_input_tokens_seen": 72340495,
      "step": 3347,
      "time_per_iteration": 2.803431510925293
    },
    {
      "auxiliary_loss_clip": 0.01300824,
      "auxiliary_loss_mlp": 0.01032158,
      "balance_loss_clip": 1.0576458,
      "balance_loss_mlp": 1.02310944,
      "epoch": 0.40257319785967655,
      "flos": 36902738436480.0,
      "grad_norm": 1.8593251194811748,
      "language_loss": 0.78827405,
      "learning_rate": 2.7123462870289848e-06,
      "loss": 0.81160378,
      "num_input_tokens_seen": 72360545,
      "step": 3348,
      "time_per_iteration": 2.900256395339966
    },
    {
      "auxiliary_loss_clip": 0.01292639,
      "auxiliary_loss_mlp": 0.01031286,
      "balance_loss_clip": 1.05042195,
      "balance_loss_mlp": 1.02320611,
      "epoch": 0.40269344075031566,
      "flos": 24353180703360.0,
      "grad_norm": 1.584575970016342,
      "language_loss": 0.81224638,
      "learning_rate": 2.711618346348102e-06,
      "loss": 0.83548564,
      "num_input_tokens_seen": 72381070,
      "step": 3349,
      "time_per_iteration": 2.821484327316284
    },
    {
      "auxiliary_loss_clip": 0.01289168,
      "auxiliary_loss_mlp": 0.01029131,
      "balance_loss_clip": 1.05208254,
      "balance_loss_mlp": 1.02139688,
      "epoch": 0.4028136836409547,
      "flos": 14389657614720.0,
      "grad_norm": 1.568844792833661,
      "language_loss": 0.63279533,
      "learning_rate": 2.7108902977152825e-06,
      "loss": 0.65597832,
      "num_input_tokens_seen": 72398970,
      "step": 3350,
      "time_per_iteration": 2.7866334915161133
    },
    {
      "auxiliary_loss_clip": 0.01240947,
      "auxiliary_loss_mlp": 0.01031113,
      "balance_loss_clip": 1.05274796,
      "balance_loss_mlp": 1.02250028,
      "epoch": 0.4029339265315938,
      "flos": 26136037284480.0,
      "grad_norm": 2.2900958353060044,
      "language_loss": 0.74994123,
      "learning_rate": 2.7101621412409704e-06,
      "loss": 0.77266181,
      "num_input_tokens_seen": 72418455,
      "step": 3351,
      "time_per_iteration": 2.7320616245269775
    },
    {
      "auxiliary_loss_clip": 0.01194902,
      "auxiliary_loss_mlp": 0.01026902,
      "balance_loss_clip": 1.0557189,
      "balance_loss_mlp": 1.01894426,
      "epoch": 0.40305416942223293,
      "flos": 23256325042560.0,
      "grad_norm": 2.864373787516934,
      "language_loss": 0.85661185,
      "learning_rate": 2.7094338770356256e-06,
      "loss": 0.87882984,
      "num_input_tokens_seen": 72437540,
      "step": 3352,
      "time_per_iteration": 2.7118382453918457
    },
    {
      "auxiliary_loss_clip": 0.01289137,
      "auxiliary_loss_mlp": 0.01029916,
      "balance_loss_clip": 1.05408549,
      "balance_loss_mlp": 1.02193749,
      "epoch": 0.403174412312872,
      "flos": 27089645506560.0,
      "grad_norm": 2.2344342030670004,
      "language_loss": 0.64371872,
      "learning_rate": 2.708705505209726e-06,
      "loss": 0.66690934,
      "num_input_tokens_seen": 72458315,
      "step": 3353,
      "time_per_iteration": 2.759544610977173
    },
    {
      "auxiliary_loss_clip": 0.01385509,
      "auxiliary_loss_mlp": 0.0103162,
      "balance_loss_clip": 1.04511189,
      "balance_loss_mlp": 1.02434278,
      "epoch": 0.4032946552035111,
      "flos": 21756336065280.0,
      "grad_norm": 2.1644730636035856,
      "language_loss": 0.91535425,
      "learning_rate": 2.7079770258737646e-06,
      "loss": 0.9395256,
      "num_input_tokens_seen": 72476225,
      "step": 3354,
      "time_per_iteration": 2.7793240547180176
    },
    {
      "auxiliary_loss_clip": 0.01337383,
      "auxiliary_loss_mlp": 0.01031125,
      "balance_loss_clip": 1.04708195,
      "balance_loss_mlp": 1.02180314,
      "epoch": 0.4034148980941502,
      "flos": 17343956448000.0,
      "grad_norm": 5.901012551227482,
      "language_loss": 0.74688482,
      "learning_rate": 2.707248439138251e-06,
      "loss": 0.77056992,
      "num_input_tokens_seen": 72492460,
      "step": 3355,
      "time_per_iteration": 2.684919595718384
    },
    {
      "auxiliary_loss_clip": 0.01286214,
      "auxiliary_loss_mlp": 0.01031774,
      "balance_loss_clip": 1.05443406,
      "balance_loss_mlp": 1.02365553,
      "epoch": 0.40353514098478926,
      "flos": 22017838055040.0,
      "grad_norm": 3.4025053253156963,
      "language_loss": 0.65404505,
      "learning_rate": 2.7065197451137114e-06,
      "loss": 0.67722493,
      "num_input_tokens_seen": 72513840,
      "step": 3356,
      "time_per_iteration": 2.720374345779419
    },
    {
      "auxiliary_loss_clip": 0.0128769,
      "auxiliary_loss_mlp": 0.01025919,
      "balance_loss_clip": 1.05275595,
      "balance_loss_mlp": 1.01802182,
      "epoch": 0.4036553838754284,
      "flos": 14246446089600.0,
      "grad_norm": 2.6490853710077613,
      "language_loss": 0.67650473,
      "learning_rate": 2.7057909439106894e-06,
      "loss": 0.69964087,
      "num_input_tokens_seen": 72531695,
      "step": 3357,
      "time_per_iteration": 2.656491994857788
    },
    {
      "auxiliary_loss_clip": 0.01237552,
      "auxiliary_loss_mlp": 0.02568078,
      "balance_loss_clip": 1.05116451,
      "balance_loss_mlp": 1.00006366,
      "epoch": 0.40377562676606743,
      "flos": 24790644443520.0,
      "grad_norm": 1.9637958988505428,
      "language_loss": 0.78431571,
      "learning_rate": 2.7050620356397417e-06,
      "loss": 0.82237196,
      "num_input_tokens_seen": 72550645,
      "step": 3358,
      "time_per_iteration": 2.728999137878418
    },
    {
      "auxiliary_loss_clip": 0.01189933,
      "auxiliary_loss_mlp": 0.01027071,
      "balance_loss_clip": 1.05630922,
      "balance_loss_mlp": 1.02004099,
      "epoch": 0.40389586965670654,
      "flos": 24061226958720.0,
      "grad_norm": 1.942299173066065,
      "language_loss": 0.72054285,
      "learning_rate": 2.7043330204114437e-06,
      "loss": 0.74271286,
      "num_input_tokens_seen": 72569355,
      "step": 3359,
      "time_per_iteration": 2.5937905311584473
    },
    {
      "auxiliary_loss_clip": 0.01187928,
      "auxiliary_loss_mlp": 0.01030927,
      "balance_loss_clip": 1.05330944,
      "balance_loss_mlp": 1.02386403,
      "epoch": 0.40401611254734565,
      "flos": 16399613934720.0,
      "grad_norm": 2.1834926539590582,
      "language_loss": 0.85486925,
      "learning_rate": 2.7036038983363862e-06,
      "loss": 0.87705779,
      "num_input_tokens_seen": 72585960,
      "step": 3360,
      "time_per_iteration": 3.4920690059661865
    },
    {
      "auxiliary_loss_clip": 0.01236791,
      "auxiliary_loss_mlp": 0.01025244,
      "balance_loss_clip": 1.05353761,
      "balance_loss_mlp": 1.01803756,
      "epoch": 0.4041363554379847,
      "flos": 23988220565760.0,
      "grad_norm": 1.7737106868961687,
      "language_loss": 0.84050667,
      "learning_rate": 2.702874669525177e-06,
      "loss": 0.86312699,
      "num_input_tokens_seen": 72604440,
      "step": 3361,
      "time_per_iteration": 2.7895946502685547
    },
    {
      "auxiliary_loss_clip": 0.01348294,
      "auxiliary_loss_mlp": 0.01034028,
      "balance_loss_clip": 1.05732775,
      "balance_loss_mlp": 1.02693534,
      "epoch": 0.4042565983286238,
      "flos": 28401964899840.0,
      "grad_norm": 2.087625241364062,
      "language_loss": 0.69874513,
      "learning_rate": 2.7021453340884394e-06,
      "loss": 0.72256839,
      "num_input_tokens_seen": 72622165,
      "step": 3362,
      "time_per_iteration": 2.74658465385437
    },
    {
      "auxiliary_loss_clip": 0.01280426,
      "auxiliary_loss_mlp": 0.02566409,
      "balance_loss_clip": 1.05014813,
      "balance_loss_mlp": 1.00007153,
      "epoch": 0.40437684121926293,
      "flos": 17710963660800.0,
      "grad_norm": 3.881042117670996,
      "language_loss": 0.72844255,
      "learning_rate": 2.7014158921368125e-06,
      "loss": 0.76691091,
      "num_input_tokens_seen": 72640490,
      "step": 3363,
      "time_per_iteration": 2.6948964595794678
    },
    {
      "auxiliary_loss_clip": 0.01193615,
      "auxiliary_loss_mlp": 0.01027109,
      "balance_loss_clip": 1.05677974,
      "balance_loss_mlp": 1.01954508,
      "epoch": 0.404497084109902,
      "flos": 24018959629440.0,
      "grad_norm": 2.4329109692580717,
      "language_loss": 0.85222018,
      "learning_rate": 2.700686343780953e-06,
      "loss": 0.87442738,
      "num_input_tokens_seen": 72660360,
      "step": 3364,
      "time_per_iteration": 2.6615543365478516
    },
    {
      "auxiliary_loss_clip": 0.01289258,
      "auxiliary_loss_mlp": 0.01028943,
      "balance_loss_clip": 1.05116153,
      "balance_loss_mlp": 1.02137649,
      "epoch": 0.4046173270005411,
      "flos": 22929861306240.0,
      "grad_norm": 1.8318350245442518,
      "language_loss": 0.88415122,
      "learning_rate": 2.699956689131532e-06,
      "loss": 0.9073332,
      "num_input_tokens_seen": 72680345,
      "step": 3365,
      "time_per_iteration": 3.7049500942230225
    },
    {
      "auxiliary_loss_clip": 0.01195282,
      "auxiliary_loss_mlp": 0.01031059,
      "balance_loss_clip": 1.05032146,
      "balance_loss_mlp": 1.02331662,
      "epoch": 0.4047375698911802,
      "flos": 20668135582080.0,
      "grad_norm": 2.43030883603922,
      "language_loss": 0.85196763,
      "learning_rate": 2.699226928299238e-06,
      "loss": 0.87423098,
      "num_input_tokens_seen": 72698365,
      "step": 3366,
      "time_per_iteration": 2.687328815460205
    },
    {
      "auxiliary_loss_clip": 0.01238754,
      "auxiliary_loss_mlp": 0.01027642,
      "balance_loss_clip": 1.05116892,
      "balance_loss_mlp": 1.01983345,
      "epoch": 0.40485781278181926,
      "flos": 28912865996160.0,
      "grad_norm": 2.9699882536677147,
      "language_loss": 0.7895304,
      "learning_rate": 2.698497061394774e-06,
      "loss": 0.81219435,
      "num_input_tokens_seen": 72716850,
      "step": 3367,
      "time_per_iteration": 2.6771328449249268
    },
    {
      "auxiliary_loss_clip": 0.01341761,
      "auxiliary_loss_mlp": 0.02565577,
      "balance_loss_clip": 1.05213165,
      "balance_loss_mlp": 1.00015962,
      "epoch": 0.40497805567245837,
      "flos": 23148377694720.0,
      "grad_norm": 1.774272911080005,
      "language_loss": 0.80549943,
      "learning_rate": 2.6977670885288627e-06,
      "loss": 0.84457278,
      "num_input_tokens_seen": 72738250,
      "step": 3368,
      "time_per_iteration": 2.8051929473876953
    },
    {
      "auxiliary_loss_clip": 0.01284338,
      "auxiliary_loss_mlp": 0.01028284,
      "balance_loss_clip": 1.05151141,
      "balance_loss_mlp": 1.02073526,
      "epoch": 0.4050982985630975,
      "flos": 16289404030080.0,
      "grad_norm": 2.017252792274504,
      "language_loss": 0.7510587,
      "learning_rate": 2.6970370098122378e-06,
      "loss": 0.77418488,
      "num_input_tokens_seen": 72755235,
      "step": 3369,
      "time_per_iteration": 3.639983892440796
    },
    {
      "auxiliary_loss_clip": 0.01191303,
      "auxiliary_loss_mlp": 0.01033945,
      "balance_loss_clip": 1.05339086,
      "balance_loss_mlp": 1.02642274,
      "epoch": 0.40521854145373654,
      "flos": 34459484353920.0,
      "grad_norm": 1.7398060869934628,
      "language_loss": 0.86501467,
      "learning_rate": 2.6963068253556535e-06,
      "loss": 0.88726711,
      "num_input_tokens_seen": 72776620,
      "step": 3370,
      "time_per_iteration": 2.755265951156616
    },
    {
      "auxiliary_loss_clip": 0.01151742,
      "auxiliary_loss_mlp": 0.01031216,
      "balance_loss_clip": 1.05219102,
      "balance_loss_mlp": 1.02316904,
      "epoch": 0.40533878434437565,
      "flos": 25331099454720.0,
      "grad_norm": 1.9053074714127805,
      "language_loss": 0.85824722,
      "learning_rate": 2.6955765352698763e-06,
      "loss": 0.88007683,
      "num_input_tokens_seen": 72796765,
      "step": 3371,
      "time_per_iteration": 3.6500113010406494
    },
    {
      "auxiliary_loss_clip": 0.01190865,
      "auxiliary_loss_mlp": 0.01031575,
      "balance_loss_clip": 1.05133247,
      "balance_loss_mlp": 1.02365375,
      "epoch": 0.40545902723501476,
      "flos": 15012061505280.0,
      "grad_norm": 1.9856770466404368,
      "language_loss": 0.732126,
      "learning_rate": 2.6948461396656923e-06,
      "loss": 0.75435042,
      "num_input_tokens_seen": 72814175,
      "step": 3372,
      "time_per_iteration": 2.544132709503174
    },
    {
      "auxiliary_loss_clip": 0.01148994,
      "auxiliary_loss_mlp": 0.01027077,
      "balance_loss_clip": 1.05358052,
      "balance_loss_mlp": 1.01938248,
      "epoch": 0.4055792701256538,
      "flos": 25521103422720.0,
      "grad_norm": 2.2043164021009742,
      "language_loss": 0.74543768,
      "learning_rate": 2.6941156386539013e-06,
      "loss": 0.76719844,
      "num_input_tokens_seen": 72834125,
      "step": 3373,
      "time_per_iteration": 2.699934720993042
    },
    {
      "auxiliary_loss_clip": 0.01285229,
      "auxiliary_loss_mlp": 0.01026746,
      "balance_loss_clip": 1.05369568,
      "balance_loss_mlp": 1.01928687,
      "epoch": 0.4056995130162929,
      "flos": 19574583972480.0,
      "grad_norm": 2.586944821179904,
      "language_loss": 0.81043416,
      "learning_rate": 2.6933850323453203e-06,
      "loss": 0.83355391,
      "num_input_tokens_seen": 72852570,
      "step": 3374,
      "time_per_iteration": 2.699056386947632
    },
    {
      "auxiliary_loss_clip": 0.01192111,
      "auxiliary_loss_mlp": 0.01027072,
      "balance_loss_clip": 1.05659914,
      "balance_loss_mlp": 1.01939797,
      "epoch": 0.405819755906932,
      "flos": 15413794191360.0,
      "grad_norm": 2.29812651259609,
      "language_loss": 0.75108677,
      "learning_rate": 2.6926543208507806e-06,
      "loss": 0.77327859,
      "num_input_tokens_seen": 72871250,
      "step": 3375,
      "time_per_iteration": 2.611058235168457
    },
    {
      "auxiliary_loss_clip": 0.01239724,
      "auxiliary_loss_mlp": 0.01031476,
      "balance_loss_clip": 1.05340123,
      "balance_loss_mlp": 1.02376318,
      "epoch": 0.4059399987975711,
      "flos": 21433930565760.0,
      "grad_norm": 5.665297264710379,
      "language_loss": 0.79952765,
      "learning_rate": 2.6919235042811316e-06,
      "loss": 0.82223964,
      "num_input_tokens_seen": 72890035,
      "step": 3376,
      "time_per_iteration": 2.6198229789733887
    },
    {
      "auxiliary_loss_clip": 0.01339252,
      "auxiliary_loss_mlp": 0.01032999,
      "balance_loss_clip": 1.0509578,
      "balance_loss_mlp": 1.02535725,
      "epoch": 0.4060602416882102,
      "flos": 25556942217600.0,
      "grad_norm": 2.4522781729053857,
      "language_loss": 0.76213086,
      "learning_rate": 2.691192582747237e-06,
      "loss": 0.78585333,
      "num_input_tokens_seen": 72909665,
      "step": 3377,
      "time_per_iteration": 2.802574634552002
    },
    {
      "auxiliary_loss_clip": 0.01191214,
      "auxiliary_loss_mlp": 0.01030672,
      "balance_loss_clip": 1.0553112,
      "balance_loss_mlp": 1.02288759,
      "epoch": 0.40618048457884925,
      "flos": 23766759262080.0,
      "grad_norm": 1.8393495861858566,
      "language_loss": 0.73829019,
      "learning_rate": 2.6904615563599765e-06,
      "loss": 0.76050907,
      "num_input_tokens_seen": 72929465,
      "step": 3378,
      "time_per_iteration": 2.6538519859313965
    },
    {
      "auxiliary_loss_clip": 0.01334183,
      "auxiliary_loss_mlp": 0.01027656,
      "balance_loss_clip": 1.05053568,
      "balance_loss_mlp": 1.0200628,
      "epoch": 0.40630072746948837,
      "flos": 17639681120640.0,
      "grad_norm": 2.1454294761298596,
      "language_loss": 0.83649242,
      "learning_rate": 2.6897304252302477e-06,
      "loss": 0.86011076,
      "num_input_tokens_seen": 72946785,
      "step": 3379,
      "time_per_iteration": 2.778083324432373
    },
    {
      "auxiliary_loss_clip": 0.01243756,
      "auxiliary_loss_mlp": 0.00999922,
      "balance_loss_clip": 1.02318835,
      "balance_loss_mlp": 0.99879593,
      "epoch": 0.4064209703601275,
      "flos": 60836053063680.0,
      "grad_norm": 0.7885054982236578,
      "language_loss": 0.54816544,
      "learning_rate": 2.688999189468962e-06,
      "loss": 0.5706023,
      "num_input_tokens_seen": 73003215,
      "step": 3380,
      "time_per_iteration": 3.159789562225342
    },
    {
      "auxiliary_loss_clip": 0.01245686,
      "auxiliary_loss_mlp": 0.01027492,
      "balance_loss_clip": 1.05834889,
      "balance_loss_mlp": 1.02022302,
      "epoch": 0.40654121325076653,
      "flos": 24024346669440.0,
      "grad_norm": 2.4309364226370267,
      "language_loss": 0.76403844,
      "learning_rate": 2.6882678491870464e-06,
      "loss": 0.78677022,
      "num_input_tokens_seen": 73023650,
      "step": 3381,
      "time_per_iteration": 2.6280388832092285
    },
    {
      "auxiliary_loss_clip": 0.01244315,
      "auxiliary_loss_mlp": 0.01022029,
      "balance_loss_clip": 1.05476868,
      "balance_loss_mlp": 1.01446486,
      "epoch": 0.40666145614140564,
      "flos": 27344252085120.0,
      "grad_norm": 2.46550835983336,
      "language_loss": 0.71469152,
      "learning_rate": 2.6875364044954453e-06,
      "loss": 0.73735499,
      "num_input_tokens_seen": 73043880,
      "step": 3382,
      "time_per_iteration": 2.734229326248169
    },
    {
      "auxiliary_loss_clip": 0.01287938,
      "auxiliary_loss_mlp": 0.01028428,
      "balance_loss_clip": 1.04781783,
      "balance_loss_mlp": 1.0209893,
      "epoch": 0.40678169903204475,
      "flos": 26176724415360.0,
      "grad_norm": 1.6450951256662354,
      "language_loss": 0.82492268,
      "learning_rate": 2.6868048555051185e-06,
      "loss": 0.8480863,
      "num_input_tokens_seen": 73065410,
      "step": 3383,
      "time_per_iteration": 2.678741455078125
    },
    {
      "auxiliary_loss_clip": 0.01297468,
      "auxiliary_loss_mlp": 0.01032404,
      "balance_loss_clip": 1.05079639,
      "balance_loss_mlp": 1.02437234,
      "epoch": 0.4069019419226838,
      "flos": 28622420622720.0,
      "grad_norm": 3.3359341886798597,
      "language_loss": 0.85958177,
      "learning_rate": 2.686073202327041e-06,
      "loss": 0.88288051,
      "num_input_tokens_seen": 73084410,
      "step": 3384,
      "time_per_iteration": 2.7544875144958496
    },
    {
      "auxiliary_loss_clip": 0.01283012,
      "auxiliary_loss_mlp": 0.01024419,
      "balance_loss_clip": 1.0473032,
      "balance_loss_mlp": 1.0171113,
      "epoch": 0.4070221848133229,
      "flos": 25229006023680.0,
      "grad_norm": 2.4043769439025886,
      "language_loss": 0.73507953,
      "learning_rate": 2.6853414450722043e-06,
      "loss": 0.75815386,
      "num_input_tokens_seen": 73104075,
      "step": 3385,
      "time_per_iteration": 2.723113775253296
    },
    {
      "auxiliary_loss_clip": 0.01236028,
      "auxiliary_loss_mlp": 0.01023336,
      "balance_loss_clip": 1.05233157,
      "balance_loss_mlp": 1.01594543,
      "epoch": 0.40714242770396203,
      "flos": 18405224709120.0,
      "grad_norm": 2.2481775859278024,
      "language_loss": 0.85182691,
      "learning_rate": 2.684609583851616e-06,
      "loss": 0.87442052,
      "num_input_tokens_seen": 73122250,
      "step": 3386,
      "time_per_iteration": 3.5689849853515625
    },
    {
      "auxiliary_loss_clip": 0.01383572,
      "auxiliary_loss_mlp": 0.01025945,
      "balance_loss_clip": 1.04825401,
      "balance_loss_mlp": 1.01845253,
      "epoch": 0.4072626705946011,
      "flos": 30228920403840.0,
      "grad_norm": 2.3011722095520666,
      "language_loss": 0.80828285,
      "learning_rate": 2.683877618776297e-06,
      "loss": 0.83237803,
      "num_input_tokens_seen": 73144505,
      "step": 3387,
      "time_per_iteration": 2.855247974395752
    },
    {
      "auxiliary_loss_clip": 0.01291084,
      "auxiliary_loss_mlp": 0.01029219,
      "balance_loss_clip": 1.04971421,
      "balance_loss_mlp": 1.02151501,
      "epoch": 0.4073829134852402,
      "flos": 21834549930240.0,
      "grad_norm": 2.7817630467354624,
      "language_loss": 0.73974144,
      "learning_rate": 2.6831455499572876e-06,
      "loss": 0.7629444,
      "num_input_tokens_seen": 73162440,
      "step": 3388,
      "time_per_iteration": 2.761523723602295
    },
    {
      "auxiliary_loss_clip": 0.01191434,
      "auxiliary_loss_mlp": 0.01028519,
      "balance_loss_clip": 1.05478692,
      "balance_loss_mlp": 1.02098441,
      "epoch": 0.40750315637587925,
      "flos": 25260211964160.0,
      "grad_norm": 2.931507565750685,
      "language_loss": 0.77997255,
      "learning_rate": 2.682413377505641e-06,
      "loss": 0.80217206,
      "num_input_tokens_seen": 73181245,
      "step": 3389,
      "time_per_iteration": 2.6485581398010254
    },
    {
      "auxiliary_loss_clip": 0.01237629,
      "auxiliary_loss_mlp": 0.01027588,
      "balance_loss_clip": 1.05131698,
      "balance_loss_mlp": 1.0207572,
      "epoch": 0.40762339926651836,
      "flos": 19712767593600.0,
      "grad_norm": 2.106073116027979,
      "language_loss": 0.76870501,
      "learning_rate": 2.6816811015324284e-06,
      "loss": 0.79135716,
      "num_input_tokens_seen": 73199295,
      "step": 3390,
      "time_per_iteration": 2.6342320442199707
    },
    {
      "auxiliary_loss_clip": 0.0108283,
      "auxiliary_loss_mlp": 0.01001768,
      "balance_loss_clip": 1.0183692,
      "balance_loss_mlp": 1.00067711,
      "epoch": 0.40774364215715747,
      "flos": 71449307314560.0,
      "grad_norm": 0.7232778722611614,
      "language_loss": 0.56665003,
      "learning_rate": 2.6809487221487343e-06,
      "loss": 0.58749604,
      "num_input_tokens_seen": 73258780,
      "step": 3391,
      "time_per_iteration": 4.140889406204224
    },
    {
      "auxiliary_loss_clip": 0.01236449,
      "auxiliary_loss_mlp": 0.01027684,
      "balance_loss_clip": 1.0509789,
      "balance_loss_mlp": 1.02010846,
      "epoch": 0.4078638850477965,
      "flos": 15084134144640.0,
      "grad_norm": 2.8674100224474133,
      "language_loss": 0.81620264,
      "learning_rate": 2.6802162394656605e-06,
      "loss": 0.83884406,
      "num_input_tokens_seen": 73275490,
      "step": 3392,
      "time_per_iteration": 2.679086446762085
    },
    {
      "auxiliary_loss_clip": 0.01283614,
      "auxiliary_loss_mlp": 0.01025484,
      "balance_loss_clip": 1.04862738,
      "balance_loss_mlp": 1.01823568,
      "epoch": 0.40798412793843564,
      "flos": 23842890138240.0,
      "grad_norm": 1.8174719761581786,
      "language_loss": 0.720411,
      "learning_rate": 2.679483653594324e-06,
      "loss": 0.74350202,
      "num_input_tokens_seen": 73297260,
      "step": 3393,
      "time_per_iteration": 2.670738458633423
    },
    {
      "auxiliary_loss_clip": 0.012394,
      "auxiliary_loss_mlp": 0.01022608,
      "balance_loss_clip": 1.05057955,
      "balance_loss_mlp": 1.01555634,
      "epoch": 0.40810437082907475,
      "flos": 21065774117760.0,
      "grad_norm": 2.5515420024287825,
      "language_loss": 0.76714998,
      "learning_rate": 2.678750964645857e-06,
      "loss": 0.78977007,
      "num_input_tokens_seen": 73316340,
      "step": 3394,
      "time_per_iteration": 2.6913084983825684
    },
    {
      "auxiliary_loss_clip": 0.01247226,
      "auxiliary_loss_mlp": 0.01031985,
      "balance_loss_clip": 1.05836761,
      "balance_loss_mlp": 1.02427208,
      "epoch": 0.4082246137197138,
      "flos": 11321377948800.0,
      "grad_norm": 2.746921485815622,
      "language_loss": 0.83854157,
      "learning_rate": 2.6780181727314094e-06,
      "loss": 0.86133367,
      "num_input_tokens_seen": 73331245,
      "step": 3395,
      "time_per_iteration": 2.5898985862731934
    },
    {
      "auxiliary_loss_clip": 0.01339933,
      "auxiliary_loss_mlp": 0.02564638,
      "balance_loss_clip": 1.04974008,
      "balance_loss_mlp": 1.00016177,
      "epoch": 0.4083448566103529,
      "flos": 19062569554560.0,
      "grad_norm": 1.935232502812014,
      "language_loss": 0.7800256,
      "learning_rate": 2.6772852779621435e-06,
      "loss": 0.81907129,
      "num_input_tokens_seen": 73349105,
      "step": 3396,
      "time_per_iteration": 3.7508294582366943
    },
    {
      "auxiliary_loss_clip": 0.01245087,
      "auxiliary_loss_mlp": 0.02562749,
      "balance_loss_clip": 1.05972219,
      "balance_loss_mlp": 1.00016069,
      "epoch": 0.408465099500992,
      "flos": 23550254035200.0,
      "grad_norm": 1.9788830862283877,
      "language_loss": 0.86872286,
      "learning_rate": 2.676552280449239e-06,
      "loss": 0.90680122,
      "num_input_tokens_seen": 73368990,
      "step": 3397,
      "time_per_iteration": 3.6710002422332764
    },
    {
      "auxiliary_loss_clip": 0.01233857,
      "auxiliary_loss_mlp": 0.01028657,
      "balance_loss_clip": 1.05018198,
      "balance_loss_mlp": 1.02114654,
      "epoch": 0.4085853423916311,
      "flos": 12750012558720.0,
      "grad_norm": 2.6786452360537254,
      "language_loss": 0.76243079,
      "learning_rate": 2.6758191803038917e-06,
      "loss": 0.78505599,
      "num_input_tokens_seen": 73387485,
      "step": 3398,
      "time_per_iteration": 2.663933515548706
    },
    {
      "auxiliary_loss_clip": 0.01431709,
      "auxiliary_loss_mlp": 0.01031739,
      "balance_loss_clip": 1.04695463,
      "balance_loss_mlp": 1.02371645,
      "epoch": 0.4087055852822702,
      "flos": 24353072962560.0,
      "grad_norm": 1.8276939095655227,
      "language_loss": 0.82641292,
      "learning_rate": 2.6750859776373125e-06,
      "loss": 0.8510474,
      "num_input_tokens_seen": 73406940,
      "step": 3399,
      "time_per_iteration": 2.811918258666992
    },
    {
      "auxiliary_loss_clip": 0.0135641,
      "auxiliary_loss_mlp": 0.01002041,
      "balance_loss_clip": 1.01945746,
      "balance_loss_mlp": 1.00090206,
      "epoch": 0.4088258281729093,
      "flos": 66387950720640.0,
      "grad_norm": 0.7703250316028225,
      "language_loss": 0.60370618,
      "learning_rate": 2.674352672560727e-06,
      "loss": 0.62729067,
      "num_input_tokens_seen": 73468385,
      "step": 3400,
      "time_per_iteration": 3.43412709236145
    },
    {
      "auxiliary_loss_clip": 0.01242519,
      "auxiliary_loss_mlp": 0.01031876,
      "balance_loss_clip": 1.0482899,
      "balance_loss_mlp": 1.02457428,
      "epoch": 0.40894607106354836,
      "flos": 20449260057600.0,
      "grad_norm": 1.7174132667274329,
      "language_loss": 0.77247441,
      "learning_rate": 2.673619265185377e-06,
      "loss": 0.79521847,
      "num_input_tokens_seen": 73488225,
      "step": 3401,
      "time_per_iteration": 2.9197418689727783
    },
    {
      "auxiliary_loss_clip": 0.01242157,
      "auxiliary_loss_mlp": 0.01025486,
      "balance_loss_clip": 1.05114114,
      "balance_loss_mlp": 1.01817584,
      "epoch": 0.40906631395418747,
      "flos": 27053627143680.0,
      "grad_norm": 2.6396740919602584,
      "language_loss": 0.78435791,
      "learning_rate": 2.672885755622521e-06,
      "loss": 0.80703425,
      "num_input_tokens_seen": 73510640,
      "step": 3402,
      "time_per_iteration": 2.7261698246002197
    },
    {
      "auxiliary_loss_clip": 0.01379716,
      "auxiliary_loss_mlp": 0.01022121,
      "balance_loss_clip": 1.04465413,
      "balance_loss_mlp": 1.0150485,
      "epoch": 0.4091865568448266,
      "flos": 25484151306240.0,
      "grad_norm": 8.592940016407335,
      "language_loss": 0.70203084,
      "learning_rate": 2.67215214398343e-06,
      "loss": 0.72604918,
      "num_input_tokens_seen": 73530655,
      "step": 3403,
      "time_per_iteration": 2.765868663787842
    },
    {
      "auxiliary_loss_clip": 0.01387296,
      "auxiliary_loss_mlp": 0.01030014,
      "balance_loss_clip": 1.04481435,
      "balance_loss_mlp": 1.02243185,
      "epoch": 0.40930679973546563,
      "flos": 28657864368000.0,
      "grad_norm": 2.666016552108034,
      "language_loss": 0.7809478,
      "learning_rate": 2.671418430379393e-06,
      "loss": 0.80512089,
      "num_input_tokens_seen": 73549340,
      "step": 3404,
      "time_per_iteration": 2.8611412048339844
    },
    {
      "auxiliary_loss_clip": 0.0118844,
      "auxiliary_loss_mlp": 0.0102266,
      "balance_loss_clip": 1.05174673,
      "balance_loss_mlp": 1.01515007,
      "epoch": 0.40942704262610474,
      "flos": 20886292834560.0,
      "grad_norm": 2.914430393149479,
      "language_loss": 0.83528447,
      "learning_rate": 2.670684614921715e-06,
      "loss": 0.85739541,
      "num_input_tokens_seen": 73568315,
      "step": 3405,
      "time_per_iteration": 2.612776517868042
    },
    {
      "auxiliary_loss_clip": 0.01292322,
      "auxiliary_loss_mlp": 0.01026303,
      "balance_loss_clip": 1.04967308,
      "balance_loss_mlp": 1.01864374,
      "epoch": 0.4095472855167438,
      "flos": 21618080616960.0,
      "grad_norm": 2.312974371386244,
      "language_loss": 0.69203871,
      "learning_rate": 2.6699506977217128e-06,
      "loss": 0.71522498,
      "num_input_tokens_seen": 73588490,
      "step": 3406,
      "time_per_iteration": 2.77194881439209
    },
    {
      "auxiliary_loss_clip": 0.01236599,
      "auxiliary_loss_mlp": 0.0102612,
      "balance_loss_clip": 1.05367649,
      "balance_loss_mlp": 1.01918817,
      "epoch": 0.4096675284073829,
      "flos": 27926112499200.0,
      "grad_norm": 3.107727204260342,
      "language_loss": 0.70348537,
      "learning_rate": 2.6692166788907233e-06,
      "loss": 0.72611254,
      "num_input_tokens_seen": 73608685,
      "step": 3407,
      "time_per_iteration": 3.036036252975464
    },
    {
      "auxiliary_loss_clip": 0.01291122,
      "auxiliary_loss_mlp": 0.01035465,
      "balance_loss_clip": 1.05040908,
      "balance_loss_mlp": 1.02787733,
      "epoch": 0.409787771298022,
      "flos": 19206607092480.0,
      "grad_norm": 2.0828762609037326,
      "language_loss": 0.77089024,
      "learning_rate": 2.6684825585400957e-06,
      "loss": 0.79415607,
      "num_input_tokens_seen": 73627630,
      "step": 3408,
      "time_per_iteration": 2.7924749851226807
    },
    {
      "auxiliary_loss_clip": 0.01183495,
      "auxiliary_loss_mlp": 0.01002401,
      "balance_loss_clip": 1.01866484,
      "balance_loss_mlp": 1.00131595,
      "epoch": 0.4099080141886611,
      "flos": 59269234832640.0,
      "grad_norm": 0.8056190039259631,
      "language_loss": 0.65086895,
      "learning_rate": 2.6677483367811947e-06,
      "loss": 0.672728,
      "num_input_tokens_seen": 73687670,
      "step": 3409,
      "time_per_iteration": 3.4256017208099365
    },
    {
      "auxiliary_loss_clip": 0.01241701,
      "auxiliary_loss_mlp": 0.01024396,
      "balance_loss_clip": 1.05159616,
      "balance_loss_mlp": 1.01668668,
      "epoch": 0.4100282570793002,
      "flos": 21906443001600.0,
      "grad_norm": 1.8344471732317216,
      "language_loss": 0.75822699,
      "learning_rate": 2.6670140137254028e-06,
      "loss": 0.78088796,
      "num_input_tokens_seen": 73707145,
      "step": 3410,
      "time_per_iteration": 2.741098403930664
    },
    {
      "auxiliary_loss_clip": 0.01383336,
      "auxiliary_loss_mlp": 0.01030093,
      "balance_loss_clip": 1.04530144,
      "balance_loss_mlp": 1.02259493,
      "epoch": 0.4101484999699393,
      "flos": 18551596631040.0,
      "grad_norm": 3.2703778335006413,
      "language_loss": 0.898633,
      "learning_rate": 2.666279589484115e-06,
      "loss": 0.92276734,
      "num_input_tokens_seen": 73725045,
      "step": 3411,
      "time_per_iteration": 2.812255620956421
    },
    {
      "auxiliary_loss_clip": 0.01389256,
      "auxiliary_loss_mlp": 0.01026674,
      "balance_loss_clip": 1.04498994,
      "balance_loss_mlp": 1.01936018,
      "epoch": 0.41026874286057835,
      "flos": 19094529680640.0,
      "grad_norm": 1.9749993411546891,
      "language_loss": 0.81185961,
      "learning_rate": 2.6655450641687435e-06,
      "loss": 0.83601886,
      "num_input_tokens_seen": 73742610,
      "step": 3412,
      "time_per_iteration": 2.8383944034576416
    },
    {
      "auxiliary_loss_clip": 0.01188366,
      "auxiliary_loss_mlp": 0.01024924,
      "balance_loss_clip": 1.05402684,
      "balance_loss_mlp": 1.01760447,
      "epoch": 0.41038898575121746,
      "flos": 31209568588800.0,
      "grad_norm": 1.9658199344511,
      "language_loss": 0.69017488,
      "learning_rate": 2.664810437890715e-06,
      "loss": 0.71230775,
      "num_input_tokens_seen": 73764280,
      "step": 3413,
      "time_per_iteration": 4.144192218780518
    },
    {
      "auxiliary_loss_clip": 0.01423754,
      "auxiliary_loss_mlp": 0.01025421,
      "balance_loss_clip": 1.05111098,
      "balance_loss_mlp": 1.01808977,
      "epoch": 0.41050922864185657,
      "flos": 14355865895040.0,
      "grad_norm": 1.9018670832741602,
      "language_loss": 0.79471815,
      "learning_rate": 2.6640757107614714e-06,
      "loss": 0.81920993,
      "num_input_tokens_seen": 73782375,
      "step": 3414,
      "time_per_iteration": 2.881629228591919
    },
    {
      "auxiliary_loss_clip": 0.01342642,
      "auxiliary_loss_mlp": 0.01029634,
      "balance_loss_clip": 1.05431986,
      "balance_loss_mlp": 1.02189386,
      "epoch": 0.4106294715324956,
      "flos": 30956290813440.0,
      "grad_norm": 2.7070817046834827,
      "language_loss": 0.69281614,
      "learning_rate": 2.6633408828924697e-06,
      "loss": 0.71653885,
      "num_input_tokens_seen": 73801240,
      "step": 3415,
      "time_per_iteration": 2.853459358215332
    },
    {
      "auxiliary_loss_clip": 0.01344783,
      "auxiliary_loss_mlp": 0.01026667,
      "balance_loss_clip": 1.05038309,
      "balance_loss_mlp": 1.01972282,
      "epoch": 0.41074971442313474,
      "flos": 24457321209600.0,
      "grad_norm": 1.579221958658815,
      "language_loss": 0.69871843,
      "learning_rate": 2.662605954395185e-06,
      "loss": 0.72243285,
      "num_input_tokens_seen": 73821200,
      "step": 3416,
      "time_per_iteration": 2.899362325668335
    },
    {
      "auxiliary_loss_clip": 0.01240445,
      "auxiliary_loss_mlp": 0.01027903,
      "balance_loss_clip": 1.04999065,
      "balance_loss_mlp": 1.02108157,
      "epoch": 0.41086995731377385,
      "flos": 21542991235200.0,
      "grad_norm": 2.0480351944001582,
      "language_loss": 0.83772099,
      "learning_rate": 2.6618709253811027e-06,
      "loss": 0.86040449,
      "num_input_tokens_seen": 73840655,
      "step": 3417,
      "time_per_iteration": 3.673060894012451
    },
    {
      "auxiliary_loss_clip": 0.01184151,
      "auxiliary_loss_mlp": 0.01024392,
      "balance_loss_clip": 1.05265832,
      "balance_loss_mlp": 1.01805282,
      "epoch": 0.4109902002044129,
      "flos": 20702753314560.0,
      "grad_norm": 2.0385630439023705,
      "language_loss": 0.87677896,
      "learning_rate": 2.6611357959617277e-06,
      "loss": 0.89886439,
      "num_input_tokens_seen": 73860275,
      "step": 3418,
      "time_per_iteration": 2.703909397125244
    },
    {
      "auxiliary_loss_clip": 0.01329736,
      "auxiliary_loss_mlp": 0.01022778,
      "balance_loss_clip": 1.04660356,
      "balance_loss_mlp": 1.01548862,
      "epoch": 0.411110443095052,
      "flos": 18179992477440.0,
      "grad_norm": 2.174090862382698,
      "language_loss": 0.91230798,
      "learning_rate": 2.660400566248578e-06,
      "loss": 0.93583322,
      "num_input_tokens_seen": 73878400,
      "step": 3419,
      "time_per_iteration": 2.7969768047332764
    },
    {
      "auxiliary_loss_clip": 0.01337312,
      "auxiliary_loss_mlp": 0.01022051,
      "balance_loss_clip": 1.04858851,
      "balance_loss_mlp": 1.01427567,
      "epoch": 0.41123068598569107,
      "flos": 14575244209920.0,
      "grad_norm": 2.910423674675818,
      "language_loss": 0.67426181,
      "learning_rate": 2.6596652363531876e-06,
      "loss": 0.69785535,
      "num_input_tokens_seen": 73894275,
      "step": 3420,
      "time_per_iteration": 2.7155120372772217
    },
    {
      "auxiliary_loss_clip": 0.01187601,
      "auxiliary_loss_mlp": 0.01035496,
      "balance_loss_clip": 1.0543251,
      "balance_loss_mlp": 1.02829576,
      "epoch": 0.4113509288763302,
      "flos": 21177995184000.0,
      "grad_norm": 1.8916494772563757,
      "language_loss": 0.78286296,
      "learning_rate": 2.6589298063871055e-06,
      "loss": 0.80509394,
      "num_input_tokens_seen": 73914450,
      "step": 3421,
      "time_per_iteration": 2.6984000205993652
    },
    {
      "auxiliary_loss_clip": 0.01189152,
      "auxiliary_loss_mlp": 0.01030726,
      "balance_loss_clip": 1.05382705,
      "balance_loss_mlp": 1.02287602,
      "epoch": 0.4114711717669693,
      "flos": 18442212739200.0,
      "grad_norm": 2.0503536982815715,
      "language_loss": 0.7028572,
      "learning_rate": 2.658194276461895e-06,
      "loss": 0.72505593,
      "num_input_tokens_seen": 73932375,
      "step": 3422,
      "time_per_iteration": 3.558162212371826
    },
    {
      "auxiliary_loss_clip": 0.01291955,
      "auxiliary_loss_mlp": 0.01027787,
      "balance_loss_clip": 1.04692912,
      "balance_loss_mlp": 1.01949644,
      "epoch": 0.41159141465760835,
      "flos": 27233395735680.0,
      "grad_norm": 2.04710064849765,
      "language_loss": 0.66770971,
      "learning_rate": 2.6574586466891368e-06,
      "loss": 0.69090718,
      "num_input_tokens_seen": 73952850,
      "step": 3423,
      "time_per_iteration": 3.7355690002441406
    },
    {
      "auxiliary_loss_clip": 0.01288194,
      "auxiliary_loss_mlp": 0.02562237,
      "balance_loss_clip": 1.05085421,
      "balance_loss_mlp": 1.00008619,
      "epoch": 0.41171165754824746,
      "flos": 20006876154240.0,
      "grad_norm": 3.5174989301280957,
      "language_loss": 0.64636385,
      "learning_rate": 2.6567229171804247e-06,
      "loss": 0.68486822,
      "num_input_tokens_seen": 73970735,
      "step": 3424,
      "time_per_iteration": 2.793581247329712
    },
    {
      "auxiliary_loss_clip": 0.01344347,
      "auxiliary_loss_mlp": 0.01028255,
      "balance_loss_clip": 1.04875171,
      "balance_loss_mlp": 1.02030945,
      "epoch": 0.41183190043888657,
      "flos": 18004318035840.0,
      "grad_norm": 3.907731571582285,
      "language_loss": 0.87689269,
      "learning_rate": 2.655987088047368e-06,
      "loss": 0.90061867,
      "num_input_tokens_seen": 73989080,
      "step": 3425,
      "time_per_iteration": 2.7631752490997314
    },
    {
      "auxiliary_loss_clip": 0.01289289,
      "auxiliary_loss_mlp": 0.01030542,
      "balance_loss_clip": 1.04984164,
      "balance_loss_mlp": 1.02273393,
      "epoch": 0.4119521433295256,
      "flos": 27163370171520.0,
      "grad_norm": 8.407624754882306,
      "language_loss": 0.78635609,
      "learning_rate": 2.6552511594015912e-06,
      "loss": 0.80955434,
      "num_input_tokens_seen": 74009470,
      "step": 3426,
      "time_per_iteration": 2.792881727218628
    },
    {
      "auxiliary_loss_clip": 0.01289618,
      "auxiliary_loss_mlp": 0.01030034,
      "balance_loss_clip": 1.04701424,
      "balance_loss_mlp": 1.02196991,
      "epoch": 0.41207238622016473,
      "flos": 15122020014720.0,
      "grad_norm": 4.187623977959082,
      "language_loss": 0.84724671,
      "learning_rate": 2.654515131354735e-06,
      "loss": 0.87044322,
      "num_input_tokens_seen": 74027735,
      "step": 3427,
      "time_per_iteration": 2.701958656311035
    },
    {
      "auxiliary_loss_clip": 0.01337083,
      "auxiliary_loss_mlp": 0.01025392,
      "balance_loss_clip": 1.05029607,
      "balance_loss_mlp": 1.01819181,
      "epoch": 0.41219262911080384,
      "flos": 27052872958080.0,
      "grad_norm": 2.2264868434221703,
      "language_loss": 0.8512516,
      "learning_rate": 2.653779004018453e-06,
      "loss": 0.87487632,
      "num_input_tokens_seen": 74048300,
      "step": 3428,
      "time_per_iteration": 2.770838499069214
    },
    {
      "auxiliary_loss_clip": 0.01285869,
      "auxiliary_loss_mlp": 0.0102458,
      "balance_loss_clip": 1.04987931,
      "balance_loss_mlp": 1.01748729,
      "epoch": 0.4123128720014429,
      "flos": 24686360282880.0,
      "grad_norm": 2.056065244570253,
      "language_loss": 0.82363629,
      "learning_rate": 2.653042777504417e-06,
      "loss": 0.84674078,
      "num_input_tokens_seen": 74070890,
      "step": 3429,
      "time_per_iteration": 2.8045990467071533
    },
    {
      "auxiliary_loss_clip": 0.01198208,
      "auxiliary_loss_mlp": 0.01023749,
      "balance_loss_clip": 1.05017483,
      "balance_loss_mlp": 1.01559508,
      "epoch": 0.412433114892082,
      "flos": 26244774731520.0,
      "grad_norm": 2.0731037432394404,
      "language_loss": 0.79862177,
      "learning_rate": 2.6523064519243105e-06,
      "loss": 0.82084137,
      "num_input_tokens_seen": 74090460,
      "step": 3430,
      "time_per_iteration": 2.760378360748291
    },
    {
      "auxiliary_loss_clip": 0.01241199,
      "auxiliary_loss_mlp": 0.01029383,
      "balance_loss_clip": 1.05485082,
      "balance_loss_mlp": 1.02154779,
      "epoch": 0.4125533577827211,
      "flos": 21361031913600.0,
      "grad_norm": 2.6877310455088943,
      "language_loss": 0.78998733,
      "learning_rate": 2.6515700273898333e-06,
      "loss": 0.81269312,
      "num_input_tokens_seen": 74108335,
      "step": 3431,
      "time_per_iteration": 2.740703821182251
    },
    {
      "auxiliary_loss_clip": 0.01290396,
      "auxiliary_loss_mlp": 0.01031893,
      "balance_loss_clip": 1.05524659,
      "balance_loss_mlp": 1.02404356,
      "epoch": 0.4126736006733602,
      "flos": 26067556005120.0,
      "grad_norm": 3.1487213183405847,
      "language_loss": 0.69150746,
      "learning_rate": 2.6508335040127018e-06,
      "loss": 0.71473038,
      "num_input_tokens_seen": 74128030,
      "step": 3432,
      "time_per_iteration": 2.7816851139068604
    },
    {
      "auxiliary_loss_clip": 0.01244467,
      "auxiliary_loss_mlp": 0.01025317,
      "balance_loss_clip": 1.05397391,
      "balance_loss_mlp": 1.01857901,
      "epoch": 0.4127938435639993,
      "flos": 25666146541440.0,
      "grad_norm": 1.522179845436325,
      "language_loss": 0.76766682,
      "learning_rate": 2.6500968819046446e-06,
      "loss": 0.79036474,
      "num_input_tokens_seen": 74148330,
      "step": 3433,
      "time_per_iteration": 2.8021676540374756
    },
    {
      "auxiliary_loss_clip": 0.01329626,
      "auxiliary_loss_mlp": 0.01026428,
      "balance_loss_clip": 1.0444839,
      "balance_loss_mlp": 1.01916218,
      "epoch": 0.4129140864546384,
      "flos": 17995914253440.0,
      "grad_norm": 2.429788153725156,
      "language_loss": 0.59093642,
      "learning_rate": 2.649360161177408e-06,
      "loss": 0.61449695,
      "num_input_tokens_seen": 74163390,
      "step": 3434,
      "time_per_iteration": 2.7132630348205566
    },
    {
      "auxiliary_loss_clip": 0.01246623,
      "auxiliary_loss_mlp": 0.01026567,
      "balance_loss_clip": 1.0535388,
      "balance_loss_mlp": 1.01837087,
      "epoch": 0.41303432934527745,
      "flos": 23732895715200.0,
      "grad_norm": 2.666224697727589,
      "language_loss": 0.73453641,
      "learning_rate": 2.6486233419427504e-06,
      "loss": 0.75726825,
      "num_input_tokens_seen": 74183205,
      "step": 3435,
      "time_per_iteration": 2.7287209033966064
    },
    {
      "auxiliary_loss_clip": 0.01334895,
      "auxiliary_loss_mlp": 0.01027155,
      "balance_loss_clip": 1.05178642,
      "balance_loss_mlp": 1.01918268,
      "epoch": 0.41315457223591656,
      "flos": 19755286318080.0,
      "grad_norm": 2.5994008615321924,
      "language_loss": 0.75279236,
      "learning_rate": 2.6478864243124484e-06,
      "loss": 0.77641284,
      "num_input_tokens_seen": 74202870,
      "step": 3436,
      "time_per_iteration": 2.726447582244873
    },
    {
      "auxiliary_loss_clip": 0.01240117,
      "auxiliary_loss_mlp": 0.01022721,
      "balance_loss_clip": 1.05079842,
      "balance_loss_mlp": 1.0145607,
      "epoch": 0.4132748151265556,
      "flos": 20923316778240.0,
      "grad_norm": 2.106694929150256,
      "language_loss": 0.8519429,
      "learning_rate": 2.6471494083982903e-06,
      "loss": 0.87457126,
      "num_input_tokens_seen": 74222255,
      "step": 3437,
      "time_per_iteration": 2.696732759475708
    },
    {
      "auxiliary_loss_clip": 0.01340516,
      "auxiliary_loss_mlp": 0.01027242,
      "balance_loss_clip": 1.04787827,
      "balance_loss_mlp": 1.01936817,
      "epoch": 0.4133950580171947,
      "flos": 32232520016640.0,
      "grad_norm": 2.3383256361539324,
      "language_loss": 0.74661726,
      "learning_rate": 2.6464122943120818e-06,
      "loss": 0.77029479,
      "num_input_tokens_seen": 74242480,
      "step": 3438,
      "time_per_iteration": 3.9563517570495605
    },
    {
      "auxiliary_loss_clip": 0.01335185,
      "auxiliary_loss_mlp": 0.01028722,
      "balance_loss_clip": 1.05068946,
      "balance_loss_mlp": 1.0211103,
      "epoch": 0.41351530090783384,
      "flos": 23292487059840.0,
      "grad_norm": 3.954108249018504,
      "language_loss": 0.81703019,
      "learning_rate": 2.645675082165642e-06,
      "loss": 0.84066927,
      "num_input_tokens_seen": 74258690,
      "step": 3439,
      "time_per_iteration": 2.7422218322753906
    },
    {
      "auxiliary_loss_clip": 0.01288689,
      "auxiliary_loss_mlp": 0.01029532,
      "balance_loss_clip": 1.05482066,
      "balance_loss_mlp": 1.02163696,
      "epoch": 0.4136355437984729,
      "flos": 25593571111680.0,
      "grad_norm": 2.525371250857849,
      "language_loss": 0.75165063,
      "learning_rate": 2.644937772070806e-06,
      "loss": 0.7748329,
      "num_input_tokens_seen": 74277135,
      "step": 3440,
      "time_per_iteration": 2.77412748336792
    },
    {
      "auxiliary_loss_clip": 0.01191515,
      "auxiliary_loss_mlp": 0.01027287,
      "balance_loss_clip": 1.0548749,
      "balance_loss_mlp": 1.02002692,
      "epoch": 0.413755786689112,
      "flos": 19828615933440.0,
      "grad_norm": 4.227914518325167,
      "language_loss": 0.83190191,
      "learning_rate": 2.6442003641394225e-06,
      "loss": 0.85408992,
      "num_input_tokens_seen": 74294730,
      "step": 3441,
      "time_per_iteration": 2.604081869125366
    },
    {
      "auxiliary_loss_clip": 0.01288316,
      "auxiliary_loss_mlp": 0.01024531,
      "balance_loss_clip": 1.04704285,
      "balance_loss_mlp": 1.01710463,
      "epoch": 0.4138760295797511,
      "flos": 26870446759680.0,
      "grad_norm": 1.735743921684924,
      "language_loss": 0.84071338,
      "learning_rate": 2.643462858483356e-06,
      "loss": 0.86384189,
      "num_input_tokens_seen": 74315015,
      "step": 3442,
      "time_per_iteration": 2.7406692504882812
    },
    {
      "auxiliary_loss_clip": 0.01394011,
      "auxiliary_loss_mlp": 0.01026911,
      "balance_loss_clip": 1.05126333,
      "balance_loss_mlp": 1.01931751,
      "epoch": 0.41399627247039017,
      "flos": 16399254798720.0,
      "grad_norm": 1.9783840137364428,
      "language_loss": 0.7296164,
      "learning_rate": 2.6427252552144856e-06,
      "loss": 0.7538256,
      "num_input_tokens_seen": 74333665,
      "step": 3443,
      "time_per_iteration": 3.7323341369628906
    },
    {
      "auxiliary_loss_clip": 0.0118824,
      "auxiliary_loss_mlp": 0.01028534,
      "balance_loss_clip": 1.05247414,
      "balance_loss_mlp": 1.02098775,
      "epoch": 0.4141165153610293,
      "flos": 22930220442240.0,
      "grad_norm": 4.019175047902539,
      "language_loss": 0.75255948,
      "learning_rate": 2.6419875544447044e-06,
      "loss": 0.77472723,
      "num_input_tokens_seen": 74355065,
      "step": 3444,
      "time_per_iteration": 2.677738904953003
    },
    {
      "auxiliary_loss_clip": 0.01190873,
      "auxiliary_loss_mlp": 0.01027539,
      "balance_loss_clip": 1.05345976,
      "balance_loss_mlp": 1.01962376,
      "epoch": 0.4142367582516684,
      "flos": 25192556697600.0,
      "grad_norm": 4.548010104554467,
      "language_loss": 0.71635532,
      "learning_rate": 2.6412497562859218e-06,
      "loss": 0.73853946,
      "num_input_tokens_seen": 74376345,
      "step": 3445,
      "time_per_iteration": 2.71471905708313
    },
    {
      "auxiliary_loss_clip": 0.01246065,
      "auxiliary_loss_mlp": 0.01025024,
      "balance_loss_clip": 1.05312979,
      "balance_loss_mlp": 1.01751363,
      "epoch": 0.41435700114230745,
      "flos": 21690476478720.0,
      "grad_norm": 2.4943536854922197,
      "language_loss": 0.76508611,
      "learning_rate": 2.6405118608500617e-06,
      "loss": 0.78779703,
      "num_input_tokens_seen": 74395170,
      "step": 3446,
      "time_per_iteration": 2.6716742515563965
    },
    {
      "auxiliary_loss_clip": 0.01329587,
      "auxiliary_loss_mlp": 0.0102971,
      "balance_loss_clip": 1.05298138,
      "balance_loss_mlp": 1.02241993,
      "epoch": 0.41447724403294656,
      "flos": 25995160143360.0,
      "grad_norm": 2.106277937127679,
      "language_loss": 0.81415582,
      "learning_rate": 2.6397738682490613e-06,
      "loss": 0.83774871,
      "num_input_tokens_seen": 74416070,
      "step": 3447,
      "time_per_iteration": 3.639651298522949
    },
    {
      "auxiliary_loss_clip": 0.01191194,
      "auxiliary_loss_mlp": 0.01027677,
      "balance_loss_clip": 1.05397439,
      "balance_loss_mlp": 1.01974988,
      "epoch": 0.41459748692358567,
      "flos": 18259678800000.0,
      "grad_norm": 1.822489201483627,
      "language_loss": 0.75232446,
      "learning_rate": 2.6390357785948734e-06,
      "loss": 0.77451319,
      "num_input_tokens_seen": 74433185,
      "step": 3448,
      "time_per_iteration": 2.6237621307373047
    },
    {
      "auxiliary_loss_clip": 0.0123899,
      "auxiliary_loss_mlp": 0.01024683,
      "balance_loss_clip": 1.05391073,
      "balance_loss_mlp": 1.01671338,
      "epoch": 0.4147177298142247,
      "flos": 24168456034560.0,
      "grad_norm": 1.8716311464410627,
      "language_loss": 0.80341017,
      "learning_rate": 2.6382975919994667e-06,
      "loss": 0.82604688,
      "num_input_tokens_seen": 74453760,
      "step": 3449,
      "time_per_iteration": 3.6483283042907715
    },
    {
      "auxiliary_loss_clip": 0.01290623,
      "auxiliary_loss_mlp": 0.01028896,
      "balance_loss_clip": 1.05079269,
      "balance_loss_mlp": 1.02166557,
      "epoch": 0.41483797270486383,
      "flos": 20084659056000.0,
      "grad_norm": 2.025649773561189,
      "language_loss": 0.7311272,
      "learning_rate": 2.637559308574822e-06,
      "loss": 0.75432241,
      "num_input_tokens_seen": 74473505,
      "step": 3450,
      "time_per_iteration": 2.7601804733276367
    },
    {
      "auxiliary_loss_clip": 0.0119078,
      "auxiliary_loss_mlp": 0.01027261,
      "balance_loss_clip": 1.05383062,
      "balance_loss_mlp": 1.01978016,
      "epoch": 0.4149582155955029,
      "flos": 30081040110720.0,
      "grad_norm": 2.291885187387876,
      "language_loss": 0.7115587,
      "learning_rate": 2.6368209284329376e-06,
      "loss": 0.73373902,
      "num_input_tokens_seen": 74494135,
      "step": 3451,
      "time_per_iteration": 2.7165627479553223
    },
    {
      "auxiliary_loss_clip": 0.01240799,
      "auxiliary_loss_mlp": 0.01035212,
      "balance_loss_clip": 1.05159295,
      "balance_loss_mlp": 1.02721274,
      "epoch": 0.415078458486142,
      "flos": 16764394504320.0,
      "grad_norm": 2.159340070245855,
      "language_loss": 0.75457895,
      "learning_rate": 2.636082451685825e-06,
      "loss": 0.77733904,
      "num_input_tokens_seen": 74512335,
      "step": 3452,
      "time_per_iteration": 2.714768886566162
    },
    {
      "auxiliary_loss_clip": 0.0129165,
      "auxiliary_loss_mlp": 0.01037597,
      "balance_loss_clip": 1.05421591,
      "balance_loss_mlp": 1.02953267,
      "epoch": 0.4151987013767811,
      "flos": 26033692458240.0,
      "grad_norm": 4.418626119274522,
      "language_loss": 0.86444253,
      "learning_rate": 2.6353438784455094e-06,
      "loss": 0.88773501,
      "num_input_tokens_seen": 74535620,
      "step": 3453,
      "time_per_iteration": 2.750631332397461
    },
    {
      "auxiliary_loss_clip": 0.01288077,
      "auxiliary_loss_mlp": 0.01032798,
      "balance_loss_clip": 1.05043745,
      "balance_loss_mlp": 1.02506101,
      "epoch": 0.41531894426742016,
      "flos": 24608002763520.0,
      "grad_norm": 3.0326179560812716,
      "language_loss": 0.71790469,
      "learning_rate": 2.6346052088240326e-06,
      "loss": 0.74111342,
      "num_input_tokens_seen": 74555140,
      "step": 3454,
      "time_per_iteration": 2.7449517250061035
    },
    {
      "auxiliary_loss_clip": 0.01290428,
      "auxiliary_loss_mlp": 0.01026279,
      "balance_loss_clip": 1.05258536,
      "balance_loss_mlp": 1.0186317,
      "epoch": 0.4154391871580593,
      "flos": 14975791747200.0,
      "grad_norm": 2.1229418480673026,
      "language_loss": 0.77351755,
      "learning_rate": 2.63386644293345e-06,
      "loss": 0.79668462,
      "num_input_tokens_seen": 74571485,
      "step": 3455,
      "time_per_iteration": 2.736422061920166
    },
    {
      "auxiliary_loss_clip": 0.01340046,
      "auxiliary_loss_mlp": 0.01024508,
      "balance_loss_clip": 1.04702592,
      "balance_loss_mlp": 1.0170579,
      "epoch": 0.4155594300486984,
      "flos": 14647173194880.0,
      "grad_norm": 2.2451186699814785,
      "language_loss": 0.83104342,
      "learning_rate": 2.633127580885833e-06,
      "loss": 0.854689,
      "num_input_tokens_seen": 74585985,
      "step": 3456,
      "time_per_iteration": 2.6913840770721436
    },
    {
      "auxiliary_loss_clip": 0.01189626,
      "auxiliary_loss_mlp": 0.01026873,
      "balance_loss_clip": 1.05533004,
      "balance_loss_mlp": 1.01968729,
      "epoch": 0.41567967293933744,
      "flos": 29497276275840.0,
      "grad_norm": 8.270469014984116,
      "language_loss": 0.64535695,
      "learning_rate": 2.632388622793265e-06,
      "loss": 0.66752195,
      "num_input_tokens_seen": 74605140,
      "step": 3457,
      "time_per_iteration": 2.7149815559387207
    },
    {
      "auxiliary_loss_clip": 0.01242714,
      "auxiliary_loss_mlp": 0.01037383,
      "balance_loss_clip": 1.05354428,
      "balance_loss_mlp": 1.02997684,
      "epoch": 0.41579991582997655,
      "flos": 19238387650560.0,
      "grad_norm": 1.7708306661256419,
      "language_loss": 0.67720616,
      "learning_rate": 2.6316495687678457e-06,
      "loss": 0.7000072,
      "num_input_tokens_seen": 74623790,
      "step": 3458,
      "time_per_iteration": 2.654083013534546
    },
    {
      "auxiliary_loss_clip": 0.0138611,
      "auxiliary_loss_mlp": 0.01022203,
      "balance_loss_clip": 1.04687047,
      "balance_loss_mlp": 1.01471066,
      "epoch": 0.41592015872061566,
      "flos": 24462061804800.0,
      "grad_norm": 3.462554846085933,
      "language_loss": 0.76662314,
      "learning_rate": 2.6309104189216887e-06,
      "loss": 0.79070628,
      "num_input_tokens_seen": 74641355,
      "step": 3459,
      "time_per_iteration": 2.9080283641815186
    },
    {
      "auxiliary_loss_clip": 0.01329838,
      "auxiliary_loss_mlp": 0.02569634,
      "balance_loss_clip": 1.04704988,
      "balance_loss_mlp": 1.00015163,
      "epoch": 0.4160404016112547,
      "flos": 20775651966720.0,
      "grad_norm": 2.2308392140511857,
      "language_loss": 0.7474463,
      "learning_rate": 2.630171173366923e-06,
      "loss": 0.78644097,
      "num_input_tokens_seen": 74657155,
      "step": 3460,
      "time_per_iteration": 2.806871175765991
    },
    {
      "auxiliary_loss_clip": 0.01389093,
      "auxiliary_loss_mlp": 0.01027018,
      "balance_loss_clip": 1.04782808,
      "balance_loss_mlp": 1.01934052,
      "epoch": 0.41616064450189383,
      "flos": 13916462820480.0,
      "grad_norm": 3.7243125509293673,
      "language_loss": 0.73957539,
      "learning_rate": 2.629431832215691e-06,
      "loss": 0.76373649,
      "num_input_tokens_seen": 74671960,
      "step": 3461,
      "time_per_iteration": 2.834475517272949
    },
    {
      "auxiliary_loss_clip": 0.01284005,
      "auxiliary_loss_mlp": 0.01027782,
      "balance_loss_clip": 1.05241072,
      "balance_loss_mlp": 1.02055526,
      "epoch": 0.41628088739253294,
      "flos": 20010826650240.0,
      "grad_norm": 1.7626758313345239,
      "language_loss": 0.87154514,
      "learning_rate": 2.628692395580151e-06,
      "loss": 0.89466298,
      "num_input_tokens_seen": 74692050,
      "step": 3462,
      "time_per_iteration": 2.815366506576538
    },
    {
      "auxiliary_loss_clip": 0.01426457,
      "auxiliary_loss_mlp": 0.01032805,
      "balance_loss_clip": 1.04092336,
      "balance_loss_mlp": 1.02491903,
      "epoch": 0.416401130283172,
      "flos": 29168801377920.0,
      "grad_norm": 2.3203596239577577,
      "language_loss": 0.79634529,
      "learning_rate": 2.6279528635724747e-06,
      "loss": 0.82093793,
      "num_input_tokens_seen": 74712205,
      "step": 3463,
      "time_per_iteration": 2.888267755508423
    },
    {
      "auxiliary_loss_clip": 0.01240644,
      "auxiliary_loss_mlp": 0.0103276,
      "balance_loss_clip": 1.05039001,
      "balance_loss_mlp": 1.02428985,
      "epoch": 0.4165213731738111,
      "flos": 16246813478400.0,
      "grad_norm": 3.731732230332921,
      "language_loss": 0.78120089,
      "learning_rate": 2.627213236304848e-06,
      "loss": 0.80393487,
      "num_input_tokens_seen": 74729005,
      "step": 3464,
      "time_per_iteration": 3.7810938358306885
    },
    {
      "auxiliary_loss_clip": 0.01242335,
      "auxiliary_loss_mlp": 0.01025373,
      "balance_loss_clip": 1.05150509,
      "balance_loss_mlp": 1.01752901,
      "epoch": 0.4166416160644502,
      "flos": 33765438787200.0,
      "grad_norm": 1.9433567396067204,
      "language_loss": 0.70655692,
      "learning_rate": 2.626473513889472e-06,
      "loss": 0.72923398,
      "num_input_tokens_seen": 74751385,
      "step": 3465,
      "time_per_iteration": 2.698026180267334
    },
    {
      "auxiliary_loss_clip": 0.01236263,
      "auxiliary_loss_mlp": 0.01027241,
      "balance_loss_clip": 1.05233169,
      "balance_loss_mlp": 1.01983261,
      "epoch": 0.41676185895508927,
      "flos": 20917498775040.0,
      "grad_norm": 2.197467739205825,
      "language_loss": 0.8246485,
      "learning_rate": 2.625733696438562e-06,
      "loss": 0.84728354,
      "num_input_tokens_seen": 74768890,
      "step": 3466,
      "time_per_iteration": 2.690098285675049
    },
    {
      "auxiliary_loss_clip": 0.01288211,
      "auxiliary_loss_mlp": 0.01033754,
      "balance_loss_clip": 1.05062962,
      "balance_loss_mlp": 1.02613711,
      "epoch": 0.4168821018457284,
      "flos": 18406122549120.0,
      "grad_norm": 1.6923923185990672,
      "language_loss": 0.75099432,
      "learning_rate": 2.6249937840643476e-06,
      "loss": 0.77421391,
      "num_input_tokens_seen": 74787195,
      "step": 3467,
      "time_per_iteration": 2.658403158187866
    },
    {
      "auxiliary_loss_clip": 0.01195116,
      "auxiliary_loss_mlp": 0.02567838,
      "balance_loss_clip": 1.05813634,
      "balance_loss_mlp": 1.00021863,
      "epoch": 0.41700234473636744,
      "flos": 18698399516160.0,
      "grad_norm": 1.8901763269783447,
      "language_loss": 0.67046702,
      "learning_rate": 2.6242537768790733e-06,
      "loss": 0.7080965,
      "num_input_tokens_seen": 74806350,
      "step": 3468,
      "time_per_iteration": 3.5315098762512207
    },
    {
      "auxiliary_loss_clip": 0.01243615,
      "auxiliary_loss_mlp": 0.01032404,
      "balance_loss_clip": 1.05612171,
      "balance_loss_mlp": 1.02474809,
      "epoch": 0.41712258762700655,
      "flos": 31033283616000.0,
      "grad_norm": 5.097838478021265,
      "language_loss": 0.68654919,
      "learning_rate": 2.6235136749949975e-06,
      "loss": 0.70930934,
      "num_input_tokens_seen": 74829800,
      "step": 3469,
      "time_per_iteration": 2.7521822452545166
    },
    {
      "auxiliary_loss_clip": 0.01190097,
      "auxiliary_loss_mlp": 0.010241,
      "balance_loss_clip": 1.05454552,
      "balance_loss_mlp": 1.01634264,
      "epoch": 0.41724283051764566,
      "flos": 35914763877120.0,
      "grad_norm": 2.333605120141151,
      "language_loss": 0.6106025,
      "learning_rate": 2.6227734785243924e-06,
      "loss": 0.63274449,
      "num_input_tokens_seen": 74849760,
      "step": 3470,
      "time_per_iteration": 2.7114744186401367
    },
    {
      "auxiliary_loss_clip": 0.01432293,
      "auxiliary_loss_mlp": 0.01026463,
      "balance_loss_clip": 1.0445298,
      "balance_loss_mlp": 1.01909554,
      "epoch": 0.4173630734082847,
      "flos": 25333649320320.0,
      "grad_norm": 1.9118279290737445,
      "language_loss": 0.79138803,
      "learning_rate": 2.6220331875795466e-06,
      "loss": 0.81597561,
      "num_input_tokens_seen": 74869110,
      "step": 3471,
      "time_per_iteration": 2.814940929412842
    },
    {
      "auxiliary_loss_clip": 0.01238541,
      "auxiliary_loss_mlp": 0.01028401,
      "balance_loss_clip": 1.05196273,
      "balance_loss_mlp": 1.02097118,
      "epoch": 0.4174833162989238,
      "flos": 26685398868480.0,
      "grad_norm": 1.7464027061665646,
      "language_loss": 0.74936521,
      "learning_rate": 2.62129280227276e-06,
      "loss": 0.7720347,
      "num_input_tokens_seen": 74889110,
      "step": 3472,
      "time_per_iteration": 2.6665444374084473
    },
    {
      "auxiliary_loss_clip": 0.01247451,
      "auxiliary_loss_mlp": 0.01027899,
      "balance_loss_clip": 1.05562901,
      "balance_loss_mlp": 1.02070808,
      "epoch": 0.41760355918956293,
      "flos": 74739584010240.0,
      "grad_norm": 2.4194097598613276,
      "language_loss": 0.68604165,
      "learning_rate": 2.62055232271635e-06,
      "loss": 0.70879519,
      "num_input_tokens_seen": 74916260,
      "step": 3473,
      "time_per_iteration": 4.023009300231934
    },
    {
      "auxiliary_loss_clip": 0.0133433,
      "auxiliary_loss_mlp": 0.01030292,
      "balance_loss_clip": 1.04724002,
      "balance_loss_mlp": 1.02285063,
      "epoch": 0.417723802080202,
      "flos": 14317513148160.0,
      "grad_norm": 2.5186017570183346,
      "language_loss": 0.87773311,
      "learning_rate": 2.619811749022646e-06,
      "loss": 0.90137935,
      "num_input_tokens_seen": 74931570,
      "step": 3474,
      "time_per_iteration": 2.8331046104431152
    },
    {
      "auxiliary_loss_clip": 0.01245047,
      "auxiliary_loss_mlp": 0.01028495,
      "balance_loss_clip": 1.05641317,
      "balance_loss_mlp": 1.020437,
      "epoch": 0.4178440449708411,
      "flos": 14643797316480.0,
      "grad_norm": 2.4597500506819867,
      "language_loss": 0.70997238,
      "learning_rate": 2.6190710813039917e-06,
      "loss": 0.7327078,
      "num_input_tokens_seen": 74944695,
      "step": 3475,
      "time_per_iteration": 2.594266176223755
    },
    {
      "auxiliary_loss_clip": 0.01387783,
      "auxiliary_loss_mlp": 0.02571702,
      "balance_loss_clip": 1.04560041,
      "balance_loss_mlp": 1.00018656,
      "epoch": 0.4179642878614802,
      "flos": 21507296094720.0,
      "grad_norm": 3.467024486296244,
      "language_loss": 0.84057593,
      "learning_rate": 2.618330319672747e-06,
      "loss": 0.88017076,
      "num_input_tokens_seen": 74964115,
      "step": 3476,
      "time_per_iteration": 3.7282044887542725
    },
    {
      "auxiliary_loss_clip": 0.01191928,
      "auxiliary_loss_mlp": 0.01029632,
      "balance_loss_clip": 1.05496597,
      "balance_loss_mlp": 1.02168107,
      "epoch": 0.41808453075211927,
      "flos": 18441997257600.0,
      "grad_norm": 2.651459237973374,
      "language_loss": 0.91905099,
      "learning_rate": 2.617589464241284e-06,
      "loss": 0.94126654,
      "num_input_tokens_seen": 74978515,
      "step": 3477,
      "time_per_iteration": 2.7227745056152344
    },
    {
      "auxiliary_loss_clip": 0.01248438,
      "auxiliary_loss_mlp": 0.01028441,
      "balance_loss_clip": 1.05239415,
      "balance_loss_mlp": 1.0211935,
      "epoch": 0.4182047736427584,
      "flos": 20301020628480.0,
      "grad_norm": 2.06703165967826,
      "language_loss": 0.74273658,
      "learning_rate": 2.6168485151219914e-06,
      "loss": 0.76550537,
      "num_input_tokens_seen": 74998135,
      "step": 3478,
      "time_per_iteration": 2.8631722927093506
    },
    {
      "auxiliary_loss_clip": 0.01240661,
      "auxiliary_loss_mlp": 0.01027181,
      "balance_loss_clip": 1.05475175,
      "balance_loss_mlp": 1.019418,
      "epoch": 0.4183250165333975,
      "flos": 18876623823360.0,
      "grad_norm": 2.4356102059242524,
      "language_loss": 0.71638489,
      "learning_rate": 2.616107472427269e-06,
      "loss": 0.73906332,
      "num_input_tokens_seen": 75012830,
      "step": 3479,
      "time_per_iteration": 2.649883508682251
    },
    {
      "auxiliary_loss_clip": 0.01248157,
      "auxiliary_loss_mlp": 0.01027184,
      "balance_loss_clip": 1.05478501,
      "balance_loss_mlp": 1.01985264,
      "epoch": 0.41844525942403654,
      "flos": 17740050698880.0,
      "grad_norm": 2.590023919821982,
      "language_loss": 0.76269937,
      "learning_rate": 2.615366336269533e-06,
      "loss": 0.78545284,
      "num_input_tokens_seen": 75026495,
      "step": 3480,
      "time_per_iteration": 2.6002180576324463
    },
    {
      "auxiliary_loss_clip": 0.01194133,
      "auxiliary_loss_mlp": 0.01027464,
      "balance_loss_clip": 1.0544951,
      "balance_loss_mlp": 1.01946557,
      "epoch": 0.41856550231467565,
      "flos": 18361377181440.0,
      "grad_norm": 2.4651797773111346,
      "language_loss": 0.80277061,
      "learning_rate": 2.6146251067612126e-06,
      "loss": 0.82498658,
      "num_input_tokens_seen": 75041970,
      "step": 3481,
      "time_per_iteration": 2.656378984451294
    },
    {
      "auxiliary_loss_clip": 0.0125083,
      "auxiliary_loss_mlp": 0.01024734,
      "balance_loss_clip": 1.06088769,
      "balance_loss_mlp": 1.0173583,
      "epoch": 0.41868574520531476,
      "flos": 22781801445120.0,
      "grad_norm": 2.4231134727503805,
      "language_loss": 0.82494718,
      "learning_rate": 2.6138837840147525e-06,
      "loss": 0.84770286,
      "num_input_tokens_seen": 75061005,
      "step": 3482,
      "time_per_iteration": 2.6421399116516113
    },
    {
      "auxiliary_loss_clip": 0.01336995,
      "auxiliary_loss_mlp": 0.01028532,
      "balance_loss_clip": 1.04931748,
      "balance_loss_mlp": 1.02108741,
      "epoch": 0.4188059880959538,
      "flos": 13699167494400.0,
      "grad_norm": 2.644039456723297,
      "language_loss": 0.75954604,
      "learning_rate": 2.6131423681426103e-06,
      "loss": 0.78320134,
      "num_input_tokens_seen": 75076920,
      "step": 3483,
      "time_per_iteration": 2.713409900665283
    },
    {
      "auxiliary_loss_clip": 0.0119062,
      "auxiliary_loss_mlp": 0.01030631,
      "balance_loss_clip": 1.05510783,
      "balance_loss_mlp": 1.02302265,
      "epoch": 0.41892623098659293,
      "flos": 37818281220480.0,
      "grad_norm": 1.7009372524814963,
      "language_loss": 0.72774798,
      "learning_rate": 2.6124008592572587e-06,
      "loss": 0.74996042,
      "num_input_tokens_seen": 75100905,
      "step": 3484,
      "time_per_iteration": 2.742433786392212
    },
    {
      "auxiliary_loss_clip": 0.01195454,
      "auxiliary_loss_mlp": 0.01027951,
      "balance_loss_clip": 1.05525708,
      "balance_loss_mlp": 1.02051854,
      "epoch": 0.419046473877232,
      "flos": 23258874908160.0,
      "grad_norm": 3.165018618754427,
      "language_loss": 0.81570005,
      "learning_rate": 2.6116592574711835e-06,
      "loss": 0.83793408,
      "num_input_tokens_seen": 75119205,
      "step": 3485,
      "time_per_iteration": 2.6686594486236572
    },
    {
      "auxiliary_loss_clip": 0.01197809,
      "auxiliary_loss_mlp": 0.01034556,
      "balance_loss_clip": 1.05726624,
      "balance_loss_mlp": 1.02650917,
      "epoch": 0.4191667167678711,
      "flos": 20741034234240.0,
      "grad_norm": 3.730774587817408,
      "language_loss": 0.84339887,
      "learning_rate": 2.6109175628968853e-06,
      "loss": 0.86572248,
      "num_input_tokens_seen": 75138970,
      "step": 3486,
      "time_per_iteration": 2.6197946071624756
    },
    {
      "auxiliary_loss_clip": 0.0123374,
      "auxiliary_loss_mlp": 0.01031101,
      "balance_loss_clip": 1.0501647,
      "balance_loss_mlp": 1.02399647,
      "epoch": 0.4192869596585102,
      "flos": 23586416052480.0,
      "grad_norm": 1.9574662678005559,
      "language_loss": 0.82956469,
      "learning_rate": 2.610175775646878e-06,
      "loss": 0.85221314,
      "num_input_tokens_seen": 75157550,
      "step": 3487,
      "time_per_iteration": 2.727952003479004
    },
    {
      "auxiliary_loss_clip": 0.01285731,
      "auxiliary_loss_mlp": 0.01026204,
      "balance_loss_clip": 1.04791939,
      "balance_loss_mlp": 1.01822257,
      "epoch": 0.41940720254914926,
      "flos": 25081269384960.0,
      "grad_norm": 2.30933263204728,
      "language_loss": 0.73413873,
      "learning_rate": 2.6094338958336907e-06,
      "loss": 0.75725812,
      "num_input_tokens_seen": 75176220,
      "step": 3488,
      "time_per_iteration": 2.7601747512817383
    },
    {
      "auxiliary_loss_clip": 0.01291296,
      "auxiliary_loss_mlp": 0.01031392,
      "balance_loss_clip": 1.05446768,
      "balance_loss_mlp": 1.02427876,
      "epoch": 0.41952744543978837,
      "flos": 15554132628480.0,
      "grad_norm": 2.437947556974515,
      "language_loss": 0.82288826,
      "learning_rate": 2.608691923569867e-06,
      "loss": 0.84611511,
      "num_input_tokens_seen": 75193095,
      "step": 3489,
      "time_per_iteration": 2.795093059539795
    },
    {
      "auxiliary_loss_clip": 0.01244702,
      "auxiliary_loss_mlp": 0.01034777,
      "balance_loss_clip": 1.05605948,
      "balance_loss_mlp": 1.02725506,
      "epoch": 0.4196476883304275,
      "flos": 24644775312000.0,
      "grad_norm": 1.7835820277788854,
      "language_loss": 0.758035,
      "learning_rate": 2.6079498589679616e-06,
      "loss": 0.78082979,
      "num_input_tokens_seen": 75214185,
      "step": 3490,
      "time_per_iteration": 2.8356852531433105
    },
    {
      "auxiliary_loss_clip": 0.01439934,
      "auxiliary_loss_mlp": 0.01030998,
      "balance_loss_clip": 1.04547167,
      "balance_loss_mlp": 1.021873,
      "epoch": 0.41976793122106654,
      "flos": 24531333183360.0,
      "grad_norm": 1.8673736910226,
      "language_loss": 0.76319587,
      "learning_rate": 2.6072077021405465e-06,
      "loss": 0.78790516,
      "num_input_tokens_seen": 75233020,
      "step": 3491,
      "time_per_iteration": 3.7237002849578857
    },
    {
      "auxiliary_loss_clip": 0.01345931,
      "auxiliary_loss_mlp": 0.01027387,
      "balance_loss_clip": 1.04845047,
      "balance_loss_mlp": 1.01988256,
      "epoch": 0.41988817411170565,
      "flos": 21175301664000.0,
      "grad_norm": 1.7737324433810486,
      "language_loss": 0.69377434,
      "learning_rate": 2.6064654532002054e-06,
      "loss": 0.71750754,
      "num_input_tokens_seen": 75252030,
      "step": 3492,
      "time_per_iteration": 2.7856907844543457
    },
    {
      "auxiliary_loss_clip": 0.01194255,
      "auxiliary_loss_mlp": 0.01029041,
      "balance_loss_clip": 1.05702901,
      "balance_loss_mlp": 1.0211072,
      "epoch": 0.42000841700234476,
      "flos": 31649402626560.0,
      "grad_norm": 2.697961330916967,
      "language_loss": 0.75965565,
      "learning_rate": 2.6057231122595375e-06,
      "loss": 0.7818886,
      "num_input_tokens_seen": 75273340,
      "step": 3493,
      "time_per_iteration": 2.6755478382110596
    },
    {
      "auxiliary_loss_clip": 0.01288445,
      "auxiliary_loss_mlp": 0.01036159,
      "balance_loss_clip": 1.04874575,
      "balance_loss_mlp": 1.02823138,
      "epoch": 0.4201286598929838,
      "flos": 21281525159040.0,
      "grad_norm": 1.7801986371993321,
      "language_loss": 0.72842264,
      "learning_rate": 2.604980679431154e-06,
      "loss": 0.75166863,
      "num_input_tokens_seen": 75291580,
      "step": 3494,
      "time_per_iteration": 3.660038471221924
    },
    {
      "auxiliary_loss_clip": 0.01242639,
      "auxiliary_loss_mlp": 0.01026519,
      "balance_loss_clip": 1.05201375,
      "balance_loss_mlp": 1.01917529,
      "epoch": 0.4202489027836229,
      "flos": 18546532813440.0,
      "grad_norm": 1.9277142987086304,
      "language_loss": 0.744946,
      "learning_rate": 2.604238154827684e-06,
      "loss": 0.76763755,
      "num_input_tokens_seen": 75308205,
      "step": 3495,
      "time_per_iteration": 2.688359260559082
    },
    {
      "auxiliary_loss_clip": 0.01242999,
      "auxiliary_loss_mlp": 0.01024976,
      "balance_loss_clip": 1.05353546,
      "balance_loss_mlp": 1.01716769,
      "epoch": 0.42036914567426203,
      "flos": 19317643009920.0,
      "grad_norm": 2.3149940429365965,
      "language_loss": 0.72489792,
      "learning_rate": 2.6034955385617656e-06,
      "loss": 0.74757767,
      "num_input_tokens_seen": 75326535,
      "step": 3496,
      "time_per_iteration": 2.6302592754364014
    },
    {
      "auxiliary_loss_clip": 0.01250018,
      "auxiliary_loss_mlp": 0.00999584,
      "balance_loss_clip": 1.02473354,
      "balance_loss_mlp": 0.99845713,
      "epoch": 0.4204893885649011,
      "flos": 67842942935040.0,
      "grad_norm": 0.7166707974741341,
      "language_loss": 0.61576146,
      "learning_rate": 2.6027528307460544e-06,
      "loss": 0.63825744,
      "num_input_tokens_seen": 75390540,
      "step": 3497,
      "time_per_iteration": 3.358452558517456
    },
    {
      "auxiliary_loss_clip": 0.01193236,
      "auxiliary_loss_mlp": 0.01036281,
      "balance_loss_clip": 1.05498719,
      "balance_loss_mlp": 1.0287168,
      "epoch": 0.4206096314555402,
      "flos": 21908777385600.0,
      "grad_norm": 1.680685894449732,
      "language_loss": 0.86466801,
      "learning_rate": 2.602010031493217e-06,
      "loss": 0.88696325,
      "num_input_tokens_seen": 75408770,
      "step": 3498,
      "time_per_iteration": 2.657012462615967
    },
    {
      "auxiliary_loss_clip": 0.01344378,
      "auxiliary_loss_mlp": 0.01023714,
      "balance_loss_clip": 1.05530214,
      "balance_loss_mlp": 1.01627564,
      "epoch": 0.42072987434617926,
      "flos": 29278185269760.0,
      "grad_norm": 2.536653802429296,
      "language_loss": 0.86763251,
      "learning_rate": 2.6012671409159367e-06,
      "loss": 0.89131343,
      "num_input_tokens_seen": 75430105,
      "step": 3499,
      "time_per_iteration": 2.7806475162506104
    },
    {
      "auxiliary_loss_clip": 0.01285965,
      "auxiliary_loss_mlp": 0.01029114,
      "balance_loss_clip": 1.05105913,
      "balance_loss_mlp": 1.02158272,
      "epoch": 0.42085011723681837,
      "flos": 27600726170880.0,
      "grad_norm": 1.7557637558502752,
      "language_loss": 0.81484747,
      "learning_rate": 2.6005241591269097e-06,
      "loss": 0.83799827,
      "num_input_tokens_seen": 75449475,
      "step": 3500,
      "time_per_iteration": 3.6815176010131836
    },
    {
      "auxiliary_loss_clip": 0.01334414,
      "auxiliary_loss_mlp": 0.01027675,
      "balance_loss_clip": 1.05321789,
      "balance_loss_mlp": 1.0208441,
      "epoch": 0.4209703601274575,
      "flos": 27818632028160.0,
      "grad_norm": 1.6509348905637071,
      "language_loss": 0.80016756,
      "learning_rate": 2.5997810862388454e-06,
      "loss": 0.8237884,
      "num_input_tokens_seen": 75469315,
      "step": 3501,
      "time_per_iteration": 3.645993232727051
    },
    {
      "auxiliary_loss_clip": 0.01287249,
      "auxiliary_loss_mlp": 0.0102792,
      "balance_loss_clip": 1.04890513,
      "balance_loss_mlp": 1.02051735,
      "epoch": 0.42109060301809653,
      "flos": 27525529048320.0,
      "grad_norm": 2.39139876279663,
      "language_loss": 0.7574538,
      "learning_rate": 2.599037922364467e-06,
      "loss": 0.78060555,
      "num_input_tokens_seen": 75488215,
      "step": 3502,
      "time_per_iteration": 2.761430501937866
    },
    {
      "auxiliary_loss_clip": 0.01339757,
      "auxiliary_loss_mlp": 0.01030712,
      "balance_loss_clip": 1.05566716,
      "balance_loss_mlp": 1.02276063,
      "epoch": 0.42121084590873564,
      "flos": 29314275459840.0,
      "grad_norm": 2.559251718837426,
      "language_loss": 0.75700676,
      "learning_rate": 2.5982946676165112e-06,
      "loss": 0.78071141,
      "num_input_tokens_seen": 75507985,
      "step": 3503,
      "time_per_iteration": 2.7895472049713135
    },
    {
      "auxiliary_loss_clip": 0.01262329,
      "auxiliary_loss_mlp": 0.01001469,
      "balance_loss_clip": 1.03492129,
      "balance_loss_mlp": 1.0004437,
      "epoch": 0.42133108879937475,
      "flos": 67398835178880.0,
      "grad_norm": 0.7287752425719596,
      "language_loss": 0.57558417,
      "learning_rate": 2.5975513221077313e-06,
      "loss": 0.59822226,
      "num_input_tokens_seen": 75571955,
      "step": 3504,
      "time_per_iteration": 3.3751442432403564
    },
    {
      "auxiliary_loss_clip": 0.01283874,
      "auxiliary_loss_mlp": 0.01027483,
      "balance_loss_clip": 1.04909086,
      "balance_loss_mlp": 1.02027118,
      "epoch": 0.4214513316900138,
      "flos": 23106038538240.0,
      "grad_norm": 2.3404125564368625,
      "language_loss": 0.88575697,
      "learning_rate": 2.5968078859508897e-06,
      "loss": 0.90887058,
      "num_input_tokens_seen": 75589155,
      "step": 3505,
      "time_per_iteration": 2.732548713684082
    },
    {
      "auxiliary_loss_clip": 0.01237519,
      "auxiliary_loss_mlp": 0.01028227,
      "balance_loss_clip": 1.05269003,
      "balance_loss_mlp": 1.02075839,
      "epoch": 0.4215715745806529,
      "flos": 15336190857600.0,
      "grad_norm": 2.309272240206507,
      "language_loss": 0.80328012,
      "learning_rate": 2.5960643592587673e-06,
      "loss": 0.82593763,
      "num_input_tokens_seen": 75606565,
      "step": 3506,
      "time_per_iteration": 2.6634674072265625
    },
    {
      "auxiliary_loss_clip": 0.01335402,
      "auxiliary_loss_mlp": 0.01032307,
      "balance_loss_clip": 1.04752696,
      "balance_loss_mlp": 1.02492166,
      "epoch": 0.42169181747129203,
      "flos": 22127257860480.0,
      "grad_norm": 1.8946584905134976,
      "language_loss": 0.81443924,
      "learning_rate": 2.5953207421441553e-06,
      "loss": 0.83811635,
      "num_input_tokens_seen": 75625165,
      "step": 3507,
      "time_per_iteration": 2.8137407302856445
    },
    {
      "auxiliary_loss_clip": 0.01353082,
      "auxiliary_loss_mlp": 0.01037899,
      "balance_loss_clip": 1.05727375,
      "balance_loss_mlp": 1.03073192,
      "epoch": 0.4218120603619311,
      "flos": 22630724841600.0,
      "grad_norm": 2.7956781925204393,
      "language_loss": 0.75119758,
      "learning_rate": 2.5945770347198603e-06,
      "loss": 0.77510738,
      "num_input_tokens_seen": 75643320,
      "step": 3508,
      "time_per_iteration": 2.8254239559173584
    },
    {
      "auxiliary_loss_clip": 0.01193069,
      "auxiliary_loss_mlp": 0.01029225,
      "balance_loss_clip": 1.04927611,
      "balance_loss_mlp": 1.02203369,
      "epoch": 0.4219323032525702,
      "flos": 19682818629120.0,
      "grad_norm": 2.0022972274092914,
      "language_loss": 0.82005894,
      "learning_rate": 2.593833237098701e-06,
      "loss": 0.84228188,
      "num_input_tokens_seen": 75660920,
      "step": 3509,
      "time_per_iteration": 2.8157765865325928
    },
    {
      "auxiliary_loss_clip": 0.01240527,
      "auxiliary_loss_mlp": 0.01026393,
      "balance_loss_clip": 1.05050433,
      "balance_loss_mlp": 1.01870131,
      "epoch": 0.4220525461432093,
      "flos": 30190747224960.0,
      "grad_norm": 2.0750702509307026,
      "language_loss": 0.62558603,
      "learning_rate": 2.593089349393512e-06,
      "loss": 0.64825523,
      "num_input_tokens_seen": 75681410,
      "step": 3510,
      "time_per_iteration": 2.743659496307373
    },
    {
      "auxiliary_loss_clip": 0.01248207,
      "auxiliary_loss_mlp": 0.01028625,
      "balance_loss_clip": 1.05893517,
      "balance_loss_mlp": 1.02047729,
      "epoch": 0.42217278903384836,
      "flos": 24315941278080.0,
      "grad_norm": 2.172459671529353,
      "language_loss": 0.83858824,
      "learning_rate": 2.592345371717141e-06,
      "loss": 0.86135662,
      "num_input_tokens_seen": 75700940,
      "step": 3511,
      "time_per_iteration": 2.6558923721313477
    },
    {
      "auxiliary_loss_clip": 0.01244635,
      "auxiliary_loss_mlp": 0.01030623,
      "balance_loss_clip": 1.05870664,
      "balance_loss_mlp": 1.02308273,
      "epoch": 0.42229303192448747,
      "flos": 17092474352640.0,
      "grad_norm": 2.195073841806471,
      "language_loss": 0.71847153,
      "learning_rate": 2.591601304182448e-06,
      "loss": 0.74122411,
      "num_input_tokens_seen": 75718910,
      "step": 3512,
      "time_per_iteration": 2.663313388824463
    },
    {
      "auxiliary_loss_clip": 0.01290839,
      "auxiliary_loss_mlp": 0.01025918,
      "balance_loss_clip": 1.05605173,
      "balance_loss_mlp": 1.01904035,
      "epoch": 0.4224132748151266,
      "flos": 22784530878720.0,
      "grad_norm": 1.7726631750954647,
      "language_loss": 0.79306513,
      "learning_rate": 2.5908571469023067e-06,
      "loss": 0.81623274,
      "num_input_tokens_seen": 75738395,
      "step": 3513,
      "time_per_iteration": 2.679189443588257
    },
    {
      "auxiliary_loss_clip": 0.01191506,
      "auxiliary_loss_mlp": 0.0102597,
      "balance_loss_clip": 1.05522776,
      "balance_loss_mlp": 1.01875818,
      "epoch": 0.42253351770576564,
      "flos": 17819090576640.0,
      "grad_norm": 2.762360493613974,
      "language_loss": 0.75418633,
      "learning_rate": 2.5901128999896067e-06,
      "loss": 0.77636111,
      "num_input_tokens_seen": 75753825,
      "step": 3514,
      "time_per_iteration": 2.5917623043060303
    },
    {
      "auxiliary_loss_clip": 0.01240815,
      "auxiliary_loss_mlp": 0.01024775,
      "balance_loss_clip": 1.05527091,
      "balance_loss_mlp": 1.01707959,
      "epoch": 0.42265376059640475,
      "flos": 28512390286080.0,
      "grad_norm": 3.532590861987271,
      "language_loss": 0.68374795,
      "learning_rate": 2.5893685635572487e-06,
      "loss": 0.70640385,
      "num_input_tokens_seen": 75774675,
      "step": 3515,
      "time_per_iteration": 2.8240742683410645
    },
    {
      "auxiliary_loss_clip": 0.01290365,
      "auxiliary_loss_mlp": 0.01031291,
      "balance_loss_clip": 1.05556083,
      "balance_loss_mlp": 1.02345014,
      "epoch": 0.4227740034870438,
      "flos": 16253349753600.0,
      "grad_norm": 2.1362133545207143,
      "language_loss": 0.69550979,
      "learning_rate": 2.5886241377181483e-06,
      "loss": 0.7187264,
      "num_input_tokens_seen": 75793545,
      "step": 3516,
      "time_per_iteration": 2.7421350479125977
    },
    {
      "auxiliary_loss_clip": 0.01240934,
      "auxiliary_loss_mlp": 0.01027598,
      "balance_loss_clip": 1.0548811,
      "balance_loss_mlp": 1.01970053,
      "epoch": 0.4228942463776829,
      "flos": 25295691623040.0,
      "grad_norm": 1.8695213007405325,
      "language_loss": 0.81148392,
      "learning_rate": 2.587879622585234e-06,
      "loss": 0.83416933,
      "num_input_tokens_seen": 75812145,
      "step": 3517,
      "time_per_iteration": 3.6194565296173096
    },
    {
      "auxiliary_loss_clip": 0.01243433,
      "auxiliary_loss_mlp": 0.0102351,
      "balance_loss_clip": 1.05889523,
      "balance_loss_mlp": 1.0158658,
      "epoch": 0.423014489268322,
      "flos": 26395779507840.0,
      "grad_norm": 2.5785791369569226,
      "language_loss": 0.75266647,
      "learning_rate": 2.5871350182714486e-06,
      "loss": 0.77533597,
      "num_input_tokens_seen": 75833025,
      "step": 3518,
      "time_per_iteration": 2.6951751708984375
    },
    {
      "auxiliary_loss_clip": 0.01191745,
      "auxiliary_loss_mlp": 0.01028828,
      "balance_loss_clip": 1.05594182,
      "balance_loss_mlp": 1.02137172,
      "epoch": 0.4231347321589611,
      "flos": 17274002711040.0,
      "grad_norm": 2.288602604767143,
      "language_loss": 0.80711192,
      "learning_rate": 2.586390324889748e-06,
      "loss": 0.82931763,
      "num_input_tokens_seen": 75848925,
      "step": 3519,
      "time_per_iteration": 2.532463312149048
    },
    {
      "auxiliary_loss_clip": 0.01242615,
      "auxiliary_loss_mlp": 0.01025955,
      "balance_loss_clip": 1.05659795,
      "balance_loss_mlp": 1.01857853,
      "epoch": 0.4232549750496002,
      "flos": 22999635475200.0,
      "grad_norm": 1.924565262351091,
      "language_loss": 0.67689395,
      "learning_rate": 2.5856455425531003e-06,
      "loss": 0.6995796,
      "num_input_tokens_seen": 75870400,
      "step": 3520,
      "time_per_iteration": 3.6152708530426025
    },
    {
      "auxiliary_loss_clip": 0.01236732,
      "auxiliary_loss_mlp": 0.01024535,
      "balance_loss_clip": 1.05536962,
      "balance_loss_mlp": 1.01678681,
      "epoch": 0.4233752179402393,
      "flos": 21248343970560.0,
      "grad_norm": 2.0270736140241645,
      "language_loss": 0.80835187,
      "learning_rate": 2.5849006713744902e-06,
      "loss": 0.83096457,
      "num_input_tokens_seen": 75889195,
      "step": 3521,
      "time_per_iteration": 2.7182681560516357
    },
    {
      "auxiliary_loss_clip": 0.01293415,
      "auxiliary_loss_mlp": 0.01025247,
      "balance_loss_clip": 1.05491924,
      "balance_loss_mlp": 1.01773977,
      "epoch": 0.42349546083087836,
      "flos": 20704297599360.0,
      "grad_norm": 2.007678195692539,
      "language_loss": 0.73223752,
      "learning_rate": 2.5841557114669135e-06,
      "loss": 0.75542414,
      "num_input_tokens_seen": 75906055,
      "step": 3522,
      "time_per_iteration": 2.65470027923584
    },
    {
      "auxiliary_loss_clip": 0.01197817,
      "auxiliary_loss_mlp": 0.01032439,
      "balance_loss_clip": 1.05752158,
      "balance_loss_mlp": 1.02420735,
      "epoch": 0.42361570372151747,
      "flos": 18585065128320.0,
      "grad_norm": 4.5315287505972295,
      "language_loss": 0.67348766,
      "learning_rate": 2.58341066294338e-06,
      "loss": 0.69579017,
      "num_input_tokens_seen": 75922720,
      "step": 3523,
      "time_per_iteration": 2.595306158065796
    },
    {
      "auxiliary_loss_clip": 0.01393603,
      "auxiliary_loss_mlp": 0.02568763,
      "balance_loss_clip": 1.05111432,
      "balance_loss_mlp": 1.00006914,
      "epoch": 0.4237359466121566,
      "flos": 20959478795520.0,
      "grad_norm": 2.1974688104485027,
      "language_loss": 0.85728723,
      "learning_rate": 2.5826655259169124e-06,
      "loss": 0.89691091,
      "num_input_tokens_seen": 75941375,
      "step": 3524,
      "time_per_iteration": 2.7252016067504883
    },
    {
      "auxiliary_loss_clip": 0.01195535,
      "auxiliary_loss_mlp": 0.01031708,
      "balance_loss_clip": 1.05896306,
      "balance_loss_mlp": 1.02424836,
      "epoch": 0.42385618950279563,
      "flos": 18038181582720.0,
      "grad_norm": 2.236034283812978,
      "language_loss": 0.90505362,
      "learning_rate": 2.5819203005005475e-06,
      "loss": 0.92732596,
      "num_input_tokens_seen": 75958710,
      "step": 3525,
      "time_per_iteration": 3.5041046142578125
    },
    {
      "auxiliary_loss_clip": 0.01284529,
      "auxiliary_loss_mlp": 0.01023084,
      "balance_loss_clip": 1.05291319,
      "balance_loss_mlp": 1.01540673,
      "epoch": 0.42397643239343474,
      "flos": 23769129559680.0,
      "grad_norm": 1.7779120112742763,
      "language_loss": 0.78672218,
      "learning_rate": 2.581174986807336e-06,
      "loss": 0.8097983,
      "num_input_tokens_seen": 75978945,
      "step": 3526,
      "time_per_iteration": 2.706537961959839
    },
    {
      "auxiliary_loss_clip": 0.01238185,
      "auxiliary_loss_mlp": 0.02566968,
      "balance_loss_clip": 1.0537256,
      "balance_loss_mlp": 1.00009394,
      "epoch": 0.42409667528407385,
      "flos": 16545088016640.0,
      "grad_norm": 3.676865209884386,
      "language_loss": 0.91295719,
      "learning_rate": 2.580429584950341e-06,
      "loss": 0.95100874,
      "num_input_tokens_seen": 75994695,
      "step": 3527,
      "time_per_iteration": 3.511000871658325
    },
    {
      "auxiliary_loss_clip": 0.01264212,
      "auxiliary_loss_mlp": 0.01029149,
      "balance_loss_clip": 1.06081188,
      "balance_loss_mlp": 1.02115631,
      "epoch": 0.4242169181747129,
      "flos": 16034186920320.0,
      "grad_norm": 2.1064937604000256,
      "language_loss": 0.66694582,
      "learning_rate": 2.5796840950426397e-06,
      "loss": 0.68987948,
      "num_input_tokens_seen": 76011780,
      "step": 3528,
      "time_per_iteration": 2.763439893722534
    },
    {
      "auxiliary_loss_clip": 0.01239245,
      "auxiliary_loss_mlp": 0.01025938,
      "balance_loss_clip": 1.05467749,
      "balance_loss_mlp": 1.0185504,
      "epoch": 0.424337161065352,
      "flos": 20084012611200.0,
      "grad_norm": 1.7263638505378958,
      "language_loss": 0.66059816,
      "learning_rate": 2.578938517197322e-06,
      "loss": 0.68324995,
      "num_input_tokens_seen": 76029875,
      "step": 3529,
      "time_per_iteration": 2.7407844066619873
    },
    {
      "auxiliary_loss_clip": 0.01287547,
      "auxiliary_loss_mlp": 0.01034665,
      "balance_loss_clip": 1.05277312,
      "balance_loss_mlp": 1.02670765,
      "epoch": 0.4244574039559911,
      "flos": 23878369797120.0,
      "grad_norm": 2.610715029345118,
      "language_loss": 0.62515181,
      "learning_rate": 2.5781928515274916e-06,
      "loss": 0.64837396,
      "num_input_tokens_seen": 76048595,
      "step": 3530,
      "time_per_iteration": 2.7033684253692627
    },
    {
      "auxiliary_loss_clip": 0.01247299,
      "auxiliary_loss_mlp": 0.01027434,
      "balance_loss_clip": 1.05833101,
      "balance_loss_mlp": 1.01998913,
      "epoch": 0.4245776468466302,
      "flos": 17565920542080.0,
      "grad_norm": 2.2528227067783058,
      "language_loss": 0.68186784,
      "learning_rate": 2.577447098146265e-06,
      "loss": 0.70461524,
      "num_input_tokens_seen": 76065770,
      "step": 3531,
      "time_per_iteration": 2.627251625061035
    },
    {
      "auxiliary_loss_clip": 0.01346442,
      "auxiliary_loss_mlp": 0.01025031,
      "balance_loss_clip": 1.05580533,
      "balance_loss_mlp": 1.01691329,
      "epoch": 0.4246978897372693,
      "flos": 27776256958080.0,
      "grad_norm": 1.5664870670784932,
      "language_loss": 0.78998137,
      "learning_rate": 2.5767012571667724e-06,
      "loss": 0.81369615,
      "num_input_tokens_seen": 76085250,
      "step": 3532,
      "time_per_iteration": 2.7716145515441895
    },
    {
      "auxiliary_loss_clip": 0.01246512,
      "auxiliary_loss_mlp": 0.0102769,
      "balance_loss_clip": 1.05437255,
      "balance_loss_mlp": 1.01951849,
      "epoch": 0.42481813262790835,
      "flos": 15596615439360.0,
      "grad_norm": 1.7232185349614153,
      "language_loss": 0.68571705,
      "learning_rate": 2.5759553287021587e-06,
      "loss": 0.70845908,
      "num_input_tokens_seen": 76103580,
      "step": 3533,
      "time_per_iteration": 2.645460367202759
    },
    {
      "auxiliary_loss_clip": 0.0129921,
      "auxiliary_loss_mlp": 0.01027542,
      "balance_loss_clip": 1.05862033,
      "balance_loss_mlp": 1.01982963,
      "epoch": 0.42493837551854746,
      "flos": 23951088881280.0,
      "grad_norm": 1.9034215753723833,
      "language_loss": 0.78174138,
      "learning_rate": 2.5752093128655786e-06,
      "loss": 0.80500883,
      "num_input_tokens_seen": 76121825,
      "step": 3534,
      "time_per_iteration": 2.6956915855407715
    },
    {
      "auxiliary_loss_clip": 0.01284537,
      "auxiliary_loss_mlp": 0.01028682,
      "balance_loss_clip": 1.05015421,
      "balance_loss_mlp": 1.02119601,
      "epoch": 0.4250586184091866,
      "flos": 20813466009600.0,
      "grad_norm": 2.0592459710215674,
      "language_loss": 0.74063557,
      "learning_rate": 2.574463209770204e-06,
      "loss": 0.76376772,
      "num_input_tokens_seen": 76141140,
      "step": 3535,
      "time_per_iteration": 2.707911491394043
    },
    {
      "auxiliary_loss_clip": 0.01344018,
      "auxiliary_loss_mlp": 0.01026019,
      "balance_loss_clip": 1.0500977,
      "balance_loss_mlp": 1.01824653,
      "epoch": 0.42517886129982563,
      "flos": 30371018607360.0,
      "grad_norm": 4.808814599795064,
      "language_loss": 0.79674786,
      "learning_rate": 2.5737170195292165e-06,
      "loss": 0.82044828,
      "num_input_tokens_seen": 76164475,
      "step": 3536,
      "time_per_iteration": 2.7150142192840576
    },
    {
      "auxiliary_loss_clip": 0.01342341,
      "auxiliary_loss_mlp": 0.01027364,
      "balance_loss_clip": 1.05123389,
      "balance_loss_mlp": 1.01960397,
      "epoch": 0.42529910419046474,
      "flos": 20080636732800.0,
      "grad_norm": 1.8592996368004149,
      "language_loss": 0.78512931,
      "learning_rate": 2.572970742255814e-06,
      "loss": 0.80882633,
      "num_input_tokens_seen": 76182965,
      "step": 3537,
      "time_per_iteration": 2.8216636180877686
    },
    {
      "auxiliary_loss_clip": 0.01241737,
      "auxiliary_loss_mlp": 0.01027467,
      "balance_loss_clip": 1.05636072,
      "balance_loss_mlp": 1.02031469,
      "epoch": 0.42541934708110385,
      "flos": 22632448694400.0,
      "grad_norm": 1.9981243260482835,
      "language_loss": 0.81669378,
      "learning_rate": 2.5722243780632046e-06,
      "loss": 0.83938587,
      "num_input_tokens_seen": 76201230,
      "step": 3538,
      "time_per_iteration": 2.6477410793304443
    },
    {
      "auxiliary_loss_clip": 0.0130493,
      "auxiliary_loss_mlp": 0.0100691,
      "balance_loss_clip": 1.02494049,
      "balance_loss_mlp": 1.00572419,
      "epoch": 0.4255395899717429,
      "flos": 66200676186240.0,
      "grad_norm": 0.7531108739722716,
      "language_loss": 0.60396814,
      "learning_rate": 2.5714779270646125e-06,
      "loss": 0.62708652,
      "num_input_tokens_seen": 76262000,
      "step": 3539,
      "time_per_iteration": 3.4196176528930664
    },
    {
      "auxiliary_loss_clip": 0.01292804,
      "auxiliary_loss_mlp": 0.02571622,
      "balance_loss_clip": 1.05537045,
      "balance_loss_mlp": 1.00000596,
      "epoch": 0.425659832862382,
      "flos": 17931814433280.0,
      "grad_norm": 2.5360214313249685,
      "language_loss": 0.77856195,
      "learning_rate": 2.5707313893732735e-06,
      "loss": 0.8172062,
      "num_input_tokens_seen": 76280540,
      "step": 3540,
      "time_per_iteration": 2.672560453414917
    },
    {
      "auxiliary_loss_clip": 0.01485771,
      "auxiliary_loss_mlp": 0.01025651,
      "balance_loss_clip": 1.04229975,
      "balance_loss_mlp": 1.01756287,
      "epoch": 0.4257800757530211,
      "flos": 24022550989440.0,
      "grad_norm": 1.7469175447855103,
      "language_loss": 0.77361822,
      "learning_rate": 2.5699847651024364e-06,
      "loss": 0.79873252,
      "num_input_tokens_seen": 76301180,
      "step": 3541,
      "time_per_iteration": 2.835223913192749
    },
    {
      "auxiliary_loss_clip": 0.01238979,
      "auxiliary_loss_mlp": 0.01030724,
      "balance_loss_clip": 1.05731893,
      "balance_loss_mlp": 1.02310967,
      "epoch": 0.4259003186436602,
      "flos": 23696015425920.0,
      "grad_norm": 2.6879814013380248,
      "language_loss": 0.76816285,
      "learning_rate": 2.5692380543653627e-06,
      "loss": 0.79085982,
      "num_input_tokens_seen": 76319335,
      "step": 3542,
      "time_per_iteration": 2.7077231407165527
    },
    {
      "auxiliary_loss_clip": 0.01247944,
      "auxiliary_loss_mlp": 0.02570472,
      "balance_loss_clip": 1.05578637,
      "balance_loss_mlp": 1.0000155,
      "epoch": 0.4260205615342993,
      "flos": 15259772672640.0,
      "grad_norm": 2.22753368745173,
      "language_loss": 0.69766974,
      "learning_rate": 2.5684912572753293e-06,
      "loss": 0.73585391,
      "num_input_tokens_seen": 76335010,
      "step": 3543,
      "time_per_iteration": 3.5436928272247314
    },
    {
      "auxiliary_loss_clip": 0.01190955,
      "auxiliary_loss_mlp": 0.01028041,
      "balance_loss_clip": 1.05663276,
      "balance_loss_mlp": 1.02052498,
      "epoch": 0.4261408044249384,
      "flos": 30665306736000.0,
      "grad_norm": 2.0257777959589793,
      "language_loss": 0.84058696,
      "learning_rate": 2.5677443739456245e-06,
      "loss": 0.86277694,
      "num_input_tokens_seen": 76356670,
      "step": 3544,
      "time_per_iteration": 2.7449755668640137
    },
    {
      "auxiliary_loss_clip": 0.01303327,
      "auxiliary_loss_mlp": 0.01027143,
      "balance_loss_clip": 1.06181717,
      "balance_loss_mlp": 1.0194422,
      "epoch": 0.42626104731557746,
      "flos": 23257905240960.0,
      "grad_norm": 2.909293083878819,
      "language_loss": 0.79077256,
      "learning_rate": 2.5669974044895495e-06,
      "loss": 0.81407726,
      "num_input_tokens_seen": 76373065,
      "step": 3545,
      "time_per_iteration": 2.6300253868103027
    },
    {
      "auxiliary_loss_clip": 0.01348719,
      "auxiliary_loss_mlp": 0.01027657,
      "balance_loss_clip": 1.05229509,
      "balance_loss_mlp": 1.02042139,
      "epoch": 0.42638129020621657,
      "flos": 25884770670720.0,
      "grad_norm": 2.9824126445702817,
      "language_loss": 0.79919058,
      "learning_rate": 2.5662503490204187e-06,
      "loss": 0.8229543,
      "num_input_tokens_seen": 76393230,
      "step": 3546,
      "time_per_iteration": 3.4174816608428955
    },
    {
      "auxiliary_loss_clip": 0.01288971,
      "auxiliary_loss_mlp": 0.01030351,
      "balance_loss_clip": 1.05053997,
      "balance_loss_mlp": 1.02286768,
      "epoch": 0.4265015330968556,
      "flos": 26502362138880.0,
      "grad_norm": 2.2356793285365613,
      "language_loss": 0.75934434,
      "learning_rate": 2.5655032076515603e-06,
      "loss": 0.78253758,
      "num_input_tokens_seen": 76412555,
      "step": 3547,
      "time_per_iteration": 2.7116644382476807
    },
    {
      "auxiliary_loss_clip": 0.01197079,
      "auxiliary_loss_mlp": 0.01030391,
      "balance_loss_clip": 1.05496716,
      "balance_loss_mlp": 1.02274954,
      "epoch": 0.42662177598749473,
      "flos": 24389522288640.0,
      "grad_norm": 2.1308708996788868,
      "language_loss": 0.82761085,
      "learning_rate": 2.5647559804963155e-06,
      "loss": 0.84988546,
      "num_input_tokens_seen": 76432485,
      "step": 3548,
      "time_per_iteration": 2.671523332595825
    },
    {
      "auxiliary_loss_clip": 0.0139785,
      "auxiliary_loss_mlp": 0.01033024,
      "balance_loss_clip": 1.05335832,
      "balance_loss_mlp": 1.02537036,
      "epoch": 0.42674201887813384,
      "flos": 23148629089920.0,
      "grad_norm": 2.7195347735023625,
      "language_loss": 0.78757823,
      "learning_rate": 2.5640086676680364e-06,
      "loss": 0.81188697,
      "num_input_tokens_seen": 76453980,
      "step": 3549,
      "time_per_iteration": 2.8430466651916504
    },
    {
      "auxiliary_loss_clip": 0.01243135,
      "auxiliary_loss_mlp": 0.0103212,
      "balance_loss_clip": 1.05592179,
      "balance_loss_mlp": 1.0237813,
      "epoch": 0.4268622617687729,
      "flos": 21689614552320.0,
      "grad_norm": 2.99722508486976,
      "language_loss": 0.80737597,
      "learning_rate": 2.5632612692800923e-06,
      "loss": 0.83012849,
      "num_input_tokens_seen": 76473045,
      "step": 3550,
      "time_per_iteration": 2.71234130859375
    },
    {
      "auxiliary_loss_clip": 0.01341115,
      "auxiliary_loss_mlp": 0.01031169,
      "balance_loss_clip": 1.05361176,
      "balance_loss_mlp": 1.02284825,
      "epoch": 0.426982504659412,
      "flos": 23440151871360.0,
      "grad_norm": 2.325844986068865,
      "language_loss": 0.75789869,
      "learning_rate": 2.5625137854458603e-06,
      "loss": 0.78162152,
      "num_input_tokens_seen": 76492060,
      "step": 3551,
      "time_per_iteration": 3.8944945335388184
    },
    {
      "auxiliary_loss_clip": 0.01294618,
      "auxiliary_loss_mlp": 0.01023542,
      "balance_loss_clip": 1.05379248,
      "balance_loss_mlp": 1.01624608,
      "epoch": 0.4271027475500511,
      "flos": 18916556768640.0,
      "grad_norm": 1.9674791835302412,
      "language_loss": 0.80052876,
      "learning_rate": 2.561766216278735e-06,
      "loss": 0.82371032,
      "num_input_tokens_seen": 76509655,
      "step": 3552,
      "time_per_iteration": 2.780121326446533
    },
    {
      "auxiliary_loss_clip": 0.01396706,
      "auxiliary_loss_mlp": 0.01026419,
      "balance_loss_clip": 1.05531931,
      "balance_loss_mlp": 1.01825285,
      "epoch": 0.4272229904406902,
      "flos": 26870554500480.0,
      "grad_norm": 2.324379885559823,
      "language_loss": 0.81247318,
      "learning_rate": 2.561018561892121e-06,
      "loss": 0.83670449,
      "num_input_tokens_seen": 76528795,
      "step": 3553,
      "time_per_iteration": 3.659587860107422
    },
    {
      "auxiliary_loss_clip": 0.01292031,
      "auxiliary_loss_mlp": 0.01034632,
      "balance_loss_clip": 1.0537591,
      "balance_loss_mlp": 1.02694893,
      "epoch": 0.4273432333313293,
      "flos": 23951376190080.0,
      "grad_norm": 1.9947263611112562,
      "language_loss": 0.76795536,
      "learning_rate": 2.5602708223994363e-06,
      "loss": 0.79122204,
      "num_input_tokens_seen": 76550660,
      "step": 3554,
      "time_per_iteration": 2.8169944286346436
    },
    {
      "auxiliary_loss_clip": 0.01344315,
      "auxiliary_loss_mlp": 0.01025721,
      "balance_loss_clip": 1.04896462,
      "balance_loss_mlp": 1.01838386,
      "epoch": 0.4274634762219684,
      "flos": 29570354496000.0,
      "grad_norm": 2.266426271754617,
      "language_loss": 0.67705476,
      "learning_rate": 2.559522997914115e-06,
      "loss": 0.70075512,
      "num_input_tokens_seen": 76570240,
      "step": 3555,
      "time_per_iteration": 2.7629234790802
    },
    {
      "auxiliary_loss_clip": 0.01194764,
      "auxiliary_loss_mlp": 0.01026932,
      "balance_loss_clip": 1.06098175,
      "balance_loss_mlp": 1.0199163,
      "epoch": 0.42758371911260745,
      "flos": 21434146047360.0,
      "grad_norm": 2.0701687610964714,
      "language_loss": 0.84992635,
      "learning_rate": 2.558775088549599e-06,
      "loss": 0.87214333,
      "num_input_tokens_seen": 76589820,
      "step": 3556,
      "time_per_iteration": 2.6729769706726074
    },
    {
      "auxiliary_loss_clip": 0.01154757,
      "auxiliary_loss_mlp": 0.01032571,
      "balance_loss_clip": 1.0561868,
      "balance_loss_mlp": 1.0241611,
      "epoch": 0.42770396200324656,
      "flos": 14752822072320.0,
      "grad_norm": 2.3433573477305027,
      "language_loss": 0.66219997,
      "learning_rate": 2.5580270944193467e-06,
      "loss": 0.68407321,
      "num_input_tokens_seen": 76606640,
      "step": 3557,
      "time_per_iteration": 2.6235625743865967
    },
    {
      "auxiliary_loss_clip": 0.01090595,
      "auxiliary_loss_mlp": 0.01000382,
      "balance_loss_clip": 1.02603984,
      "balance_loss_mlp": 0.99940491,
      "epoch": 0.4278242048938857,
      "flos": 70654712601600.0,
      "grad_norm": 0.7473540014464652,
      "language_loss": 0.55488926,
      "learning_rate": 2.557279015636827e-06,
      "loss": 0.57579911,
      "num_input_tokens_seen": 76667050,
      "step": 3558,
      "time_per_iteration": 3.2072298526763916
    },
    {
      "auxiliary_loss_clip": 0.01139343,
      "auxiliary_loss_mlp": 0.01002525,
      "balance_loss_clip": 1.02489305,
      "balance_loss_mlp": 1.00150585,
      "epoch": 0.42794444778452473,
      "flos": 69366165033600.0,
      "grad_norm": 0.768860048668422,
      "language_loss": 0.61214817,
      "learning_rate": 2.5565308523155245e-06,
      "loss": 0.63356686,
      "num_input_tokens_seen": 76726650,
      "step": 3559,
      "time_per_iteration": 3.180443286895752
    },
    {
      "auxiliary_loss_clip": 0.01382405,
      "auxiliary_loss_mlp": 0.01030204,
      "balance_loss_clip": 1.05183423,
      "balance_loss_mlp": 1.02234864,
      "epoch": 0.42806469067516384,
      "flos": 18215328481920.0,
      "grad_norm": 3.033121383110696,
      "language_loss": 0.82201719,
      "learning_rate": 2.5557826045689336e-06,
      "loss": 0.84614325,
      "num_input_tokens_seen": 76742890,
      "step": 3560,
      "time_per_iteration": 2.705798625946045
    },
    {
      "auxiliary_loss_clip": 0.01272862,
      "auxiliary_loss_mlp": 0.01004213,
      "balance_loss_clip": 1.04388881,
      "balance_loss_mlp": 1.00318229,
      "epoch": 0.4281849335658029,
      "flos": 54535814432640.0,
      "grad_norm": 0.8308411567922598,
      "language_loss": 0.58800876,
      "learning_rate": 2.5550342725105643e-06,
      "loss": 0.61077952,
      "num_input_tokens_seen": 76801055,
      "step": 3561,
      "time_per_iteration": 3.204542398452759
    },
    {
      "auxiliary_loss_clip": 0.01246304,
      "auxiliary_loss_mlp": 0.01024074,
      "balance_loss_clip": 1.06065798,
      "balance_loss_mlp": 1.01617098,
      "epoch": 0.428305176456442,
      "flos": 17274828723840.0,
      "grad_norm": 1.7546188931214184,
      "language_loss": 0.80925834,
      "learning_rate": 2.554285856253937e-06,
      "loss": 0.83196217,
      "num_input_tokens_seen": 76819890,
      "step": 3562,
      "time_per_iteration": 2.6638576984405518
    },
    {
      "auxiliary_loss_clip": 0.0128681,
      "auxiliary_loss_mlp": 0.01029719,
      "balance_loss_clip": 1.0550617,
      "balance_loss_mlp": 1.02189898,
      "epoch": 0.4284254193470811,
      "flos": 26359509749760.0,
      "grad_norm": 1.9230491824097171,
      "language_loss": 0.77853775,
      "learning_rate": 2.5535373559125855e-06,
      "loss": 0.8017031,
      "num_input_tokens_seen": 76840255,
      "step": 3563,
      "time_per_iteration": 2.777161121368408
    },
    {
      "auxiliary_loss_clip": 0.01495372,
      "auxiliary_loss_mlp": 0.01029191,
      "balance_loss_clip": 1.04932904,
      "balance_loss_mlp": 1.02054238,
      "epoch": 0.42854566223772017,
      "flos": 29714248379520.0,
      "grad_norm": 1.6257580476609694,
      "language_loss": 0.82303089,
      "learning_rate": 2.552788771600057e-06,
      "loss": 0.8482765,
      "num_input_tokens_seen": 76860565,
      "step": 3564,
      "time_per_iteration": 3.0004918575286865
    },
    {
      "auxiliary_loss_clip": 0.01345335,
      "auxiliary_loss_mlp": 0.01039259,
      "balance_loss_clip": 1.05518508,
      "balance_loss_mlp": 1.03132606,
      "epoch": 0.4286659051283593,
      "flos": 22018161277440.0,
      "grad_norm": 2.315677009081112,
      "language_loss": 0.82332397,
      "learning_rate": 2.5520401034299118e-06,
      "loss": 0.84716994,
      "num_input_tokens_seen": 76878325,
      "step": 3565,
      "time_per_iteration": 2.8085553646087646
    },
    {
      "auxiliary_loss_clip": 0.01246376,
      "auxiliary_loss_mlp": 0.01030316,
      "balance_loss_clip": 1.05730081,
      "balance_loss_mlp": 1.02145243,
      "epoch": 0.4287861480189984,
      "flos": 13334422838400.0,
      "grad_norm": 2.1996771887952025,
      "language_loss": 0.87950635,
      "learning_rate": 2.551291351515722e-06,
      "loss": 0.9022733,
      "num_input_tokens_seen": 76895340,
      "step": 3566,
      "time_per_iteration": 2.5949506759643555
    },
    {
      "auxiliary_loss_clip": 0.01340485,
      "auxiliary_loss_mlp": 0.02570751,
      "balance_loss_clip": 1.05006886,
      "balance_loss_mlp": 1.00000691,
      "epoch": 0.42890639090963745,
      "flos": 26651535321600.0,
      "grad_norm": 1.8784400070445777,
      "language_loss": 0.85652399,
      "learning_rate": 2.5505425159710726e-06,
      "loss": 0.89563638,
      "num_input_tokens_seen": 76915150,
      "step": 3567,
      "time_per_iteration": 2.7702322006225586
    },
    {
      "auxiliary_loss_clip": 0.01299989,
      "auxiliary_loss_mlp": 0.02570409,
      "balance_loss_clip": 1.05268931,
      "balance_loss_mlp": 1.00000262,
      "epoch": 0.42902663380027656,
      "flos": 24055768091520.0,
      "grad_norm": 3.127848622508519,
      "language_loss": 0.82758331,
      "learning_rate": 2.549793596909561e-06,
      "loss": 0.86628729,
      "num_input_tokens_seen": 76933770,
      "step": 3568,
      "time_per_iteration": 3.7199530601501465
    },
    {
      "auxiliary_loss_clip": 0.01295814,
      "auxiliary_loss_mlp": 0.01028183,
      "balance_loss_clip": 1.05844581,
      "balance_loss_mlp": 1.02053237,
      "epoch": 0.42914687669091567,
      "flos": 15632561975040.0,
      "grad_norm": 3.4916275386461795,
      "language_loss": 0.66671491,
      "learning_rate": 2.5490445944447976e-06,
      "loss": 0.68995488,
      "num_input_tokens_seen": 76952265,
      "step": 3569,
      "time_per_iteration": 2.708069086074829
    },
    {
      "auxiliary_loss_clip": 0.01241267,
      "auxiliary_loss_mlp": 0.0103119,
      "balance_loss_clip": 1.05464792,
      "balance_loss_mlp": 1.0232451,
      "epoch": 0.4292671195815547,
      "flos": 31467802440960.0,
      "grad_norm": 2.9548041056716023,
      "language_loss": 0.65143079,
      "learning_rate": 2.548295508690406e-06,
      "loss": 0.67415535,
      "num_input_tokens_seen": 76973560,
      "step": 3570,
      "time_per_iteration": 2.70430850982666
    },
    {
      "auxiliary_loss_clip": 0.01248164,
      "auxiliary_loss_mlp": 0.01035875,
      "balance_loss_clip": 1.05547071,
      "balance_loss_mlp": 1.02717876,
      "epoch": 0.42938736247219383,
      "flos": 30257756046720.0,
      "grad_norm": 1.912112274365101,
      "language_loss": 0.76339102,
      "learning_rate": 2.5475463397600217e-06,
      "loss": 0.78623146,
      "num_input_tokens_seen": 76993640,
      "step": 3571,
      "time_per_iteration": 2.771775245666504
    },
    {
      "auxiliary_loss_clip": 0.01195107,
      "auxiliary_loss_mlp": 0.0103172,
      "balance_loss_clip": 1.05662882,
      "balance_loss_mlp": 1.02364016,
      "epoch": 0.42950760536283294,
      "flos": 29349683291520.0,
      "grad_norm": 2.03358795252261,
      "language_loss": 0.77524245,
      "learning_rate": 2.546797087767293e-06,
      "loss": 0.79751074,
      "num_input_tokens_seen": 77013765,
      "step": 3572,
      "time_per_iteration": 3.571986675262451
    },
    {
      "auxiliary_loss_clip": 0.0138744,
      "auxiliary_loss_mlp": 0.01027328,
      "balance_loss_clip": 1.05182791,
      "balance_loss_mlp": 1.02007961,
      "epoch": 0.429627848253472,
      "flos": 26869943969280.0,
      "grad_norm": 1.780893920072732,
      "language_loss": 0.87017858,
      "learning_rate": 2.546047752825881e-06,
      "loss": 0.89432621,
      "num_input_tokens_seen": 77034370,
      "step": 3573,
      "time_per_iteration": 2.744964361190796
    },
    {
      "auxiliary_loss_clip": 0.01392909,
      "auxiliary_loss_mlp": 0.01027765,
      "balance_loss_clip": 1.04789031,
      "balance_loss_mlp": 1.02050853,
      "epoch": 0.4297480911441111,
      "flos": 13881270470400.0,
      "grad_norm": 2.1006968125619294,
      "language_loss": 0.93210018,
      "learning_rate": 2.5452983350494595e-06,
      "loss": 0.95630693,
      "num_input_tokens_seen": 77049925,
      "step": 3574,
      "time_per_iteration": 2.7038044929504395
    },
    {
      "auxiliary_loss_clip": 0.01242646,
      "auxiliary_loss_mlp": 0.02569194,
      "balance_loss_clip": 1.05448759,
      "balance_loss_mlp": 0.99999237,
      "epoch": 0.4298683340347502,
      "flos": 20741141975040.0,
      "grad_norm": 2.460125088345148,
      "language_loss": 0.64920831,
      "learning_rate": 2.544548834551713e-06,
      "loss": 0.68732673,
      "num_input_tokens_seen": 77068930,
      "step": 3575,
      "time_per_iteration": 2.6535279750823975
    },
    {
      "auxiliary_loss_clip": 0.01338062,
      "auxiliary_loss_mlp": 0.02570295,
      "balance_loss_clip": 1.051669,
      "balance_loss_mlp": 1.00001121,
      "epoch": 0.4299885769253893,
      "flos": 20882126856960.0,
      "grad_norm": 3.5564630932949743,
      "language_loss": 0.94711471,
      "learning_rate": 2.5437992514463424e-06,
      "loss": 0.98619831,
      "num_input_tokens_seen": 77082255,
      "step": 3576,
      "time_per_iteration": 2.75175404548645
    },
    {
      "auxiliary_loss_clip": 0.01241935,
      "auxiliary_loss_mlp": 0.01030796,
      "balance_loss_clip": 1.0549624,
      "balance_loss_mlp": 1.02276683,
      "epoch": 0.4301088198160284,
      "flos": 25484618183040.0,
      "grad_norm": 1.7744226391641464,
      "language_loss": 0.88117385,
      "learning_rate": 2.5430495858470565e-06,
      "loss": 0.90390116,
      "num_input_tokens_seen": 77101725,
      "step": 3577,
      "time_per_iteration": 3.530515432357788
    },
    {
      "auxiliary_loss_clip": 0.01239811,
      "auxiliary_loss_mlp": 0.0102917,
      "balance_loss_clip": 1.05519962,
      "balance_loss_mlp": 1.02144527,
      "epoch": 0.43022906270666744,
      "flos": 18259427404800.0,
      "grad_norm": 3.5693149182415365,
      "language_loss": 0.77725315,
      "learning_rate": 2.54229983786758e-06,
      "loss": 0.79994297,
      "num_input_tokens_seen": 77119670,
      "step": 3578,
      "time_per_iteration": 2.635326623916626
    },
    {
      "auxiliary_loss_clip": 0.01295603,
      "auxiliary_loss_mlp": 0.01033119,
      "balance_loss_clip": 1.05277228,
      "balance_loss_mlp": 1.02541184,
      "epoch": 0.43034930559730655,
      "flos": 23399536567680.0,
      "grad_norm": 2.3768172865900765,
      "language_loss": 0.85085183,
      "learning_rate": 2.541550007621651e-06,
      "loss": 0.87413907,
      "num_input_tokens_seen": 77138160,
      "step": 3579,
      "time_per_iteration": 3.6216514110565186
    },
    {
      "auxiliary_loss_clip": 0.01240587,
      "auxiliary_loss_mlp": 0.01031794,
      "balance_loss_clip": 1.05714202,
      "balance_loss_mlp": 1.0236814,
      "epoch": 0.43046954848794566,
      "flos": 28184382264960.0,
      "grad_norm": 1.7799199470541633,
      "language_loss": 0.79562271,
      "learning_rate": 2.5408000952230156e-06,
      "loss": 0.81834656,
      "num_input_tokens_seen": 77156950,
      "step": 3580,
      "time_per_iteration": 2.6858022212982178
    },
    {
      "auxiliary_loss_clip": 0.01257112,
      "auxiliary_loss_mlp": 0.01033124,
      "balance_loss_clip": 1.05297065,
      "balance_loss_mlp": 1.0245471,
      "epoch": 0.4305897913785847,
      "flos": 28580476515840.0,
      "grad_norm": 2.661135266159236,
      "language_loss": 0.90802622,
      "learning_rate": 2.5400501007854357e-06,
      "loss": 0.93092853,
      "num_input_tokens_seen": 77176395,
      "step": 3581,
      "time_per_iteration": 2.792750120162964
    },
    {
      "auxiliary_loss_clip": 0.01390454,
      "auxiliary_loss_mlp": 0.01030005,
      "balance_loss_clip": 1.04789948,
      "balance_loss_mlp": 1.02279305,
      "epoch": 0.43071003426922383,
      "flos": 20448721353600.0,
      "grad_norm": 2.265690662587956,
      "language_loss": 0.75883347,
      "learning_rate": 2.539300024422685e-06,
      "loss": 0.78303808,
      "num_input_tokens_seen": 77194340,
      "step": 3582,
      "time_per_iteration": 2.710742473602295
    },
    {
      "auxiliary_loss_clip": 0.01142882,
      "auxiliary_loss_mlp": 0.0100747,
      "balance_loss_clip": 1.01827943,
      "balance_loss_mlp": 1.00647438,
      "epoch": 0.43083027715986294,
      "flos": 51997969883520.0,
      "grad_norm": 0.7900334355245108,
      "language_loss": 0.60900342,
      "learning_rate": 2.538549866248549e-06,
      "loss": 0.63050699,
      "num_input_tokens_seen": 77249320,
      "step": 3583,
      "time_per_iteration": 3.137239694595337
    },
    {
      "auxiliary_loss_clip": 0.0124247,
      "auxiliary_loss_mlp": 0.01024725,
      "balance_loss_clip": 1.05470502,
      "balance_loss_mlp": 1.01649904,
      "epoch": 0.430950520050502,
      "flos": 16690885320960.0,
      "grad_norm": 1.8993112083641563,
      "language_loss": 0.81144607,
      "learning_rate": 2.5377996263768274e-06,
      "loss": 0.83411801,
      "num_input_tokens_seen": 77267400,
      "step": 3584,
      "time_per_iteration": 2.583667278289795
    },
    {
      "auxiliary_loss_clip": 0.01240363,
      "auxiliary_loss_mlp": 0.01031262,
      "balance_loss_clip": 1.05289495,
      "balance_loss_mlp": 1.02335858,
      "epoch": 0.4310707629411411,
      "flos": 24608433726720.0,
      "grad_norm": 1.8824157123991863,
      "language_loss": 0.68417001,
      "learning_rate": 2.5370493049213293e-06,
      "loss": 0.70688629,
      "num_input_tokens_seen": 77287045,
      "step": 3585,
      "time_per_iteration": 2.7586658000946045
    },
    {
      "auxiliary_loss_clip": 0.01529355,
      "auxiliary_loss_mlp": 0.01025333,
      "balance_loss_clip": 1.0461638,
      "balance_loss_mlp": 1.01732206,
      "epoch": 0.4311910058317802,
      "flos": 26432983019520.0,
      "grad_norm": 2.1720308980229377,
      "language_loss": 0.79835868,
      "learning_rate": 2.536298901995878e-06,
      "loss": 0.82390547,
      "num_input_tokens_seen": 77306255,
      "step": 3586,
      "time_per_iteration": 3.0233190059661865
    },
    {
      "auxiliary_loss_clip": 0.01293489,
      "auxiliary_loss_mlp": 0.01030726,
      "balance_loss_clip": 1.05288196,
      "balance_loss_mlp": 1.02257216,
      "epoch": 0.43131124872241927,
      "flos": 25155891889920.0,
      "grad_norm": 1.7635418619708798,
      "language_loss": 0.80211282,
      "learning_rate": 2.535548417714311e-06,
      "loss": 0.82535499,
      "num_input_tokens_seen": 77325555,
      "step": 3587,
      "time_per_iteration": 2.9549143314361572
    },
    {
      "auxiliary_loss_clip": 0.01149977,
      "auxiliary_loss_mlp": 0.01026013,
      "balance_loss_clip": 1.05243182,
      "balance_loss_mlp": 1.01881313,
      "epoch": 0.4314314916130584,
      "flos": 21614812479360.0,
      "grad_norm": 1.6330469918827237,
      "language_loss": 0.87451494,
      "learning_rate": 2.534797852190474e-06,
      "loss": 0.89627481,
      "num_input_tokens_seen": 77345735,
      "step": 3588,
      "time_per_iteration": 2.6934356689453125
    },
    {
      "auxiliary_loss_clip": 0.0124257,
      "auxiliary_loss_mlp": 0.01033316,
      "balance_loss_clip": 1.05428982,
      "balance_loss_mlp": 1.02579379,
      "epoch": 0.4315517345036975,
      "flos": 19275016544640.0,
      "grad_norm": 1.850993931918705,
      "language_loss": 0.81556547,
      "learning_rate": 2.5340472055382283e-06,
      "loss": 0.83832431,
      "num_input_tokens_seen": 77361765,
      "step": 3589,
      "time_per_iteration": 2.755305528640747
    },
    {
      "auxiliary_loss_clip": 0.01343848,
      "auxiliary_loss_mlp": 0.01026337,
      "balance_loss_clip": 1.04861331,
      "balance_loss_mlp": 1.01864839,
      "epoch": 0.43167197739433655,
      "flos": 24273853516800.0,
      "grad_norm": 2.1415975395444273,
      "language_loss": 0.81018138,
      "learning_rate": 2.5332964778714468e-06,
      "loss": 0.83388323,
      "num_input_tokens_seen": 77378950,
      "step": 3590,
      "time_per_iteration": 2.811648368835449
    },
    {
      "auxiliary_loss_clip": 0.01337526,
      "auxiliary_loss_mlp": 0.01029241,
      "balance_loss_clip": 1.05323923,
      "balance_loss_mlp": 1.02155209,
      "epoch": 0.43179222028497566,
      "flos": 16867816738560.0,
      "grad_norm": 1.7102871796397119,
      "language_loss": 0.66320336,
      "learning_rate": 2.5325456693040123e-06,
      "loss": 0.68687099,
      "num_input_tokens_seen": 77396145,
      "step": 3591,
      "time_per_iteration": 2.6743550300598145
    },
    {
      "auxiliary_loss_clip": 0.01154895,
      "auxiliary_loss_mlp": 0.01034173,
      "balance_loss_clip": 1.05504107,
      "balance_loss_mlp": 1.02541077,
      "epoch": 0.43191246317561477,
      "flos": 17639214243840.0,
      "grad_norm": 9.69028263982647,
      "language_loss": 0.74740815,
      "learning_rate": 2.531794779949824e-06,
      "loss": 0.76929879,
      "num_input_tokens_seen": 77414045,
      "step": 3592,
      "time_per_iteration": 2.6792287826538086
    },
    {
      "auxiliary_loss_clip": 0.01333214,
      "auxiliary_loss_mlp": 0.01029084,
      "balance_loss_clip": 1.04881895,
      "balance_loss_mlp": 1.02191377,
      "epoch": 0.4320327060662538,
      "flos": 23878800760320.0,
      "grad_norm": 1.8523865754337328,
      "language_loss": 0.87996209,
      "learning_rate": 2.5310438099227903e-06,
      "loss": 0.90358508,
      "num_input_tokens_seen": 77431310,
      "step": 3593,
      "time_per_iteration": 2.860227584838867
    },
    {
      "auxiliary_loss_clip": 0.01133563,
      "auxiliary_loss_mlp": 0.01001135,
      "balance_loss_clip": 1.01718211,
      "balance_loss_mlp": 1.00014555,
      "epoch": 0.43215294895689293,
      "flos": 66394917959040.0,
      "grad_norm": 0.8290640411962298,
      "language_loss": 0.5334813,
      "learning_rate": 2.530292759336833e-06,
      "loss": 0.55482829,
      "num_input_tokens_seen": 77492045,
      "step": 3594,
      "time_per_iteration": 3.3653109073638916
    },
    {
      "auxiliary_loss_clip": 0.01293295,
      "auxiliary_loss_mlp": 0.01034321,
      "balance_loss_clip": 1.05447459,
      "balance_loss_mlp": 1.02607131,
      "epoch": 0.432273191847532,
      "flos": 20594267262720.0,
      "grad_norm": 14.952921790676909,
      "language_loss": 0.69639206,
      "learning_rate": 2.5295416283058855e-06,
      "loss": 0.71966827,
      "num_input_tokens_seen": 77510910,
      "step": 3595,
      "time_per_iteration": 4.447504281997681
    },
    {
      "auxiliary_loss_clip": 0.01290613,
      "auxiliary_loss_mlp": 0.02568062,
      "balance_loss_clip": 1.05196655,
      "balance_loss_mlp": 1.00007296,
      "epoch": 0.4323934347381711,
      "flos": 19282127437440.0,
      "grad_norm": 1.5641422205372977,
      "language_loss": 0.66200602,
      "learning_rate": 2.5287904169438943e-06,
      "loss": 0.7005927,
      "num_input_tokens_seen": 77530115,
      "step": 3596,
      "time_per_iteration": 2.6793129444122314
    },
    {
      "auxiliary_loss_clip": 0.01494587,
      "auxiliary_loss_mlp": 0.01029799,
      "balance_loss_clip": 1.04786921,
      "balance_loss_mlp": 1.02169824,
      "epoch": 0.4325136776288102,
      "flos": 21726315273600.0,
      "grad_norm": 2.9997293001812775,
      "language_loss": 0.6419096,
      "learning_rate": 2.528039125364817e-06,
      "loss": 0.66715348,
      "num_input_tokens_seen": 77548920,
      "step": 3597,
      "time_per_iteration": 3.8477768898010254
    },
    {
      "auxiliary_loss_clip": 0.01340983,
      "auxiliary_loss_mlp": 0.01028044,
      "balance_loss_clip": 1.05060482,
      "balance_loss_mlp": 1.01968741,
      "epoch": 0.43263392051944927,
      "flos": 22340746344960.0,
      "grad_norm": 2.246251639864913,
      "language_loss": 0.75869066,
      "learning_rate": 2.5272877536826246e-06,
      "loss": 0.78238094,
      "num_input_tokens_seen": 77567715,
      "step": 3598,
      "time_per_iteration": 2.7765753269195557
    },
    {
      "auxiliary_loss_clip": 0.01389218,
      "auxiliary_loss_mlp": 0.01029379,
      "balance_loss_clip": 1.04577994,
      "balance_loss_mlp": 1.02172601,
      "epoch": 0.4327541634100884,
      "flos": 29168406328320.0,
      "grad_norm": 2.3784720656511813,
      "language_loss": 0.70493144,
      "learning_rate": 2.5265363020112986e-06,
      "loss": 0.72911739,
      "num_input_tokens_seen": 77588035,
      "step": 3599,
      "time_per_iteration": 2.893726110458374
    },
    {
      "auxiliary_loss_clip": 0.01241188,
      "auxiliary_loss_mlp": 0.0102955,
      "balance_loss_clip": 1.05521214,
      "balance_loss_mlp": 1.02145553,
      "epoch": 0.4328744063007275,
      "flos": 26067448264320.0,
      "grad_norm": 1.9873171639659757,
      "language_loss": 0.8401683,
      "learning_rate": 2.5257847704648344e-06,
      "loss": 0.8628757,
      "num_input_tokens_seen": 77609265,
      "step": 3600,
      "time_per_iteration": 2.739973545074463
    },
    {
      "auxiliary_loss_clip": 0.01190616,
      "auxiliary_loss_mlp": 0.01028425,
      "balance_loss_clip": 1.05433631,
      "balance_loss_mlp": 1.02054524,
      "epoch": 0.43299464919136654,
      "flos": 16581357774720.0,
      "grad_norm": 3.43082209638467,
      "language_loss": 0.75661325,
      "learning_rate": 2.525033159157239e-06,
      "loss": 0.77880371,
      "num_input_tokens_seen": 77625580,
      "step": 3601,
      "time_per_iteration": 2.5428669452667236
    },
    {
      "auxiliary_loss_clip": 0.01241774,
      "auxiliary_loss_mlp": 0.01036734,
      "balance_loss_clip": 1.05187821,
      "balance_loss_mlp": 1.02874708,
      "epoch": 0.43311489208200565,
      "flos": 16107265140480.0,
      "grad_norm": 1.874280818922286,
      "language_loss": 0.76558959,
      "learning_rate": 2.52428146820253e-06,
      "loss": 0.78837466,
      "num_input_tokens_seen": 77643835,
      "step": 3602,
      "time_per_iteration": 3.617720365524292
    },
    {
      "auxiliary_loss_clip": 0.01338947,
      "auxiliary_loss_mlp": 0.01030212,
      "balance_loss_clip": 1.05124879,
      "balance_loss_mlp": 1.02197504,
      "epoch": 0.43323513497264476,
      "flos": 22930220442240.0,
      "grad_norm": 1.8196948739087184,
      "language_loss": 0.82072461,
      "learning_rate": 2.52352969771474e-06,
      "loss": 0.8444162,
      "num_input_tokens_seen": 77663060,
      "step": 3603,
      "time_per_iteration": 2.7816038131713867
    },
    {
      "auxiliary_loss_clip": 0.01292657,
      "auxiliary_loss_mlp": 0.01029788,
      "balance_loss_clip": 1.05231667,
      "balance_loss_mlp": 1.02239108,
      "epoch": 0.4333553778632838,
      "flos": 25299031587840.0,
      "grad_norm": 2.6960873155031555,
      "language_loss": 0.88483429,
      "learning_rate": 2.5227778478079106e-06,
      "loss": 0.90805876,
      "num_input_tokens_seen": 77682470,
      "step": 3604,
      "time_per_iteration": 2.6740357875823975
    },
    {
      "auxiliary_loss_clip": 0.01238165,
      "auxiliary_loss_mlp": 0.01026263,
      "balance_loss_clip": 1.05163169,
      "balance_loss_mlp": 1.01874971,
      "epoch": 0.43347562075392293,
      "flos": 19387165783680.0,
      "grad_norm": 1.6665002503545883,
      "language_loss": 0.76613605,
      "learning_rate": 2.522025918596098e-06,
      "loss": 0.78878033,
      "num_input_tokens_seen": 77700770,
      "step": 3605,
      "time_per_iteration": 3.720994234085083
    },
    {
      "auxiliary_loss_clip": 0.01147625,
      "auxiliary_loss_mlp": 0.01032137,
      "balance_loss_clip": 1.05471706,
      "balance_loss_mlp": 1.02474654,
      "epoch": 0.43359586364456204,
      "flos": 26325969425280.0,
      "grad_norm": 1.5881193457736333,
      "language_loss": 0.65508318,
      "learning_rate": 2.521273910193368e-06,
      "loss": 0.67688084,
      "num_input_tokens_seen": 77723950,
      "step": 3606,
      "time_per_iteration": 2.7395517826080322
    },
    {
      "auxiliary_loss_clip": 0.01248377,
      "auxiliary_loss_mlp": 0.01036082,
      "balance_loss_clip": 1.05558491,
      "balance_loss_mlp": 1.02721262,
      "epoch": 0.4337161065352011,
      "flos": 15989261984640.0,
      "grad_norm": 2.0415932757078474,
      "language_loss": 0.8719849,
      "learning_rate": 2.5205218227138006e-06,
      "loss": 0.89482945,
      "num_input_tokens_seen": 77736905,
      "step": 3607,
      "time_per_iteration": 2.6291465759277344
    },
    {
      "auxiliary_loss_clip": 0.01190595,
      "auxiliary_loss_mlp": 0.01023918,
      "balance_loss_clip": 1.05354416,
      "balance_loss_mlp": 1.01556742,
      "epoch": 0.4338363494258402,
      "flos": 20224710184320.0,
      "grad_norm": 2.0502764172025065,
      "language_loss": 0.79223502,
      "learning_rate": 2.519769656271486e-06,
      "loss": 0.81438017,
      "num_input_tokens_seen": 77754325,
      "step": 3608,
      "time_per_iteration": 2.6317641735076904
    },
    {
      "auxiliary_loss_clip": 0.01381459,
      "auxiliary_loss_mlp": 0.0102927,
      "balance_loss_clip": 1.04880881,
      "balance_loss_mlp": 1.02126837,
      "epoch": 0.43395659231647926,
      "flos": 20083904870400.0,
      "grad_norm": 3.1248161950213347,
      "language_loss": 0.68024945,
      "learning_rate": 2.5190174109805285e-06,
      "loss": 0.70435667,
      "num_input_tokens_seen": 77774150,
      "step": 3609,
      "time_per_iteration": 2.6972553730010986
    },
    {
      "auxiliary_loss_clip": 0.01292437,
      "auxiliary_loss_mlp": 0.0103369,
      "balance_loss_clip": 1.05473685,
      "balance_loss_mlp": 1.02542245,
      "epoch": 0.43407683520711837,
      "flos": 19901801894400.0,
      "grad_norm": 3.124676244333142,
      "language_loss": 0.63684702,
      "learning_rate": 2.518265086955042e-06,
      "loss": 0.66010833,
      "num_input_tokens_seen": 77791870,
      "step": 3610,
      "time_per_iteration": 2.7032532691955566
    },
    {
      "auxiliary_loss_clip": 0.01192685,
      "auxiliary_loss_mlp": 0.01033582,
      "balance_loss_clip": 1.0544138,
      "balance_loss_mlp": 1.02590466,
      "epoch": 0.4341970780977575,
      "flos": 23108732058240.0,
      "grad_norm": 1.8638113803137257,
      "language_loss": 0.83531141,
      "learning_rate": 2.5175126843091534e-06,
      "loss": 0.85757411,
      "num_input_tokens_seen": 77811240,
      "step": 3611,
      "time_per_iteration": 2.632643699645996
    },
    {
      "auxiliary_loss_clip": 0.0119684,
      "auxiliary_loss_mlp": 0.01026302,
      "balance_loss_clip": 1.05409539,
      "balance_loss_mlp": 1.01867819,
      "epoch": 0.43431732098839654,
      "flos": 37408288406400.0,
      "grad_norm": 2.21269220547321,
      "language_loss": 0.75520551,
      "learning_rate": 2.5167602031570034e-06,
      "loss": 0.77743697,
      "num_input_tokens_seen": 77831425,
      "step": 3612,
      "time_per_iteration": 2.798128366470337
    },
    {
      "auxiliary_loss_clip": 0.01190805,
      "auxiliary_loss_mlp": 0.01026974,
      "balance_loss_clip": 1.05357969,
      "balance_loss_mlp": 1.01930857,
      "epoch": 0.43443756387903565,
      "flos": 31868206323840.0,
      "grad_norm": 1.676458072038454,
      "language_loss": 0.73517108,
      "learning_rate": 2.51600764361274e-06,
      "loss": 0.7573489,
      "num_input_tokens_seen": 77852950,
      "step": 3613,
      "time_per_iteration": 2.6677186489105225
    },
    {
      "auxiliary_loss_clip": 0.01194664,
      "auxiliary_loss_mlp": 0.01031829,
      "balance_loss_clip": 1.05534923,
      "balance_loss_mlp": 1.02378845,
      "epoch": 0.43455780676967476,
      "flos": 23477139901440.0,
      "grad_norm": 2.8860735270827207,
      "language_loss": 0.78691703,
      "learning_rate": 2.5152550057905283e-06,
      "loss": 0.80918193,
      "num_input_tokens_seen": 77872840,
      "step": 3614,
      "time_per_iteration": 2.6827006340026855
    },
    {
      "auxiliary_loss_clip": 0.01244882,
      "auxiliary_loss_mlp": 0.025741,
      "balance_loss_clip": 1.05687308,
      "balance_loss_mlp": 1.00007641,
      "epoch": 0.4346780496603138,
      "flos": 24207060176640.0,
      "grad_norm": 2.7138049623920337,
      "language_loss": 0.77619267,
      "learning_rate": 2.5145022898045415e-06,
      "loss": 0.81438249,
      "num_input_tokens_seen": 77892025,
      "step": 3615,
      "time_per_iteration": 2.636960744857788
    },
    {
      "auxiliary_loss_clip": 0.01200049,
      "auxiliary_loss_mlp": 0.01028371,
      "balance_loss_clip": 1.05025852,
      "balance_loss_mlp": 1.02036881,
      "epoch": 0.4347982925509529,
      "flos": 17092366611840.0,
      "grad_norm": 2.4450813421506936,
      "language_loss": 0.89908433,
      "learning_rate": 2.5137494957689664e-06,
      "loss": 0.92136854,
      "num_input_tokens_seen": 77907635,
      "step": 3616,
      "time_per_iteration": 2.7191097736358643
    },
    {
      "auxiliary_loss_clip": 0.01191167,
      "auxiliary_loss_mlp": 0.00999812,
      "balance_loss_clip": 1.02324796,
      "balance_loss_mlp": 0.99863786,
      "epoch": 0.43491853544159204,
      "flos": 60945544696320.0,
      "grad_norm": 0.7650114423034884,
      "language_loss": 0.57298207,
      "learning_rate": 2.5129966237980016e-06,
      "loss": 0.59489191,
      "num_input_tokens_seen": 77970630,
      "step": 3617,
      "time_per_iteration": 3.2720253467559814
    },
    {
      "auxiliary_loss_clip": 0.01340087,
      "auxiliary_loss_mlp": 0.01034748,
      "balance_loss_clip": 1.04953003,
      "balance_loss_mlp": 1.02682042,
      "epoch": 0.4350387783322311,
      "flos": 21944652094080.0,
      "grad_norm": 1.8513623679821196,
      "language_loss": 0.78021407,
      "learning_rate": 2.512243674005857e-06,
      "loss": 0.80396241,
      "num_input_tokens_seen": 77989995,
      "step": 3618,
      "time_per_iteration": 2.7402632236480713
    },
    {
      "auxiliary_loss_clip": 0.01438711,
      "auxiliary_loss_mlp": 0.01029012,
      "balance_loss_clip": 1.04444861,
      "balance_loss_mlp": 1.02120042,
      "epoch": 0.4351590212228702,
      "flos": 25082705928960.0,
      "grad_norm": 1.9874637486426991,
      "language_loss": 0.86295724,
      "learning_rate": 2.5114906465067537e-06,
      "loss": 0.8876344,
      "num_input_tokens_seen": 78010980,
      "step": 3619,
      "time_per_iteration": 2.8907909393310547
    },
    {
      "auxiliary_loss_clip": 0.01241426,
      "auxiliary_loss_mlp": 0.01026959,
      "balance_loss_clip": 1.05144572,
      "balance_loss_mlp": 1.01917422,
      "epoch": 0.4352792641135093,
      "flos": 21506541909120.0,
      "grad_norm": 7.144919884638316,
      "language_loss": 0.74966246,
      "learning_rate": 2.5107375414149264e-06,
      "loss": 0.77234626,
      "num_input_tokens_seen": 78030225,
      "step": 3620,
      "time_per_iteration": 2.663081645965576
    },
    {
      "auxiliary_loss_clip": 0.01381836,
      "auxiliary_loss_mlp": 0.01027011,
      "balance_loss_clip": 1.04357719,
      "balance_loss_mlp": 1.01855886,
      "epoch": 0.43539950700414837,
      "flos": 16253457494400.0,
      "grad_norm": 2.345727007557022,
      "language_loss": 0.71985662,
      "learning_rate": 2.5099843588446197e-06,
      "loss": 0.74394506,
      "num_input_tokens_seen": 78048545,
      "step": 3621,
      "time_per_iteration": 4.2744810581207275
    },
    {
      "auxiliary_loss_clip": 0.01294539,
      "auxiliary_loss_mlp": 0.01029337,
      "balance_loss_clip": 1.04914558,
      "balance_loss_mlp": 1.0212549,
      "epoch": 0.4355197498947875,
      "flos": 16691819074560.0,
      "grad_norm": 1.6109583145395872,
      "language_loss": 0.61530256,
      "learning_rate": 2.509231098910091e-06,
      "loss": 0.63854134,
      "num_input_tokens_seen": 78068415,
      "step": 3622,
      "time_per_iteration": 2.827373504638672
    },
    {
      "auxiliary_loss_clip": 0.01299672,
      "auxiliary_loss_mlp": 0.01034771,
      "balance_loss_clip": 1.06062663,
      "balance_loss_mlp": 1.02624726,
      "epoch": 0.4356399927854266,
      "flos": 16362733645440.0,
      "grad_norm": 4.922012104101964,
      "language_loss": 0.74836552,
      "learning_rate": 2.508477761725611e-06,
      "loss": 0.77171004,
      "num_input_tokens_seen": 78086690,
      "step": 3623,
      "time_per_iteration": 2.73087215423584
    },
    {
      "auxiliary_loss_clip": 0.01247107,
      "auxiliary_loss_mlp": 0.0102837,
      "balance_loss_clip": 1.05420721,
      "balance_loss_mlp": 1.02018619,
      "epoch": 0.43576023567606564,
      "flos": 17202037812480.0,
      "grad_norm": 1.933452134185116,
      "language_loss": 0.81131309,
      "learning_rate": 2.507724347405458e-06,
      "loss": 0.83406782,
      "num_input_tokens_seen": 78104640,
      "step": 3624,
      "time_per_iteration": 3.55426025390625
    },
    {
      "auxiliary_loss_clip": 0.01386269,
      "auxiliary_loss_mlp": 0.01024798,
      "balance_loss_clip": 1.04603243,
      "balance_loss_mlp": 1.01703799,
      "epoch": 0.43588047856670475,
      "flos": 15917656222080.0,
      "grad_norm": 2.312905227365521,
      "language_loss": 0.81924099,
      "learning_rate": 2.5069708560639243e-06,
      "loss": 0.84335166,
      "num_input_tokens_seen": 78122550,
      "step": 3625,
      "time_per_iteration": 2.763002634048462
    },
    {
      "auxiliary_loss_clip": 0.01342296,
      "auxiliary_loss_mlp": 0.01029719,
      "balance_loss_clip": 1.05032814,
      "balance_loss_mlp": 1.02176738,
      "epoch": 0.4360007214573438,
      "flos": 23659566099840.0,
      "grad_norm": 2.146308433032509,
      "language_loss": 0.61625993,
      "learning_rate": 2.5062172878153158e-06,
      "loss": 0.63998008,
      "num_input_tokens_seen": 78141825,
      "step": 3626,
      "time_per_iteration": 2.8498942852020264
    },
    {
      "auxiliary_loss_clip": 0.01349227,
      "auxiliary_loss_mlp": 0.01030105,
      "balance_loss_clip": 1.04864907,
      "balance_loss_mlp": 1.02184927,
      "epoch": 0.4361209643479829,
      "flos": 21978767036160.0,
      "grad_norm": 2.1182757923338453,
      "language_loss": 0.87684572,
      "learning_rate": 2.505463642773947e-06,
      "loss": 0.900639,
      "num_input_tokens_seen": 78161790,
      "step": 3627,
      "time_per_iteration": 2.879458427429199
    },
    {
      "auxiliary_loss_clip": 0.01343888,
      "auxiliary_loss_mlp": 0.02574847,
      "balance_loss_clip": 1.05242205,
      "balance_loss_mlp": 1.00013793,
      "epoch": 0.43624120723862203,
      "flos": 17420159151360.0,
      "grad_norm": 3.216625369964374,
      "language_loss": 0.75334132,
      "learning_rate": 2.504709921054146e-06,
      "loss": 0.79252863,
      "num_input_tokens_seen": 78178605,
      "step": 3628,
      "time_per_iteration": 3.5762667655944824
    },
    {
      "auxiliary_loss_clip": 0.01339566,
      "auxiliary_loss_mlp": 0.01033968,
      "balance_loss_clip": 1.04541063,
      "balance_loss_mlp": 1.02565956,
      "epoch": 0.4363614501292611,
      "flos": 17895293280000.0,
      "grad_norm": 2.2298991511003967,
      "language_loss": 0.83816218,
      "learning_rate": 2.50395612277025e-06,
      "loss": 0.86189747,
      "num_input_tokens_seen": 78194460,
      "step": 3629,
      "time_per_iteration": 2.814850330352783
    },
    {
      "auxiliary_loss_clip": 0.01295758,
      "auxiliary_loss_mlp": 0.01031986,
      "balance_loss_clip": 1.05116487,
      "balance_loss_mlp": 1.02440441,
      "epoch": 0.4364816930199002,
      "flos": 20302888135680.0,
      "grad_norm": 2.307418920225434,
      "language_loss": 0.73304403,
      "learning_rate": 2.503202248036612e-06,
      "loss": 0.75632155,
      "num_input_tokens_seen": 78213315,
      "step": 3630,
      "time_per_iteration": 2.7428195476531982
    },
    {
      "auxiliary_loss_clip": 0.01191828,
      "auxiliary_loss_mlp": 0.01028653,
      "balance_loss_clip": 1.05563748,
      "balance_loss_mlp": 1.02077913,
      "epoch": 0.4366019359105393,
      "flos": 24061334699520.0,
      "grad_norm": 1.795330105084189,
      "language_loss": 0.73391038,
      "learning_rate": 2.5024482969675927e-06,
      "loss": 0.7561152,
      "num_input_tokens_seen": 78233270,
      "step": 3631,
      "time_per_iteration": 3.567664861679077
    },
    {
      "auxiliary_loss_clip": 0.01297147,
      "auxiliary_loss_mlp": 0.01025234,
      "balance_loss_clip": 1.04730368,
      "balance_loss_mlp": 1.01779246,
      "epoch": 0.43672217880117836,
      "flos": 21754109422080.0,
      "grad_norm": 2.367792406696381,
      "language_loss": 0.84377515,
      "learning_rate": 2.501694269677566e-06,
      "loss": 0.86699903,
      "num_input_tokens_seen": 78251040,
      "step": 3632,
      "time_per_iteration": 2.828904628753662
    },
    {
      "auxiliary_loss_clip": 0.01246403,
      "auxiliary_loss_mlp": 0.01027769,
      "balance_loss_clip": 1.05303621,
      "balance_loss_mlp": 1.01963282,
      "epoch": 0.4368424216918175,
      "flos": 18035200753920.0,
      "grad_norm": 2.001875209734955,
      "language_loss": 0.80396557,
      "learning_rate": 2.500940166280918e-06,
      "loss": 0.82670724,
      "num_input_tokens_seen": 78269470,
      "step": 3633,
      "time_per_iteration": 2.642428159713745
    },
    {
      "auxiliary_loss_clip": 0.01241045,
      "auxiliary_loss_mlp": 0.01026841,
      "balance_loss_clip": 1.05320048,
      "balance_loss_mlp": 1.01924753,
      "epoch": 0.4369626645824566,
      "flos": 25447127362560.0,
      "grad_norm": 1.9815927877909907,
      "language_loss": 0.79090714,
      "learning_rate": 2.500185986892045e-06,
      "loss": 0.813586,
      "num_input_tokens_seen": 78288955,
      "step": 3634,
      "time_per_iteration": 2.753901243209839
    },
    {
      "auxiliary_loss_clip": 0.01240785,
      "auxiliary_loss_mlp": 0.01030486,
      "balance_loss_clip": 1.05135584,
      "balance_loss_mlp": 1.02215934,
      "epoch": 0.43708290747309564,
      "flos": 25302694775040.0,
      "grad_norm": 2.120155099575838,
      "language_loss": 0.77652025,
      "learning_rate": 2.499431731625355e-06,
      "loss": 0.7992329,
      "num_input_tokens_seen": 78307980,
      "step": 3635,
      "time_per_iteration": 2.6922478675842285
    },
    {
      "auxiliary_loss_clip": 0.01196396,
      "auxiliary_loss_mlp": 0.0102873,
      "balance_loss_clip": 1.05610311,
      "balance_loss_mlp": 1.02066588,
      "epoch": 0.43720315036373475,
      "flos": 31575103344000.0,
      "grad_norm": 1.8491617082904117,
      "language_loss": 0.79402196,
      "learning_rate": 2.4986774005952686e-06,
      "loss": 0.81627333,
      "num_input_tokens_seen": 78330355,
      "step": 3636,
      "time_per_iteration": 2.7138524055480957
    },
    {
      "auxiliary_loss_clip": 0.01246559,
      "auxiliary_loss_mlp": 0.01035195,
      "balance_loss_clip": 1.05756688,
      "balance_loss_mlp": 1.02708864,
      "epoch": 0.43732339325437386,
      "flos": 23112000195840.0,
      "grad_norm": 2.4967405997775884,
      "language_loss": 0.84628278,
      "learning_rate": 2.4979229939162166e-06,
      "loss": 0.86910033,
      "num_input_tokens_seen": 78349135,
      "step": 3637,
      "time_per_iteration": 2.660477876663208
    },
    {
      "auxiliary_loss_clip": 0.01238013,
      "auxiliary_loss_mlp": 0.01036787,
      "balance_loss_clip": 1.05377555,
      "balance_loss_mlp": 1.02913988,
      "epoch": 0.4374436361450129,
      "flos": 27746272080000.0,
      "grad_norm": 1.6867878510196712,
      "language_loss": 0.80646408,
      "learning_rate": 2.4971685117026433e-06,
      "loss": 0.82921207,
      "num_input_tokens_seen": 78368900,
      "step": 3638,
      "time_per_iteration": 2.7222275733947754
    },
    {
      "auxiliary_loss_clip": 0.01245753,
      "auxiliary_loss_mlp": 0.01027011,
      "balance_loss_clip": 1.05547547,
      "balance_loss_mlp": 1.01898873,
      "epoch": 0.437563879035652,
      "flos": 24172370616960.0,
      "grad_norm": 2.264335382624015,
      "language_loss": 0.76747549,
      "learning_rate": 2.4964139540690018e-06,
      "loss": 0.79020309,
      "num_input_tokens_seen": 78392235,
      "step": 3639,
      "time_per_iteration": 2.700502634048462
    },
    {
      "auxiliary_loss_clip": 0.01343106,
      "auxiliary_loss_mlp": 0.01032995,
      "balance_loss_clip": 1.05207634,
      "balance_loss_mlp": 1.02408409,
      "epoch": 0.4376841219262911,
      "flos": 23477211728640.0,
      "grad_norm": 1.9039227832274725,
      "language_loss": 0.72678655,
      "learning_rate": 2.495659321129758e-06,
      "loss": 0.75054753,
      "num_input_tokens_seen": 78409980,
      "step": 3640,
      "time_per_iteration": 2.798823356628418
    },
    {
      "auxiliary_loss_clip": 0.01240436,
      "auxiliary_loss_mlp": 0.01027025,
      "balance_loss_clip": 1.05127108,
      "balance_loss_mlp": 1.01938367,
      "epoch": 0.4378043648169302,
      "flos": 25447809720960.0,
      "grad_norm": 3.896935195436091,
      "language_loss": 0.7554028,
      "learning_rate": 2.494904612999389e-06,
      "loss": 0.77807742,
      "num_input_tokens_seen": 78428690,
      "step": 3641,
      "time_per_iteration": 2.7444708347320557
    },
    {
      "auxiliary_loss_clip": 0.01130302,
      "auxiliary_loss_mlp": 0.01001747,
      "balance_loss_clip": 1.01770616,
      "balance_loss_mlp": 1.00076318,
      "epoch": 0.4379246077075693,
      "flos": 53914056986880.0,
      "grad_norm": 0.7725913654156559,
      "language_loss": 0.56476879,
      "learning_rate": 2.4941498297923843e-06,
      "loss": 0.58608925,
      "num_input_tokens_seen": 78489260,
      "step": 3642,
      "time_per_iteration": 3.1892731189727783
    },
    {
      "auxiliary_loss_clip": 0.01239828,
      "auxiliary_loss_mlp": 0.01026698,
      "balance_loss_clip": 1.05492544,
      "balance_loss_mlp": 1.01912808,
      "epoch": 0.43804485059820836,
      "flos": 20588305605120.0,
      "grad_norm": 1.743196922355191,
      "language_loss": 0.70010269,
      "learning_rate": 2.4933949716232424e-06,
      "loss": 0.72276795,
      "num_input_tokens_seen": 78506785,
      "step": 3643,
      "time_per_iteration": 2.729036569595337
    },
    {
      "auxiliary_loss_clip": 0.01249989,
      "auxiliary_loss_mlp": 0.01034384,
      "balance_loss_clip": 1.05666208,
      "balance_loss_mlp": 1.02568185,
      "epoch": 0.43816509348884747,
      "flos": 23876214981120.0,
      "grad_norm": 2.2643169847666216,
      "language_loss": 0.73676264,
      "learning_rate": 2.492640038606476e-06,
      "loss": 0.75960642,
      "num_input_tokens_seen": 78525150,
      "step": 3644,
      "time_per_iteration": 2.835496425628662
    },
    {
      "auxiliary_loss_clip": 0.01241819,
      "auxiliary_loss_mlp": 0.01027351,
      "balance_loss_clip": 1.05287921,
      "balance_loss_mlp": 1.01911986,
      "epoch": 0.4382853363794866,
      "flos": 14684448533760.0,
      "grad_norm": 2.384688142173742,
      "language_loss": 0.79066879,
      "learning_rate": 2.491885030856608e-06,
      "loss": 0.81336051,
      "num_input_tokens_seen": 78543245,
      "step": 3645,
      "time_per_iteration": 2.746540069580078
    },
    {
      "auxiliary_loss_clip": 0.01295217,
      "auxiliary_loss_mlp": 0.01028577,
      "balance_loss_clip": 1.05398142,
      "balance_loss_mlp": 1.0201968,
      "epoch": 0.43840557927012563,
      "flos": 17165301177600.0,
      "grad_norm": 3.2380095752449267,
      "language_loss": 0.82589281,
      "learning_rate": 2.4911299484881713e-06,
      "loss": 0.84913075,
      "num_input_tokens_seen": 78560775,
      "step": 3646,
      "time_per_iteration": 2.6745572090148926
    },
    {
      "auxiliary_loss_clip": 0.01287247,
      "auxiliary_loss_mlp": 0.01035343,
      "balance_loss_clip": 1.04986382,
      "balance_loss_mlp": 1.02773714,
      "epoch": 0.43852582216076474,
      "flos": 19390685316480.0,
      "grad_norm": 1.660037849228677,
      "language_loss": 0.80948502,
      "learning_rate": 2.490374791615712e-06,
      "loss": 0.83271098,
      "num_input_tokens_seen": 78580800,
      "step": 3647,
      "time_per_iteration": 4.103867292404175
    },
    {
      "auxiliary_loss_clip": 0.01199812,
      "auxiliary_loss_mlp": 0.02576741,
      "balance_loss_clip": 1.05863428,
      "balance_loss_mlp": 1.00017452,
      "epoch": 0.43864606505140386,
      "flos": 18075133699200.0,
      "grad_norm": 4.081661212294089,
      "language_loss": 0.77691877,
      "learning_rate": 2.4896195603537867e-06,
      "loss": 0.81468427,
      "num_input_tokens_seen": 78595410,
      "step": 3648,
      "time_per_iteration": 2.6325931549072266
    },
    {
      "auxiliary_loss_clip": 0.01379605,
      "auxiliary_loss_mlp": 0.01029356,
      "balance_loss_clip": 1.05219519,
      "balance_loss_mlp": 1.02155936,
      "epoch": 0.4387663079420429,
      "flos": 19644896845440.0,
      "grad_norm": 2.3413320540595377,
      "language_loss": 0.74221408,
      "learning_rate": 2.488864254816964e-06,
      "loss": 0.76630366,
      "num_input_tokens_seen": 78614100,
      "step": 3649,
      "time_per_iteration": 2.747596263885498
    },
    {
      "auxiliary_loss_clip": 0.01246241,
      "auxiliary_loss_mlp": 0.01031937,
      "balance_loss_clip": 1.05719304,
      "balance_loss_mlp": 1.02312136,
      "epoch": 0.438886550832682,
      "flos": 19719339782400.0,
      "grad_norm": 3.619054590764539,
      "language_loss": 0.68349457,
      "learning_rate": 2.4881088751198218e-06,
      "loss": 0.70627636,
      "num_input_tokens_seen": 78632260,
      "step": 3650,
      "time_per_iteration": 3.5251617431640625
    },
    {
      "auxiliary_loss_clip": 0.01293429,
      "auxiliary_loss_mlp": 0.01027402,
      "balance_loss_clip": 1.05135286,
      "balance_loss_mlp": 1.01899457,
      "epoch": 0.43900679372332113,
      "flos": 14536675981440.0,
      "grad_norm": 2.7518245982952125,
      "language_loss": 0.63593435,
      "learning_rate": 2.4873534213769517e-06,
      "loss": 0.65914273,
      "num_input_tokens_seen": 78647490,
      "step": 3651,
      "time_per_iteration": 2.66157865524292
    },
    {
      "auxiliary_loss_clip": 0.0133532,
      "auxiliary_loss_mlp": 0.01025288,
      "balance_loss_clip": 1.05147636,
      "balance_loss_mlp": 1.01768267,
      "epoch": 0.4391270366139602,
      "flos": 24056234968320.0,
      "grad_norm": 2.1379320777519215,
      "language_loss": 0.71607924,
      "learning_rate": 2.4865978937029547e-06,
      "loss": 0.7396853,
      "num_input_tokens_seen": 78666470,
      "step": 3652,
      "time_per_iteration": 2.717647075653076
    },
    {
      "auxiliary_loss_clip": 0.0138093,
      "auxiliary_loss_mlp": 0.01026313,
      "balance_loss_clip": 1.04888868,
      "balance_loss_mlp": 1.01859474,
      "epoch": 0.4392472795045993,
      "flos": 31538510363520.0,
      "grad_norm": 1.6468010401713709,
      "language_loss": 0.66342866,
      "learning_rate": 2.485842292212445e-06,
      "loss": 0.68750107,
      "num_input_tokens_seen": 78687685,
      "step": 3653,
      "time_per_iteration": 2.845449447631836
    },
    {
      "auxiliary_loss_clip": 0.01196153,
      "auxiliary_loss_mlp": 0.01030954,
      "balance_loss_clip": 1.05723822,
      "balance_loss_mlp": 1.02304435,
      "epoch": 0.4393675223952384,
      "flos": 14866300114560.0,
      "grad_norm": 1.8824213127660807,
      "language_loss": 0.80477983,
      "learning_rate": 2.485086617020045e-06,
      "loss": 0.82705092,
      "num_input_tokens_seen": 78706180,
      "step": 3654,
      "time_per_iteration": 3.5354011058807373
    },
    {
      "auxiliary_loss_clip": 0.01286508,
      "auxiliary_loss_mlp": 0.01030828,
      "balance_loss_clip": 1.04961646,
      "balance_loss_mlp": 1.02239954,
      "epoch": 0.43948776528587746,
      "flos": 14825900292480.0,
      "grad_norm": 2.2315388401603036,
      "language_loss": 0.81584716,
      "learning_rate": 2.4843308682403903e-06,
      "loss": 0.83902049,
      "num_input_tokens_seen": 78723095,
      "step": 3655,
      "time_per_iteration": 2.6608755588531494
    },
    {
      "auxiliary_loss_clip": 0.01191689,
      "auxiliary_loss_mlp": 0.01024681,
      "balance_loss_clip": 1.05412161,
      "balance_loss_mlp": 1.01719487,
      "epoch": 0.4396080081765166,
      "flos": 13914523486080.0,
      "grad_norm": 3.5788394175186817,
      "language_loss": 0.82979465,
      "learning_rate": 2.4835750459881294e-06,
      "loss": 0.85195839,
      "num_input_tokens_seen": 78739720,
      "step": 3656,
      "time_per_iteration": 2.578390121459961
    },
    {
      "auxiliary_loss_clip": 0.01287396,
      "auxiliary_loss_mlp": 0.01026843,
      "balance_loss_clip": 1.04741979,
      "balance_loss_mlp": 1.01889753,
      "epoch": 0.43972825106715563,
      "flos": 18222978078720.0,
      "grad_norm": 1.9285165040900338,
      "language_loss": 0.81532592,
      "learning_rate": 2.4828191503779177e-06,
      "loss": 0.83846831,
      "num_input_tokens_seen": 78757820,
      "step": 3657,
      "time_per_iteration": 3.7613468170166016
    },
    {
      "auxiliary_loss_clip": 0.01335855,
      "auxiliary_loss_mlp": 0.01032013,
      "balance_loss_clip": 1.04758954,
      "balance_loss_mlp": 1.02329254,
      "epoch": 0.43984849395779474,
      "flos": 16873239692160.0,
      "grad_norm": 1.9463114340833605,
      "language_loss": 0.89893234,
      "learning_rate": 2.482063181524425e-06,
      "loss": 0.92261106,
      "num_input_tokens_seen": 78773720,
      "step": 3658,
      "time_per_iteration": 2.8031158447265625
    },
    {
      "auxiliary_loss_clip": 0.0119641,
      "auxiliary_loss_mlp": 0.01030623,
      "balance_loss_clip": 1.05652285,
      "balance_loss_mlp": 1.02168787,
      "epoch": 0.43996873684843385,
      "flos": 18691504104960.0,
      "grad_norm": 2.0044187706947296,
      "language_loss": 0.81445193,
      "learning_rate": 2.4813071395423307e-06,
      "loss": 0.83672225,
      "num_input_tokens_seen": 78791285,
      "step": 3659,
      "time_per_iteration": 2.5990357398986816
    },
    {
      "auxiliary_loss_clip": 0.01241302,
      "auxiliary_loss_mlp": 0.01027201,
      "balance_loss_clip": 1.05244267,
      "balance_loss_mlp": 1.01945877,
      "epoch": 0.4400889797390729,
      "flos": 23653460787840.0,
      "grad_norm": 1.9768455460907064,
      "language_loss": 0.64964122,
      "learning_rate": 2.4805510245463263e-06,
      "loss": 0.67232627,
      "num_input_tokens_seen": 78811440,
      "step": 3660,
      "time_per_iteration": 2.690753936767578
    },
    {
      "auxiliary_loss_clip": 0.01242316,
      "auxiliary_loss_mlp": 0.01031707,
      "balance_loss_clip": 1.05158448,
      "balance_loss_mlp": 1.02352309,
      "epoch": 0.440209222629712,
      "flos": 23149203707520.0,
      "grad_norm": 2.5244088465145547,
      "language_loss": 0.60660112,
      "learning_rate": 2.4797948366511137e-06,
      "loss": 0.62934136,
      "num_input_tokens_seen": 78831150,
      "step": 3661,
      "time_per_iteration": 2.658458709716797
    },
    {
      "auxiliary_loss_clip": 0.01339094,
      "auxiliary_loss_mlp": 0.01027416,
      "balance_loss_clip": 1.04716456,
      "balance_loss_mlp": 1.01997161,
      "epoch": 0.4403294655203511,
      "flos": 24823394668800.0,
      "grad_norm": 2.8768115687304006,
      "language_loss": 0.76277453,
      "learning_rate": 2.4790385759714055e-06,
      "loss": 0.78643966,
      "num_input_tokens_seen": 78850215,
      "step": 3662,
      "time_per_iteration": 2.7350387573242188
    },
    {
      "auxiliary_loss_clip": 0.01248108,
      "auxiliary_loss_mlp": 0.01025229,
      "balance_loss_clip": 1.05920815,
      "balance_loss_mlp": 1.01718879,
      "epoch": 0.4404497084109902,
      "flos": 22565080736640.0,
      "grad_norm": 1.9554695720631914,
      "language_loss": 0.71298957,
      "learning_rate": 2.478282242621926e-06,
      "loss": 0.73572296,
      "num_input_tokens_seen": 78870675,
      "step": 3663,
      "time_per_iteration": 2.637824535369873
    },
    {
      "auxiliary_loss_clip": 0.01245779,
      "auxiliary_loss_mlp": 0.00998786,
      "balance_loss_clip": 1.01877964,
      "balance_loss_mlp": 0.9975878,
      "epoch": 0.4405699513016293,
      "flos": 64967073448320.0,
      "grad_norm": 0.8502763865470065,
      "language_loss": 0.59497631,
      "learning_rate": 2.477525836717411e-06,
      "loss": 0.61742187,
      "num_input_tokens_seen": 78938440,
      "step": 3664,
      "time_per_iteration": 3.40004563331604
    },
    {
      "auxiliary_loss_clip": 0.01239981,
      "auxiliary_loss_mlp": 0.01039485,
      "balance_loss_clip": 1.05162477,
      "balance_loss_mlp": 1.03098536,
      "epoch": 0.4406901941922684,
      "flos": 35661952978560.0,
      "grad_norm": 4.319037692219301,
      "language_loss": 0.79439437,
      "learning_rate": 2.476769358372606e-06,
      "loss": 0.81718904,
      "num_input_tokens_seen": 78960090,
      "step": 3665,
      "time_per_iteration": 2.7879490852355957
    },
    {
      "auxiliary_loss_clip": 0.01339794,
      "auxiliary_loss_mlp": 0.01027083,
      "balance_loss_clip": 1.05434799,
      "balance_loss_mlp": 1.01906574,
      "epoch": 0.44081043708290746,
      "flos": 18040767361920.0,
      "grad_norm": 2.0520763278678653,
      "language_loss": 0.74913609,
      "learning_rate": 2.4760128077022683e-06,
      "loss": 0.77280492,
      "num_input_tokens_seen": 78978225,
      "step": 3666,
      "time_per_iteration": 2.7001712322235107
    },
    {
      "auxiliary_loss_clip": 0.01287741,
      "auxiliary_loss_mlp": 0.01027757,
      "balance_loss_clip": 1.04776144,
      "balance_loss_mlp": 1.01997244,
      "epoch": 0.44093067997354657,
      "flos": 30153507799680.0,
      "grad_norm": 1.5453901491034103,
      "language_loss": 0.68307656,
      "learning_rate": 2.4752561848211672e-06,
      "loss": 0.70623153,
      "num_input_tokens_seen": 79000625,
      "step": 3667,
      "time_per_iteration": 2.8263654708862305
    },
    {
      "auxiliary_loss_clip": 0.01239709,
      "auxiliary_loss_mlp": 0.0103557,
      "balance_loss_clip": 1.05674171,
      "balance_loss_mlp": 1.02846479,
      "epoch": 0.4410509228641857,
      "flos": 23255068066560.0,
      "grad_norm": 3.9084065352777784,
      "language_loss": 0.71456718,
      "learning_rate": 2.4744994898440797e-06,
      "loss": 0.73731995,
      "num_input_tokens_seen": 79019415,
      "step": 3668,
      "time_per_iteration": 2.637880802154541
    },
    {
      "auxiliary_loss_clip": 0.01342968,
      "auxiliary_loss_mlp": 0.0102767,
      "balance_loss_clip": 1.04875469,
      "balance_loss_mlp": 1.0196228,
      "epoch": 0.44117116575482473,
      "flos": 19500571998720.0,
      "grad_norm": 1.9982485306790896,
      "language_loss": 0.83934611,
      "learning_rate": 2.473742722885797e-06,
      "loss": 0.86305249,
      "num_input_tokens_seen": 79038435,
      "step": 3669,
      "time_per_iteration": 2.7398924827575684
    },
    {
      "auxiliary_loss_clip": 0.01248248,
      "auxiliary_loss_mlp": 0.02575791,
      "balance_loss_clip": 1.05866778,
      "balance_loss_mlp": 1.00019598,
      "epoch": 0.44129140864546385,
      "flos": 27053124353280.0,
      "grad_norm": 7.76065153445084,
      "language_loss": 0.64965665,
      "learning_rate": 2.4729858840611197e-06,
      "loss": 0.68789703,
      "num_input_tokens_seen": 79057345,
      "step": 3670,
      "time_per_iteration": 2.675870418548584
    },
    {
      "auxiliary_loss_clip": 0.01191012,
      "auxiliary_loss_mlp": 0.01033135,
      "balance_loss_clip": 1.05493975,
      "balance_loss_mlp": 1.02569604,
      "epoch": 0.4414116515361029,
      "flos": 26102101910400.0,
      "grad_norm": 2.2496651145399844,
      "language_loss": 0.72850543,
      "learning_rate": 2.4722289734848605e-06,
      "loss": 0.75074685,
      "num_input_tokens_seen": 79077810,
      "step": 3671,
      "time_per_iteration": 2.721208333969116
    },
    {
      "auxiliary_loss_clip": 0.01345948,
      "auxiliary_loss_mlp": 0.01028654,
      "balance_loss_clip": 1.05645096,
      "balance_loss_mlp": 1.02092671,
      "epoch": 0.441531894426742,
      "flos": 21906083865600.0,
      "grad_norm": 1.9688862116279817,
      "language_loss": 0.77659667,
      "learning_rate": 2.471471991271841e-06,
      "loss": 0.80034274,
      "num_input_tokens_seen": 79094935,
      "step": 3672,
      "time_per_iteration": 2.7162418365478516
    },
    {
      "auxiliary_loss_clip": 0.01236974,
      "auxiliary_loss_mlp": 0.0103358,
      "balance_loss_clip": 1.05043101,
      "balance_loss_mlp": 1.02562928,
      "epoch": 0.4416521373173811,
      "flos": 23437099215360.0,
      "grad_norm": 2.041069827738184,
      "language_loss": 0.79501867,
      "learning_rate": 2.470714937536896e-06,
      "loss": 0.81772423,
      "num_input_tokens_seen": 79113660,
      "step": 3673,
      "time_per_iteration": 3.5676069259643555
    },
    {
      "auxiliary_loss_clip": 0.01394514,
      "auxiliary_loss_mlp": 0.01031219,
      "balance_loss_clip": 1.05287576,
      "balance_loss_mlp": 1.02236187,
      "epoch": 0.4417723802080202,
      "flos": 20334345471360.0,
      "grad_norm": 1.923843422672611,
      "language_loss": 0.7073096,
      "learning_rate": 2.469957812394868e-06,
      "loss": 0.73156697,
      "num_input_tokens_seen": 79132470,
      "step": 3674,
      "time_per_iteration": 2.804262638092041
    },
    {
      "auxiliary_loss_clip": 0.01192687,
      "auxiliary_loss_mlp": 0.01026836,
      "balance_loss_clip": 1.05763066,
      "balance_loss_mlp": 1.0194627,
      "epoch": 0.4418926230986593,
      "flos": 18880682060160.0,
      "grad_norm": 2.010929986598282,
      "language_loss": 0.76056063,
      "learning_rate": 2.4692006159606148e-06,
      "loss": 0.78275585,
      "num_input_tokens_seen": 79150000,
      "step": 3675,
      "time_per_iteration": 3.4874372482299805
    },
    {
      "auxiliary_loss_clip": 0.01190138,
      "auxiliary_loss_mlp": 0.01034086,
      "balance_loss_clip": 1.05263257,
      "balance_loss_mlp": 1.02636147,
      "epoch": 0.4420128659892984,
      "flos": 19464409981440.0,
      "grad_norm": 2.226479243552932,
      "language_loss": 0.78773665,
      "learning_rate": 2.468443348349e-06,
      "loss": 0.80997884,
      "num_input_tokens_seen": 79167875,
      "step": 3676,
      "time_per_iteration": 2.6435623168945312
    },
    {
      "auxiliary_loss_clip": 0.01385898,
      "auxiliary_loss_mlp": 0.01031944,
      "balance_loss_clip": 1.04609835,
      "balance_loss_mlp": 1.02393949,
      "epoch": 0.44213310887993745,
      "flos": 17894359526400.0,
      "grad_norm": 2.395775891984414,
      "language_loss": 0.82746494,
      "learning_rate": 2.467686009674902e-06,
      "loss": 0.85164338,
      "num_input_tokens_seen": 79182325,
      "step": 3677,
      "time_per_iteration": 2.729168176651001
    },
    {
      "auxiliary_loss_clip": 0.01238638,
      "auxiliary_loss_mlp": 0.01028887,
      "balance_loss_clip": 1.05047107,
      "balance_loss_mlp": 1.02087021,
      "epoch": 0.44225335177057656,
      "flos": 19204667758080.0,
      "grad_norm": 2.3097031710249687,
      "language_loss": 0.85226822,
      "learning_rate": 2.466928600053209e-06,
      "loss": 0.87494344,
      "num_input_tokens_seen": 79197630,
      "step": 3678,
      "time_per_iteration": 2.656160593032837
    },
    {
      "auxiliary_loss_clip": 0.0129276,
      "auxiliary_loss_mlp": 0.01029702,
      "balance_loss_clip": 1.05105937,
      "balance_loss_mlp": 1.02191806,
      "epoch": 0.4423735946612157,
      "flos": 23471321898240.0,
      "grad_norm": 2.3003399172885777,
      "language_loss": 0.71247113,
      "learning_rate": 2.466171119598818e-06,
      "loss": 0.73569578,
      "num_input_tokens_seen": 79217600,
      "step": 3679,
      "time_per_iteration": 2.726193428039551
    },
    {
      "auxiliary_loss_clip": 0.01244131,
      "auxiliary_loss_mlp": 0.01026163,
      "balance_loss_clip": 1.05011499,
      "balance_loss_mlp": 1.01811326,
      "epoch": 0.44249383755185473,
      "flos": 26685398868480.0,
      "grad_norm": 1.8317933104616118,
      "language_loss": 0.76972616,
      "learning_rate": 2.465413568426639e-06,
      "loss": 0.79242909,
      "num_input_tokens_seen": 79238550,
      "step": 3680,
      "time_per_iteration": 2.6508617401123047
    },
    {
      "auxiliary_loss_clip": 0.01233668,
      "auxiliary_loss_mlp": 0.01028235,
      "balance_loss_clip": 1.05008137,
      "balance_loss_mlp": 1.02080774,
      "epoch": 0.44261408044249384,
      "flos": 23147659422720.0,
      "grad_norm": 1.713300305501779,
      "language_loss": 0.81220222,
      "learning_rate": 2.464655946651591e-06,
      "loss": 0.83482128,
      "num_input_tokens_seen": 79257555,
      "step": 3681,
      "time_per_iteration": 3.636655330657959
    },
    {
      "auxiliary_loss_clip": 0.01240736,
      "auxiliary_loss_mlp": 0.01031818,
      "balance_loss_clip": 1.0536195,
      "balance_loss_mlp": 1.02439702,
      "epoch": 0.44273432333313295,
      "flos": 24462564595200.0,
      "grad_norm": 1.9606686780714202,
      "language_loss": 0.80882102,
      "learning_rate": 2.4638982543886065e-06,
      "loss": 0.8315466,
      "num_input_tokens_seen": 79277595,
      "step": 3682,
      "time_per_iteration": 2.652500629425049
    },
    {
      "auxiliary_loss_clip": 0.01242936,
      "auxiliary_loss_mlp": 0.01040348,
      "balance_loss_clip": 1.05404699,
      "balance_loss_mlp": 1.03209245,
      "epoch": 0.442854566223772,
      "flos": 17528932512000.0,
      "grad_norm": 2.285180986394316,
      "language_loss": 0.8700397,
      "learning_rate": 2.4631404917526254e-06,
      "loss": 0.89287251,
      "num_input_tokens_seen": 79294550,
      "step": 3683,
      "time_per_iteration": 3.578205108642578
    },
    {
      "auxiliary_loss_clip": 0.01238667,
      "auxiliary_loss_mlp": 0.01029855,
      "balance_loss_clip": 1.05125952,
      "balance_loss_mlp": 1.02269077,
      "epoch": 0.4429748091144111,
      "flos": 24896293320960.0,
      "grad_norm": 1.576433020144998,
      "language_loss": 0.79254675,
      "learning_rate": 2.4623826588586e-06,
      "loss": 0.81523198,
      "num_input_tokens_seen": 79314820,
      "step": 3684,
      "time_per_iteration": 2.6929454803466797
    },
    {
      "auxiliary_loss_clip": 0.01288374,
      "auxiliary_loss_mlp": 0.01032819,
      "balance_loss_clip": 1.04859853,
      "balance_loss_mlp": 1.02463531,
      "epoch": 0.4430950520050502,
      "flos": 21614704738560.0,
      "grad_norm": 1.5528572497136857,
      "language_loss": 0.82668424,
      "learning_rate": 2.461624755821492e-06,
      "loss": 0.84989613,
      "num_input_tokens_seen": 79334300,
      "step": 3685,
      "time_per_iteration": 2.6083621978759766
    },
    {
      "auxiliary_loss_clip": 0.01243317,
      "auxiliary_loss_mlp": 0.0102927,
      "balance_loss_clip": 1.0503583,
      "balance_loss_mlp": 1.02152169,
      "epoch": 0.4432152948956893,
      "flos": 24572271709440.0,
      "grad_norm": 1.6981724054760885,
      "language_loss": 0.76274276,
      "learning_rate": 2.4608667827562763e-06,
      "loss": 0.7854687,
      "num_input_tokens_seen": 79353630,
      "step": 3686,
      "time_per_iteration": 2.7795984745025635
    },
    {
      "auxiliary_loss_clip": 0.01249437,
      "auxiliary_loss_mlp": 0.01030541,
      "balance_loss_clip": 1.0567503,
      "balance_loss_mlp": 1.0218091,
      "epoch": 0.4433355377863284,
      "flos": 21762261809280.0,
      "grad_norm": 2.6305949077882547,
      "language_loss": 0.90071869,
      "learning_rate": 2.460108739777936e-06,
      "loss": 0.92351854,
      "num_input_tokens_seen": 79372765,
      "step": 3687,
      "time_per_iteration": 2.6429452896118164
    },
    {
      "auxiliary_loss_clip": 0.01290859,
      "auxiliary_loss_mlp": 0.0102617,
      "balance_loss_clip": 1.05358386,
      "balance_loss_mlp": 1.01849902,
      "epoch": 0.44345578067696745,
      "flos": 20084479488000.0,
      "grad_norm": 1.6729632591677888,
      "language_loss": 0.76426852,
      "learning_rate": 2.4593506270014656e-06,
      "loss": 0.78743881,
      "num_input_tokens_seen": 79391735,
      "step": 3688,
      "time_per_iteration": 2.6688597202301025
    },
    {
      "auxiliary_loss_clip": 0.01290612,
      "auxiliary_loss_mlp": 0.01029935,
      "balance_loss_clip": 1.04938781,
      "balance_loss_mlp": 1.02143526,
      "epoch": 0.44357602356760656,
      "flos": 24169497528960.0,
      "grad_norm": 1.829886679436439,
      "language_loss": 0.81875074,
      "learning_rate": 2.45859244454187e-06,
      "loss": 0.84195614,
      "num_input_tokens_seen": 79411525,
      "step": 3689,
      "time_per_iteration": 2.729163646697998
    },
    {
      "auxiliary_loss_clip": 0.01235654,
      "auxiliary_loss_mlp": 0.01029769,
      "balance_loss_clip": 1.05409551,
      "balance_loss_mlp": 1.02231216,
      "epoch": 0.44369626645824567,
      "flos": 22707717644160.0,
      "grad_norm": 1.9311850136490099,
      "language_loss": 0.66700423,
      "learning_rate": 2.4578341925141655e-06,
      "loss": 0.68965846,
      "num_input_tokens_seen": 79430740,
      "step": 3690,
      "time_per_iteration": 2.693181276321411
    },
    {
      "auxiliary_loss_clip": 0.01247625,
      "auxiliary_loss_mlp": 0.01025919,
      "balance_loss_clip": 1.05311584,
      "balance_loss_mlp": 1.01809883,
      "epoch": 0.4438165093488847,
      "flos": 38030225420160.0,
      "grad_norm": 2.077117068977518,
      "language_loss": 0.72281384,
      "learning_rate": 2.457075871033378e-06,
      "loss": 0.7455492,
      "num_input_tokens_seen": 79452615,
      "step": 3691,
      "time_per_iteration": 2.781879186630249
    },
    {
      "auxiliary_loss_clip": 0.013362,
      "auxiliary_loss_mlp": 0.01028549,
      "balance_loss_clip": 1.04954839,
      "balance_loss_mlp": 1.02083623,
      "epoch": 0.44393675223952384,
      "flos": 15523213996800.0,
      "grad_norm": 2.2687762630086437,
      "language_loss": 0.88627636,
      "learning_rate": 2.4563174802145445e-06,
      "loss": 0.90992391,
      "num_input_tokens_seen": 79469865,
      "step": 3692,
      "time_per_iteration": 2.7024874687194824
    },
    {
      "auxiliary_loss_clip": 0.01190686,
      "auxiliary_loss_mlp": 0.009984,
      "balance_loss_clip": 1.01979434,
      "balance_loss_mlp": 0.99738044,
      "epoch": 0.44405699513016295,
      "flos": 64574893779840.0,
      "grad_norm": 0.6347478576403353,
      "language_loss": 0.48635247,
      "learning_rate": 2.455559020172712e-06,
      "loss": 0.50824332,
      "num_input_tokens_seen": 79537220,
      "step": 3693,
      "time_per_iteration": 3.3425819873809814
    },
    {
      "auxiliary_loss_clip": 0.01390471,
      "auxiliary_loss_mlp": 0.01033835,
      "balance_loss_clip": 1.05336583,
      "balance_loss_mlp": 1.02603936,
      "epoch": 0.444177238020802,
      "flos": 23987394552960.0,
      "grad_norm": 2.2295318912027806,
      "language_loss": 0.89800674,
      "learning_rate": 2.4548004910229385e-06,
      "loss": 0.92224979,
      "num_input_tokens_seen": 79554795,
      "step": 3694,
      "time_per_iteration": 2.7460951805114746
    },
    {
      "auxiliary_loss_clip": 0.01242309,
      "auxiliary_loss_mlp": 0.02572979,
      "balance_loss_clip": 1.05286455,
      "balance_loss_mlp": 1.00028479,
      "epoch": 0.4442974809114411,
      "flos": 22563069575040.0,
      "grad_norm": 16.422846651453995,
      "language_loss": 0.87031204,
      "learning_rate": 2.4540418928802913e-06,
      "loss": 0.90846491,
      "num_input_tokens_seen": 79573530,
      "step": 3695,
      "time_per_iteration": 2.7322394847869873
    },
    {
      "auxiliary_loss_clip": 0.01294676,
      "auxiliary_loss_mlp": 0.01029594,
      "balance_loss_clip": 1.05296242,
      "balance_loss_mlp": 1.02142787,
      "epoch": 0.4444177238020802,
      "flos": 17675699483520.0,
      "grad_norm": 2.1858803640084097,
      "language_loss": 0.6574589,
      "learning_rate": 2.4532832258598506e-06,
      "loss": 0.68070161,
      "num_input_tokens_seen": 79591360,
      "step": 3696,
      "time_per_iteration": 2.657613515853882
    },
    {
      "auxiliary_loss_clip": 0.01190192,
      "auxiliary_loss_mlp": 0.01028287,
      "balance_loss_clip": 1.05512357,
      "balance_loss_mlp": 1.02065778,
      "epoch": 0.4445379666927193,
      "flos": 28621594609920.0,
      "grad_norm": 1.8067751417135207,
      "language_loss": 0.80545902,
      "learning_rate": 2.4525244900767047e-06,
      "loss": 0.82764387,
      "num_input_tokens_seen": 79612175,
      "step": 3697,
      "time_per_iteration": 2.727884531021118
    },
    {
      "auxiliary_loss_clip": 0.01140289,
      "auxiliary_loss_mlp": 0.01001892,
      "balance_loss_clip": 1.02708447,
      "balance_loss_mlp": 1.00083661,
      "epoch": 0.4446582095833584,
      "flos": 70487370115200.0,
      "grad_norm": 0.7800839528244163,
      "language_loss": 0.60482812,
      "learning_rate": 2.4517656856459536e-06,
      "loss": 0.62624991,
      "num_input_tokens_seen": 79678020,
      "step": 3698,
      "time_per_iteration": 3.336541175842285
    },
    {
      "auxiliary_loss_clip": 0.01244892,
      "auxiliary_loss_mlp": 0.0103185,
      "balance_loss_clip": 1.05599344,
      "balance_loss_mlp": 1.023785,
      "epoch": 0.4447784524739975,
      "flos": 26505199313280.0,
      "grad_norm": 1.683193250307588,
      "language_loss": 0.68536592,
      "learning_rate": 2.4510068126827073e-06,
      "loss": 0.70813334,
      "num_input_tokens_seen": 79699020,
      "step": 3699,
      "time_per_iteration": 3.6834263801574707
    },
    {
      "auxiliary_loss_clip": 0.01290449,
      "auxiliary_loss_mlp": 0.01025825,
      "balance_loss_clip": 1.05224621,
      "balance_loss_mlp": 1.017802,
      "epoch": 0.44489869536463655,
      "flos": 11656209553920.0,
      "grad_norm": 2.3199011470148987,
      "language_loss": 0.8172797,
      "learning_rate": 2.450247871302086e-06,
      "loss": 0.84044242,
      "num_input_tokens_seen": 79716795,
      "step": 3700,
      "time_per_iteration": 2.723374366760254
    },
    {
      "auxiliary_loss_clip": 0.01149026,
      "auxiliary_loss_mlp": 0.01023885,
      "balance_loss_clip": 1.054497,
      "balance_loss_mlp": 1.01655674,
      "epoch": 0.44501893825527566,
      "flos": 20448469958400.0,
      "grad_norm": 2.5971370662413453,
      "language_loss": 0.8426773,
      "learning_rate": 2.44948886161922e-06,
      "loss": 0.86440647,
      "num_input_tokens_seen": 79735810,
      "step": 3701,
      "time_per_iteration": 3.511808156967163
    },
    {
      "auxiliary_loss_clip": 0.0124418,
      "auxiliary_loss_mlp": 0.01024847,
      "balance_loss_clip": 1.05395937,
      "balance_loss_mlp": 1.01728296,
      "epoch": 0.4451391811459148,
      "flos": 18261079430400.0,
      "grad_norm": 3.5906810708127233,
      "language_loss": 0.84849453,
      "learning_rate": 2.4487297837492524e-06,
      "loss": 0.87118471,
      "num_input_tokens_seen": 79754975,
      "step": 3702,
      "time_per_iteration": 2.637890577316284
    },
    {
      "auxiliary_loss_clip": 0.01336759,
      "auxiliary_loss_mlp": 0.01029919,
      "balance_loss_clip": 1.04950511,
      "balance_loss_mlp": 1.02194965,
      "epoch": 0.44525942403655383,
      "flos": 16910155895040.0,
      "grad_norm": 2.3521210355766824,
      "language_loss": 0.62255698,
      "learning_rate": 2.4479706378073323e-06,
      "loss": 0.64622372,
      "num_input_tokens_seen": 79773515,
      "step": 3703,
      "time_per_iteration": 2.702260971069336
    },
    {
      "auxiliary_loss_clip": 0.01332781,
      "auxiliary_loss_mlp": 0.01024545,
      "balance_loss_clip": 1.04517436,
      "balance_loss_mlp": 1.01713586,
      "epoch": 0.44537966692719294,
      "flos": 23258838994560.0,
      "grad_norm": 3.6008378175362146,
      "language_loss": 0.83900249,
      "learning_rate": 2.447211423908623e-06,
      "loss": 0.86257577,
      "num_input_tokens_seen": 79793560,
      "step": 3704,
      "time_per_iteration": 2.78127121925354
    },
    {
      "auxiliary_loss_clip": 0.0124196,
      "auxiliary_loss_mlp": 0.01023266,
      "balance_loss_clip": 1.05199814,
      "balance_loss_mlp": 1.01573253,
      "epoch": 0.445499909817832,
      "flos": 21724160457600.0,
      "grad_norm": 2.5810667340950846,
      "language_loss": 0.75063103,
      "learning_rate": 2.4464521421682966e-06,
      "loss": 0.7732833,
      "num_input_tokens_seen": 79811150,
      "step": 3705,
      "time_per_iteration": 2.6940622329711914
    },
    {
      "auxiliary_loss_clip": 0.01234643,
      "auxiliary_loss_mlp": 0.01021816,
      "balance_loss_clip": 1.05329978,
      "balance_loss_mlp": 1.01487541,
      "epoch": 0.4456201527084711,
      "flos": 23987969170560.0,
      "grad_norm": 1.3760451228982526,
      "language_loss": 0.87751299,
      "learning_rate": 2.4456927927015345e-06,
      "loss": 0.90007758,
      "num_input_tokens_seen": 79832190,
      "step": 3706,
      "time_per_iteration": 3.5958380699157715
    },
    {
      "auxiliary_loss_clip": 0.01296013,
      "auxiliary_loss_mlp": 0.01030456,
      "balance_loss_clip": 1.05374384,
      "balance_loss_mlp": 1.02262425,
      "epoch": 0.4457403955991102,
      "flos": 18807065136000.0,
      "grad_norm": 2.097516071621075,
      "language_loss": 0.76086938,
      "learning_rate": 2.4449333756235307e-06,
      "loss": 0.78413409,
      "num_input_tokens_seen": 79848905,
      "step": 3707,
      "time_per_iteration": 2.6564993858337402
    },
    {
      "auxiliary_loss_clip": 0.01147771,
      "auxiliary_loss_mlp": 0.01027125,
      "balance_loss_clip": 1.05492282,
      "balance_loss_mlp": 1.0196656,
      "epoch": 0.4458606384897493,
      "flos": 19207756327680.0,
      "grad_norm": 4.385392629449563,
      "language_loss": 0.79401582,
      "learning_rate": 2.4441738910494876e-06,
      "loss": 0.81576478,
      "num_input_tokens_seen": 79863640,
      "step": 3708,
      "time_per_iteration": 3.59679913520813
    },
    {
      "auxiliary_loss_clip": 0.01294464,
      "auxiliary_loss_mlp": 0.01030109,
      "balance_loss_clip": 1.04873812,
      "balance_loss_mlp": 1.02202666,
      "epoch": 0.4459808813803884,
      "flos": 21361283308800.0,
      "grad_norm": 1.9019144346155867,
      "language_loss": 0.82293272,
      "learning_rate": 2.4434143390946176e-06,
      "loss": 0.84617841,
      "num_input_tokens_seen": 79882450,
      "step": 3709,
      "time_per_iteration": 2.6975271701812744
    },
    {
      "auxiliary_loss_clip": 0.01332791,
      "auxiliary_loss_mlp": 0.01030503,
      "balance_loss_clip": 1.04926145,
      "balance_loss_mlp": 1.02262294,
      "epoch": 0.4461011242710275,
      "flos": 23288967527040.0,
      "grad_norm": 2.4009411949666437,
      "language_loss": 0.85677922,
      "learning_rate": 2.4426547198741457e-06,
      "loss": 0.8804121,
      "num_input_tokens_seen": 79900655,
      "step": 3710,
      "time_per_iteration": 2.7406575679779053
    },
    {
      "auxiliary_loss_clip": 0.01388968,
      "auxiliary_loss_mlp": 0.01028017,
      "balance_loss_clip": 1.05263972,
      "balance_loss_mlp": 1.02063751,
      "epoch": 0.44622136716166655,
      "flos": 20193001453440.0,
      "grad_norm": 2.3604504946926452,
      "language_loss": 0.74464369,
      "learning_rate": 2.441895033503305e-06,
      "loss": 0.76881355,
      "num_input_tokens_seen": 79918575,
      "step": 3711,
      "time_per_iteration": 2.691481828689575
    },
    {
      "auxiliary_loss_clip": 0.01240086,
      "auxiliary_loss_mlp": 0.01027682,
      "balance_loss_clip": 1.05272925,
      "balance_loss_mlp": 1.01921248,
      "epoch": 0.44634161005230566,
      "flos": 21283033530240.0,
      "grad_norm": 1.7615770687975116,
      "language_loss": 0.82123739,
      "learning_rate": 2.4411352800973375e-06,
      "loss": 0.8439151,
      "num_input_tokens_seen": 79937010,
      "step": 3712,
      "time_per_iteration": 2.676727294921875
    },
    {
      "auxiliary_loss_clip": 0.01335839,
      "auxiliary_loss_mlp": 0.01028496,
      "balance_loss_clip": 1.0479362,
      "balance_loss_mlp": 1.02005053,
      "epoch": 0.44646185294294477,
      "flos": 22929358515840.0,
      "grad_norm": 2.459165516222412,
      "language_loss": 0.75003099,
      "learning_rate": 2.4403754597715005e-06,
      "loss": 0.77367431,
      "num_input_tokens_seen": 79956455,
      "step": 3713,
      "time_per_iteration": 2.7106218338012695
    },
    {
      "auxiliary_loss_clip": 0.01293393,
      "auxiliary_loss_mlp": 0.01028578,
      "balance_loss_clip": 1.04811358,
      "balance_loss_mlp": 1.02048361,
      "epoch": 0.4465820958335838,
      "flos": 22637692080000.0,
      "grad_norm": 7.602396509808295,
      "language_loss": 0.93315625,
      "learning_rate": 2.4396155726410553e-06,
      "loss": 0.95637596,
      "num_input_tokens_seen": 79975065,
      "step": 3714,
      "time_per_iteration": 2.743741512298584
    },
    {
      "auxiliary_loss_clip": 0.01149275,
      "auxiliary_loss_mlp": 0.01028577,
      "balance_loss_clip": 1.05221891,
      "balance_loss_mlp": 1.02038193,
      "epoch": 0.44670233872422294,
      "flos": 22672525294080.0,
      "grad_norm": 2.559781605195445,
      "language_loss": 0.9080739,
      "learning_rate": 2.438855618821278e-06,
      "loss": 0.92985243,
      "num_input_tokens_seen": 79990865,
      "step": 3715,
      "time_per_iteration": 2.645972967147827
    },
    {
      "auxiliary_loss_clip": 0.01235582,
      "auxiliary_loss_mlp": 0.01030247,
      "balance_loss_clip": 1.04982448,
      "balance_loss_mlp": 1.02296972,
      "epoch": 0.44682258161486205,
      "flos": 23582178247680.0,
      "grad_norm": 2.1279969550035105,
      "language_loss": 0.67548454,
      "learning_rate": 2.4380955984274517e-06,
      "loss": 0.69814289,
      "num_input_tokens_seen": 80009520,
      "step": 3716,
      "time_per_iteration": 2.6464011669158936
    },
    {
      "auxiliary_loss_clip": 0.01242152,
      "auxiliary_loss_mlp": 0.01030693,
      "balance_loss_clip": 1.05128515,
      "balance_loss_mlp": 1.02278948,
      "epoch": 0.4469428245055011,
      "flos": 26501356558080.0,
      "grad_norm": 2.9328385849130942,
      "language_loss": 0.77010018,
      "learning_rate": 2.4373355115748716e-06,
      "loss": 0.79282862,
      "num_input_tokens_seen": 80030350,
      "step": 3717,
      "time_per_iteration": 2.6546247005462646
    },
    {
      "auxiliary_loss_clip": 0.01284948,
      "auxiliary_loss_mlp": 0.01025573,
      "balance_loss_clip": 1.04979134,
      "balance_loss_mlp": 1.01831007,
      "epoch": 0.4470630673961402,
      "flos": 21504925797120.0,
      "grad_norm": 1.8423883595885777,
      "language_loss": 0.72192341,
      "learning_rate": 2.436575358378842e-06,
      "loss": 0.74502861,
      "num_input_tokens_seen": 80049840,
      "step": 3718,
      "time_per_iteration": 2.702970027923584
    },
    {
      "auxiliary_loss_clip": 0.01294114,
      "auxiliary_loss_mlp": 0.010314,
      "balance_loss_clip": 1.05216324,
      "balance_loss_mlp": 1.02415764,
      "epoch": 0.44718331028677927,
      "flos": 16173986653440.0,
      "grad_norm": 7.562878547212082,
      "language_loss": 0.82822949,
      "learning_rate": 2.4358151389546782e-06,
      "loss": 0.85148466,
      "num_input_tokens_seen": 80066525,
      "step": 3719,
      "time_per_iteration": 2.644186019897461
    },
    {
      "auxiliary_loss_clip": 0.01193138,
      "auxiliary_loss_mlp": 0.01030646,
      "balance_loss_clip": 1.05595374,
      "balance_loss_mlp": 1.02249169,
      "epoch": 0.4473035531774184,
      "flos": 19681238430720.0,
      "grad_norm": 3.4924914284154545,
      "language_loss": 0.76343095,
      "learning_rate": 2.4350548534177035e-06,
      "loss": 0.78566885,
      "num_input_tokens_seen": 80083355,
      "step": 3720,
      "time_per_iteration": 2.6119513511657715
    },
    {
      "auxiliary_loss_clip": 0.01340161,
      "auxiliary_loss_mlp": 0.01030552,
      "balance_loss_clip": 1.04995191,
      "balance_loss_mlp": 1.02310765,
      "epoch": 0.4474237960680575,
      "flos": 41427590515200.0,
      "grad_norm": 1.7113879772095146,
      "language_loss": 0.66618979,
      "learning_rate": 2.434294501883254e-06,
      "loss": 0.68989694,
      "num_input_tokens_seen": 80106450,
      "step": 3721,
      "time_per_iteration": 2.9217934608459473
    },
    {
      "auxiliary_loss_clip": 0.01286668,
      "auxiliary_loss_mlp": 0.010256,
      "balance_loss_clip": 1.0491668,
      "balance_loss_mlp": 1.01843858,
      "epoch": 0.44754403895869654,
      "flos": 22891328991360.0,
      "grad_norm": 1.783134079730421,
      "language_loss": 0.65695965,
      "learning_rate": 2.433534084466674e-06,
      "loss": 0.68008232,
      "num_input_tokens_seen": 80125670,
      "step": 3722,
      "time_per_iteration": 2.6394989490509033
    },
    {
      "auxiliary_loss_clip": 0.0118811,
      "auxiliary_loss_mlp": 0.01024636,
      "balance_loss_clip": 1.05409729,
      "balance_loss_mlp": 1.01733482,
      "epoch": 0.44766428184933565,
      "flos": 25630271832960.0,
      "grad_norm": 1.4284424971931367,
      "language_loss": 0.70918661,
      "learning_rate": 2.4327736012833178e-06,
      "loss": 0.73131412,
      "num_input_tokens_seen": 80147390,
      "step": 3723,
      "time_per_iteration": 2.6845009326934814
    },
    {
      "auxiliary_loss_clip": 0.01241397,
      "auxiliary_loss_mlp": 0.01032005,
      "balance_loss_clip": 1.05439556,
      "balance_loss_mlp": 1.02439928,
      "epoch": 0.44778452473997477,
      "flos": 20448972748800.0,
      "grad_norm": 2.0754534051987648,
      "language_loss": 0.76696777,
      "learning_rate": 2.4320130524485506e-06,
      "loss": 0.78970176,
      "num_input_tokens_seen": 80166185,
      "step": 3724,
      "time_per_iteration": 2.6409871578216553
    },
    {
      "auxiliary_loss_clip": 0.01291971,
      "auxiliary_loss_mlp": 0.01034345,
      "balance_loss_clip": 1.05943751,
      "balance_loss_mlp": 1.02699518,
      "epoch": 0.4479047676306138,
      "flos": 21975462984960.0,
      "grad_norm": 1.7697018722845022,
      "language_loss": 0.7963953,
      "learning_rate": 2.431252438077746e-06,
      "loss": 0.8196584,
      "num_input_tokens_seen": 80185685,
      "step": 3725,
      "time_per_iteration": 3.630763053894043
    },
    {
      "auxiliary_loss_clip": 0.01244955,
      "auxiliary_loss_mlp": 0.02572192,
      "balance_loss_clip": 1.05243397,
      "balance_loss_mlp": 1.00029027,
      "epoch": 0.44802501052125293,
      "flos": 21467219495040.0,
      "grad_norm": 2.2355415896932214,
      "language_loss": 0.77575248,
      "learning_rate": 2.4304917582862906e-06,
      "loss": 0.81392395,
      "num_input_tokens_seen": 80204865,
      "step": 3726,
      "time_per_iteration": 2.654783010482788
    },
    {
      "auxiliary_loss_clip": 0.01190495,
      "auxiliary_loss_mlp": 0.01024386,
      "balance_loss_clip": 1.0543232,
      "balance_loss_mlp": 1.01698935,
      "epoch": 0.44814525341189204,
      "flos": 22126970551680.0,
      "grad_norm": 2.004255668168773,
      "language_loss": 0.88026631,
      "learning_rate": 2.4297310131895774e-06,
      "loss": 0.9024151,
      "num_input_tokens_seen": 80223410,
      "step": 3727,
      "time_per_iteration": 2.6098766326904297
    },
    {
      "auxiliary_loss_clip": 0.01238964,
      "auxiliary_loss_mlp": 0.01030963,
      "balance_loss_clip": 1.05191779,
      "balance_loss_mlp": 1.02343535,
      "epoch": 0.4482654963025311,
      "flos": 16653933204480.0,
      "grad_norm": 2.126159557469395,
      "language_loss": 0.75374532,
      "learning_rate": 2.4289702029030113e-06,
      "loss": 0.77644467,
      "num_input_tokens_seen": 80240880,
      "step": 3728,
      "time_per_iteration": 3.5250134468078613
    },
    {
      "auxiliary_loss_clip": 0.01241402,
      "auxiliary_loss_mlp": 0.01027456,
      "balance_loss_clip": 1.05515909,
      "balance_loss_mlp": 1.01923084,
      "epoch": 0.4483857391931702,
      "flos": 18841251905280.0,
      "grad_norm": 2.1522964378988676,
      "language_loss": 0.82931137,
      "learning_rate": 2.4282093275420057e-06,
      "loss": 0.85199994,
      "num_input_tokens_seen": 80259910,
      "step": 3729,
      "time_per_iteration": 2.6070189476013184
    },
    {
      "auxiliary_loss_clip": 0.01147564,
      "auxiliary_loss_mlp": 0.01029006,
      "balance_loss_clip": 1.05336916,
      "balance_loss_mlp": 1.02144837,
      "epoch": 0.4485059820838093,
      "flos": 20372590477440.0,
      "grad_norm": 2.0605486830990305,
      "language_loss": 0.70365649,
      "learning_rate": 2.4274483872219863e-06,
      "loss": 0.7254222,
      "num_input_tokens_seen": 80277270,
      "step": 3730,
      "time_per_iteration": 2.6517491340637207
    },
    {
      "auxiliary_loss_clip": 0.0123842,
      "auxiliary_loss_mlp": 0.01032909,
      "balance_loss_clip": 1.05215263,
      "balance_loss_mlp": 1.02541363,
      "epoch": 0.4486262249744484,
      "flos": 20047742853120.0,
      "grad_norm": 2.0805774134270507,
      "language_loss": 0.93463743,
      "learning_rate": 2.426687382058386e-06,
      "loss": 0.95735067,
      "num_input_tokens_seen": 80295550,
      "step": 3731,
      "time_per_iteration": 2.6423158645629883
    },
    {
      "auxiliary_loss_clip": 0.01134886,
      "auxiliary_loss_mlp": 0.00999598,
      "balance_loss_clip": 1.0219419,
      "balance_loss_mlp": 0.99848986,
      "epoch": 0.4487464678650875,
      "flos": 64595684776320.0,
      "grad_norm": 0.8610757040144524,
      "language_loss": 0.59829783,
      "learning_rate": 2.425926312166649e-06,
      "loss": 0.61964273,
      "num_input_tokens_seen": 80348425,
      "step": 3732,
      "time_per_iteration": 3.946404218673706
    },
    {
      "auxiliary_loss_clip": 0.01296165,
      "auxiliary_loss_mlp": 0.01032419,
      "balance_loss_clip": 1.05412507,
      "balance_loss_mlp": 1.02414,
      "epoch": 0.4488667107557266,
      "flos": 20769798049920.0,
      "grad_norm": 3.322130977112531,
      "language_loss": 0.73114991,
      "learning_rate": 2.42516517766223e-06,
      "loss": 0.75443578,
      "num_input_tokens_seen": 80366505,
      "step": 3733,
      "time_per_iteration": 2.8545007705688477
    },
    {
      "auxiliary_loss_clip": 0.01190746,
      "auxiliary_loss_mlp": 0.01032228,
      "balance_loss_clip": 1.05646288,
      "balance_loss_mlp": 1.02483988,
      "epoch": 0.44898695364636565,
      "flos": 23951735326080.0,
      "grad_norm": 2.1078443529155386,
      "language_loss": 0.68077314,
      "learning_rate": 2.4244039786605907e-06,
      "loss": 0.70300281,
      "num_input_tokens_seen": 80387510,
      "step": 3734,
      "time_per_iteration": 2.628124952316284
    },
    {
      "auxiliary_loss_clip": 0.01395806,
      "auxiliary_loss_mlp": 0.01027268,
      "balance_loss_clip": 1.04957414,
      "balance_loss_mlp": 1.01934314,
      "epoch": 0.44910719653700476,
      "flos": 18624351628800.0,
      "grad_norm": 2.727985663337072,
      "language_loss": 0.82495773,
      "learning_rate": 2.4236427152772055e-06,
      "loss": 0.84918839,
      "num_input_tokens_seen": 80405915,
      "step": 3735,
      "time_per_iteration": 3.6828114986419678
    },
    {
      "auxiliary_loss_clip": 0.01229049,
      "auxiliary_loss_mlp": 0.01005259,
      "balance_loss_clip": 1.02051556,
      "balance_loss_mlp": 1.0041678,
      "epoch": 0.4492274394276438,
      "flos": 57033435749760.0,
      "grad_norm": 0.8424922756011921,
      "language_loss": 0.57366902,
      "learning_rate": 2.422881387627557e-06,
      "loss": 0.59601206,
      "num_input_tokens_seen": 80458365,
      "step": 3736,
      "time_per_iteration": 2.948422908782959
    },
    {
      "auxiliary_loss_clip": 0.01196525,
      "auxiliary_loss_mlp": 0.01030285,
      "balance_loss_clip": 1.05260074,
      "balance_loss_mlp": 1.02288806,
      "epoch": 0.4493476823182829,
      "flos": 23254888498560.0,
      "grad_norm": 1.6175395854326855,
      "language_loss": 0.77591336,
      "learning_rate": 2.422119995827139e-06,
      "loss": 0.79818141,
      "num_input_tokens_seen": 80478490,
      "step": 3737,
      "time_per_iteration": 2.7915284633636475
    },
    {
      "auxiliary_loss_clip": 0.01150937,
      "auxiliary_loss_mlp": 0.01032159,
      "balance_loss_clip": 1.05635905,
      "balance_loss_mlp": 1.02430344,
      "epoch": 0.44946792520892204,
      "flos": 15815131827840.0,
      "grad_norm": 3.1596265851656717,
      "language_loss": 0.73565793,
      "learning_rate": 2.4213585399914528e-06,
      "loss": 0.75748897,
      "num_input_tokens_seen": 80495695,
      "step": 3738,
      "time_per_iteration": 2.652933120727539
    },
    {
      "auxiliary_loss_clip": 0.01236015,
      "auxiliary_loss_mlp": 0.01028034,
      "balance_loss_clip": 1.05277944,
      "balance_loss_mlp": 1.02045226,
      "epoch": 0.4495881680995611,
      "flos": 19610063631360.0,
      "grad_norm": 1.8780888002989298,
      "language_loss": 0.85373259,
      "learning_rate": 2.4205970202360113e-06,
      "loss": 0.87637311,
      "num_input_tokens_seen": 80515260,
      "step": 3739,
      "time_per_iteration": 2.625178098678589
    },
    {
      "auxiliary_loss_clip": 0.01382655,
      "auxiliary_loss_mlp": 0.01025631,
      "balance_loss_clip": 1.04756999,
      "balance_loss_mlp": 1.01803792,
      "epoch": 0.4497084109902002,
      "flos": 26031465815040.0,
      "grad_norm": 2.4043587564760354,
      "language_loss": 0.78227782,
      "learning_rate": 2.4198354366763354e-06,
      "loss": 0.80636072,
      "num_input_tokens_seen": 80533900,
      "step": 3740,
      "time_per_iteration": 2.770158052444458
    },
    {
      "auxiliary_loss_clip": 0.01291345,
      "auxiliary_loss_mlp": 0.01028851,
      "balance_loss_clip": 1.05092001,
      "balance_loss_mlp": 1.02100086,
      "epoch": 0.4498286538808393,
      "flos": 14793688771200.0,
      "grad_norm": 2.214545018096871,
      "language_loss": 0.78434503,
      "learning_rate": 2.4190737894279587e-06,
      "loss": 0.80754697,
      "num_input_tokens_seen": 80551270,
      "step": 3741,
      "time_per_iteration": 2.627232789993286
    },
    {
      "auxiliary_loss_clip": 0.01330592,
      "auxiliary_loss_mlp": 0.01028009,
      "balance_loss_clip": 1.04283929,
      "balance_loss_mlp": 1.02076077,
      "epoch": 0.44994889677147837,
      "flos": 15450171690240.0,
      "grad_norm": 3.2609682076343334,
      "language_loss": 0.80605197,
      "learning_rate": 2.4183120786064203e-06,
      "loss": 0.829638,
      "num_input_tokens_seen": 80568145,
      "step": 3742,
      "time_per_iteration": 2.7067580223083496
    },
    {
      "auxiliary_loss_clip": 0.01240688,
      "auxiliary_loss_mlp": 0.02572512,
      "balance_loss_clip": 1.05456722,
      "balance_loss_mlp": 1.00026679,
      "epoch": 0.4500691396621175,
      "flos": 21798316085760.0,
      "grad_norm": 2.3274356843168564,
      "language_loss": 0.85519111,
      "learning_rate": 2.417550304327273e-06,
      "loss": 0.89332312,
      "num_input_tokens_seen": 80586185,
      "step": 3743,
      "time_per_iteration": 2.6546127796173096
    },
    {
      "auxiliary_loss_clip": 0.01195008,
      "auxiliary_loss_mlp": 0.01026791,
      "balance_loss_clip": 1.05634916,
      "balance_loss_mlp": 1.01809502,
      "epoch": 0.4501893825527566,
      "flos": 32382016421760.0,
      "grad_norm": 3.6757397910318397,
      "language_loss": 0.75987935,
      "learning_rate": 2.4167884667060763e-06,
      "loss": 0.7820974,
      "num_input_tokens_seen": 80608895,
      "step": 3744,
      "time_per_iteration": 2.728443145751953
    },
    {
      "auxiliary_loss_clip": 0.01295748,
      "auxiliary_loss_mlp": 0.01027197,
      "balance_loss_clip": 1.05213737,
      "balance_loss_mlp": 1.01819658,
      "epoch": 0.45030962544339564,
      "flos": 16544944362240.0,
      "grad_norm": 2.2534084406764796,
      "language_loss": 0.87413067,
      "learning_rate": 2.4160265658584e-06,
      "loss": 0.89736009,
      "num_input_tokens_seen": 80623785,
      "step": 3745,
      "time_per_iteration": 2.6248879432678223
    },
    {
      "auxiliary_loss_clip": 0.01246317,
      "auxiliary_loss_mlp": 0.01028728,
      "balance_loss_clip": 1.05477238,
      "balance_loss_mlp": 1.02109218,
      "epoch": 0.45042986833403476,
      "flos": 19573039687680.0,
      "grad_norm": 2.4001490442870046,
      "language_loss": 0.68589705,
      "learning_rate": 2.4152646018998253e-06,
      "loss": 0.70864749,
      "num_input_tokens_seen": 80642735,
      "step": 3746,
      "time_per_iteration": 2.63167142868042
    },
    {
      "auxiliary_loss_clip": 0.01234419,
      "auxiliary_loss_mlp": 0.01027239,
      "balance_loss_clip": 1.05173075,
      "balance_loss_mlp": 1.01883531,
      "epoch": 0.45055011122467387,
      "flos": 23112467072640.0,
      "grad_norm": 1.6317378981776416,
      "language_loss": 0.71991462,
      "learning_rate": 2.4145025749459403e-06,
      "loss": 0.74253118,
      "num_input_tokens_seen": 80663760,
      "step": 3747,
      "time_per_iteration": 2.6444029808044434
    },
    {
      "auxiliary_loss_clip": 0.01493731,
      "auxiliary_loss_mlp": 0.01031818,
      "balance_loss_clip": 1.05128551,
      "balance_loss_mlp": 1.02408147,
      "epoch": 0.4506703541153129,
      "flos": 19934623946880.0,
      "grad_norm": 1.843726805120693,
      "language_loss": 0.70123398,
      "learning_rate": 2.413740485112344e-06,
      "loss": 0.72648942,
      "num_input_tokens_seen": 80682100,
      "step": 3748,
      "time_per_iteration": 2.8397233486175537
    },
    {
      "auxiliary_loss_clip": 0.01289797,
      "auxiliary_loss_mlp": 0.01030825,
      "balance_loss_clip": 1.05613923,
      "balance_loss_mlp": 1.02320719,
      "epoch": 0.45079059700595203,
      "flos": 19499530504320.0,
      "grad_norm": 2.119312321412857,
      "language_loss": 0.82376075,
      "learning_rate": 2.412978332514646e-06,
      "loss": 0.84696692,
      "num_input_tokens_seen": 80700880,
      "step": 3749,
      "time_per_iteration": 2.8666911125183105
    },
    {
      "auxiliary_loss_clip": 0.01294305,
      "auxiliary_loss_mlp": 0.01025998,
      "balance_loss_clip": 1.05308676,
      "balance_loss_mlp": 1.0180583,
      "epoch": 0.4509108398965911,
      "flos": 27636313570560.0,
      "grad_norm": 1.9738335627143062,
      "language_loss": 0.72519988,
      "learning_rate": 2.4122161172684623e-06,
      "loss": 0.74840289,
      "num_input_tokens_seen": 80721675,
      "step": 3750,
      "time_per_iteration": 4.1330931186676025
    },
    {
      "auxiliary_loss_clip": 0.01197249,
      "auxiliary_loss_mlp": 0.01023857,
      "balance_loss_clip": 1.05304468,
      "balance_loss_mlp": 1.01598287,
      "epoch": 0.4510310827872302,
      "flos": 20995712640000.0,
      "grad_norm": 11.158845244137302,
      "language_loss": 0.84432101,
      "learning_rate": 2.4114538394894216e-06,
      "loss": 0.86653209,
      "num_input_tokens_seen": 80739315,
      "step": 3751,
      "time_per_iteration": 2.780290126800537
    },
    {
      "auxiliary_loss_clip": 0.01287173,
      "auxiliary_loss_mlp": 0.01025836,
      "balance_loss_clip": 1.04719615,
      "balance_loss_mlp": 1.01845694,
      "epoch": 0.4511513256778693,
      "flos": 16216684945920.0,
      "grad_norm": 1.8320710319742433,
      "language_loss": 0.83035618,
      "learning_rate": 2.410691499293161e-06,
      "loss": 0.85348624,
      "num_input_tokens_seen": 80757470,
      "step": 3752,
      "time_per_iteration": 2.6646435260772705
    },
    {
      "auxiliary_loss_clip": 0.01240296,
      "auxiliary_loss_mlp": 0.01021993,
      "balance_loss_clip": 1.05361485,
      "balance_loss_mlp": 1.01454282,
      "epoch": 0.45127156856850836,
      "flos": 25186702780800.0,
      "grad_norm": 2.0072652166886678,
      "language_loss": 0.74380028,
      "learning_rate": 2.409929096795326e-06,
      "loss": 0.76642323,
      "num_input_tokens_seen": 80777840,
      "step": 3753,
      "time_per_iteration": 2.7151083946228027
    },
    {
      "auxiliary_loss_clip": 0.01238498,
      "auxiliary_loss_mlp": 0.01034289,
      "balance_loss_clip": 1.04946828,
      "balance_loss_mlp": 1.02646899,
      "epoch": 0.4513918114591475,
      "flos": 20412523422720.0,
      "grad_norm": 1.9641472131863893,
      "language_loss": 0.79249746,
      "learning_rate": 2.409166632111573e-06,
      "loss": 0.8152253,
      "num_input_tokens_seen": 80795975,
      "step": 3754,
      "time_per_iteration": 3.5687038898468018
    },
    {
      "auxiliary_loss_clip": 0.01247629,
      "auxiliary_loss_mlp": 0.01027031,
      "balance_loss_clip": 1.05341995,
      "balance_loss_mlp": 1.01916325,
      "epoch": 0.4515120543497866,
      "flos": 26648482665600.0,
      "grad_norm": 2.408056542237496,
      "language_loss": 0.80849463,
      "learning_rate": 2.4084041053575674e-06,
      "loss": 0.83124125,
      "num_input_tokens_seen": 80815395,
      "step": 3755,
      "time_per_iteration": 2.6511526107788086
    },
    {
      "auxiliary_loss_clip": 0.01200548,
      "auxiliary_loss_mlp": 0.01022082,
      "balance_loss_clip": 1.05422103,
      "balance_loss_mlp": 1.01439571,
      "epoch": 0.45163229724042564,
      "flos": 20595093275520.0,
      "grad_norm": 1.9625880814995473,
      "language_loss": 0.72051525,
      "learning_rate": 2.4076415166489834e-06,
      "loss": 0.74274153,
      "num_input_tokens_seen": 80834805,
      "step": 3756,
      "time_per_iteration": 2.717254638671875
    },
    {
      "auxiliary_loss_clip": 0.01299459,
      "auxiliary_loss_mlp": 0.01026354,
      "balance_loss_clip": 1.05069959,
      "balance_loss_mlp": 1.01937485,
      "epoch": 0.45175254013106475,
      "flos": 21689004021120.0,
      "grad_norm": 2.1091342424594206,
      "language_loss": 0.7922554,
      "learning_rate": 2.406878866101506e-06,
      "loss": 0.81551349,
      "num_input_tokens_seen": 80853770,
      "step": 3757,
      "time_per_iteration": 2.745755910873413
    },
    {
      "auxiliary_loss_clip": 0.01193139,
      "auxiliary_loss_mlp": 0.0103019,
      "balance_loss_clip": 1.05760837,
      "balance_loss_mlp": 1.0222919,
      "epoch": 0.45187278302170386,
      "flos": 18878850466560.0,
      "grad_norm": 5.228285237625514,
      "language_loss": 0.78506172,
      "learning_rate": 2.4061161538308273e-06,
      "loss": 0.80729508,
      "num_input_tokens_seen": 80870615,
      "step": 3758,
      "time_per_iteration": 3.465075969696045
    },
    {
      "auxiliary_loss_clip": 0.01239167,
      "auxiliary_loss_mlp": 0.01031712,
      "balance_loss_clip": 1.0539279,
      "balance_loss_mlp": 1.02395153,
      "epoch": 0.4519930259123429,
      "flos": 18582479349120.0,
      "grad_norm": 1.7994305621921198,
      "language_loss": 0.89218819,
      "learning_rate": 2.4053533799526523e-06,
      "loss": 0.91489697,
      "num_input_tokens_seen": 80886335,
      "step": 3759,
      "time_per_iteration": 2.665757894515991
    },
    {
      "auxiliary_loss_clip": 0.01280344,
      "auxiliary_loss_mlp": 0.01023454,
      "balance_loss_clip": 1.05118561,
      "balance_loss_mlp": 1.01603031,
      "epoch": 0.452113268802982,
      "flos": 25192377129600.0,
      "grad_norm": 1.9760134220883396,
      "language_loss": 0.86507332,
      "learning_rate": 2.404590544582691e-06,
      "loss": 0.88811123,
      "num_input_tokens_seen": 80904570,
      "step": 3760,
      "time_per_iteration": 2.710015296936035
    },
    {
      "auxiliary_loss_clip": 0.0139368,
      "auxiliary_loss_mlp": 0.01028552,
      "balance_loss_clip": 1.04398036,
      "balance_loss_mlp": 1.02108097,
      "epoch": 0.45223351169362114,
      "flos": 39378922312320.0,
      "grad_norm": 1.6337276316873381,
      "language_loss": 0.81255722,
      "learning_rate": 2.403827647836666e-06,
      "loss": 0.83677959,
      "num_input_tokens_seen": 80925125,
      "step": 3761,
      "time_per_iteration": 3.8277535438537598
    },
    {
      "auxiliary_loss_clip": 0.01193956,
      "auxiliary_loss_mlp": 0.01026179,
      "balance_loss_clip": 1.05554533,
      "balance_loss_mlp": 1.01867437,
      "epoch": 0.4523537545842602,
      "flos": 21582169994880.0,
      "grad_norm": 2.8573908355479145,
      "language_loss": 0.69340968,
      "learning_rate": 2.4030646898303075e-06,
      "loss": 0.71561098,
      "num_input_tokens_seen": 80946615,
      "step": 3762,
      "time_per_iteration": 2.5686209201812744
    },
    {
      "auxiliary_loss_clip": 0.01295246,
      "auxiliary_loss_mlp": 0.01029737,
      "balance_loss_clip": 1.05134761,
      "balance_loss_mlp": 1.02177942,
      "epoch": 0.4524739974748993,
      "flos": 28439527547520.0,
      "grad_norm": 2.2930931055239907,
      "language_loss": 0.81911272,
      "learning_rate": 2.4023016706793566e-06,
      "loss": 0.84236252,
      "num_input_tokens_seen": 80966410,
      "step": 3763,
      "time_per_iteration": 2.773864269256592
    },
    {
      "auxiliary_loss_clip": 0.01247756,
      "auxiliary_loss_mlp": 0.01002805,
      "balance_loss_clip": 1.01754284,
      "balance_loss_mlp": 1.00160146,
      "epoch": 0.4525942403655384,
      "flos": 61556492148480.0,
      "grad_norm": 0.7757053055278016,
      "language_loss": 0.56829971,
      "learning_rate": 2.401538590499561e-06,
      "loss": 0.59080535,
      "num_input_tokens_seen": 81026865,
      "step": 3764,
      "time_per_iteration": 3.32999849319458
    },
    {
      "auxiliary_loss_clip": 0.01245086,
      "auxiliary_loss_mlp": 0.02574962,
      "balance_loss_clip": 1.05427337,
      "balance_loss_mlp": 1.00023282,
      "epoch": 0.45271448325617747,
      "flos": 27529838680320.0,
      "grad_norm": 2.6318939199140345,
      "language_loss": 0.71916378,
      "learning_rate": 2.400775449406682e-06,
      "loss": 0.75736427,
      "num_input_tokens_seen": 81050060,
      "step": 3765,
      "time_per_iteration": 2.7250149250030518
    },
    {
      "auxiliary_loss_clip": 0.01240035,
      "auxiliary_loss_mlp": 0.01027003,
      "balance_loss_clip": 1.05167437,
      "balance_loss_mlp": 1.01952839,
      "epoch": 0.4528347261468166,
      "flos": 22452608275200.0,
      "grad_norm": 1.7769371975097383,
      "language_loss": 0.73263663,
      "learning_rate": 2.400012247516485e-06,
      "loss": 0.75530702,
      "num_input_tokens_seen": 81070625,
      "step": 3766,
      "time_per_iteration": 2.6692867279052734
    },
    {
      "auxiliary_loss_clip": 0.01343721,
      "auxiliary_loss_mlp": 0.01038278,
      "balance_loss_clip": 1.05039597,
      "balance_loss_mlp": 1.0302496,
      "epoch": 0.45295496903745563,
      "flos": 21103875469440.0,
      "grad_norm": 1.696614396341837,
      "language_loss": 0.90295804,
      "learning_rate": 2.3992489849447484e-06,
      "loss": 0.92677808,
      "num_input_tokens_seen": 81089080,
      "step": 3767,
      "time_per_iteration": 2.680875539779663
    },
    {
      "auxiliary_loss_clip": 0.01247192,
      "auxiliary_loss_mlp": 0.01027086,
      "balance_loss_clip": 1.05208731,
      "balance_loss_mlp": 1.01947474,
      "epoch": 0.45307521192809475,
      "flos": 23221168606080.0,
      "grad_norm": 1.606540719140495,
      "language_loss": 0.78940904,
      "learning_rate": 2.3984856618072584e-06,
      "loss": 0.81215179,
      "num_input_tokens_seen": 81109115,
      "step": 3768,
      "time_per_iteration": 2.6931560039520264
    },
    {
      "auxiliary_loss_clip": 0.0134466,
      "auxiliary_loss_mlp": 0.01028719,
      "balance_loss_clip": 1.05013061,
      "balance_loss_mlp": 1.02110755,
      "epoch": 0.45319545481873386,
      "flos": 15560094286080.0,
      "grad_norm": 1.9719938210876449,
      "language_loss": 0.74178606,
      "learning_rate": 2.3977222782198098e-06,
      "loss": 0.76551986,
      "num_input_tokens_seen": 81127750,
      "step": 3769,
      "time_per_iteration": 2.694934368133545
    },
    {
      "auxiliary_loss_clip": 0.01340897,
      "auxiliary_loss_mlp": 0.01031013,
      "balance_loss_clip": 1.05171669,
      "balance_loss_mlp": 1.02297258,
      "epoch": 0.4533156977093729,
      "flos": 21944759834880.0,
      "grad_norm": 2.3363946334963344,
      "language_loss": 0.75187653,
      "learning_rate": 2.3969588342982077e-06,
      "loss": 0.77559566,
      "num_input_tokens_seen": 81147125,
      "step": 3770,
      "time_per_iteration": 2.7590036392211914
    },
    {
      "auxiliary_loss_clip": 0.01238487,
      "auxiliary_loss_mlp": 0.01031703,
      "balance_loss_clip": 1.05597019,
      "balance_loss_mlp": 1.02378178,
      "epoch": 0.453435940600012,
      "flos": 24242180699520.0,
      "grad_norm": 2.0435823618122733,
      "language_loss": 0.73015046,
      "learning_rate": 2.396195330158267e-06,
      "loss": 0.75285232,
      "num_input_tokens_seen": 81167015,
      "step": 3771,
      "time_per_iteration": 2.6724274158477783
    },
    {
      "auxiliary_loss_clip": 0.01191082,
      "auxiliary_loss_mlp": 0.01028062,
      "balance_loss_clip": 1.05421352,
      "balance_loss_mlp": 1.01983023,
      "epoch": 0.45355618349065113,
      "flos": 23440367352960.0,
      "grad_norm": 1.906045884705502,
      "language_loss": 0.79778373,
      "learning_rate": 2.3954317659158094e-06,
      "loss": 0.81997514,
      "num_input_tokens_seen": 81187350,
      "step": 3772,
      "time_per_iteration": 2.6691017150878906
    },
    {
      "auxiliary_loss_clip": 0.01082304,
      "auxiliary_loss_mlp": 0.00999525,
      "balance_loss_clip": 1.01901031,
      "balance_loss_mlp": 0.99845171,
      "epoch": 0.4536764263812902,
      "flos": 66903161448960.0,
      "grad_norm": 0.8862425267990435,
      "language_loss": 0.56923795,
      "learning_rate": 2.394668141686667e-06,
      "loss": 0.59005618,
      "num_input_tokens_seen": 81249315,
      "step": 3773,
      "time_per_iteration": 3.270596742630005
    },
    {
      "auxiliary_loss_clip": 0.01239773,
      "auxiliary_loss_mlp": 0.01034375,
      "balance_loss_clip": 1.05283093,
      "balance_loss_mlp": 1.02689517,
      "epoch": 0.4537966692719293,
      "flos": 42739766254080.0,
      "grad_norm": 2.1108735666060894,
      "language_loss": 0.69795978,
      "learning_rate": 2.3939044575866813e-06,
      "loss": 0.72070134,
      "num_input_tokens_seen": 81272065,
      "step": 3774,
      "time_per_iteration": 2.843432664871216
    },
    {
      "auxiliary_loss_clip": 0.01289054,
      "auxiliary_loss_mlp": 0.02571807,
      "balance_loss_clip": 1.05097246,
      "balance_loss_mlp": 1.00015533,
      "epoch": 0.4539169121625684,
      "flos": 35549480517120.0,
      "grad_norm": 2.162069015810933,
      "language_loss": 0.75345063,
      "learning_rate": 2.3931407137317024e-06,
      "loss": 0.7920593,
      "num_input_tokens_seen": 81292220,
      "step": 3775,
      "time_per_iteration": 2.8259613513946533
    },
    {
      "auxiliary_loss_clip": 0.01241973,
      "auxiliary_loss_mlp": 0.01029348,
      "balance_loss_clip": 1.04690516,
      "balance_loss_mlp": 1.02187347,
      "epoch": 0.45403715505320746,
      "flos": 18514716341760.0,
      "grad_norm": 1.9653931279219832,
      "language_loss": 0.84858596,
      "learning_rate": 2.3923769102375907e-06,
      "loss": 0.87129915,
      "num_input_tokens_seen": 81311085,
      "step": 3776,
      "time_per_iteration": 3.652050495147705
    },
    {
      "auxiliary_loss_clip": 0.01346026,
      "auxiliary_loss_mlp": 0.01028336,
      "balance_loss_clip": 1.05186093,
      "balance_loss_mlp": 1.02016425,
      "epoch": 0.4541573979438466,
      "flos": 25045825639680.0,
      "grad_norm": 2.3976704416712358,
      "language_loss": 0.78881907,
      "learning_rate": 2.391613047220213e-06,
      "loss": 0.81256258,
      "num_input_tokens_seen": 81330985,
      "step": 3777,
      "time_per_iteration": 2.806893825531006
    },
    {
      "auxiliary_loss_clip": 0.01299572,
      "auxiliary_loss_mlp": 0.01033196,
      "balance_loss_clip": 1.04989171,
      "balance_loss_mlp": 1.02552533,
      "epoch": 0.4542776408344857,
      "flos": 18332397884160.0,
      "grad_norm": 1.82200292235073,
      "language_loss": 0.79150575,
      "learning_rate": 2.390849124795447e-06,
      "loss": 0.8148334,
      "num_input_tokens_seen": 81346985,
      "step": 3778,
      "time_per_iteration": 2.785508871078491
    },
    {
      "auxiliary_loss_clip": 0.0119102,
      "auxiliary_loss_mlp": 0.01028628,
      "balance_loss_clip": 1.05545187,
      "balance_loss_mlp": 1.02099824,
      "epoch": 0.45439788372512474,
      "flos": 20701173116160.0,
      "grad_norm": 2.4708814174936817,
      "language_loss": 0.84064639,
      "learning_rate": 2.3900851430791804e-06,
      "loss": 0.86284292,
      "num_input_tokens_seen": 81365005,
      "step": 3779,
      "time_per_iteration": 2.6401987075805664
    },
    {
      "auxiliary_loss_clip": 0.01195226,
      "auxiliary_loss_mlp": 0.0103863,
      "balance_loss_clip": 1.05603218,
      "balance_loss_mlp": 1.02990437,
      "epoch": 0.45451812661576385,
      "flos": 22309432663680.0,
      "grad_norm": 2.0466586828471334,
      "language_loss": 0.84908569,
      "learning_rate": 2.389321102187307e-06,
      "loss": 0.87142432,
      "num_input_tokens_seen": 81383785,
      "step": 3780,
      "time_per_iteration": 3.6507809162139893
    },
    {
      "auxiliary_loss_clip": 0.01288145,
      "auxiliary_loss_mlp": 0.02574227,
      "balance_loss_clip": 1.05324721,
      "balance_loss_mlp": 1.00027108,
      "epoch": 0.4546383695064029,
      "flos": 21763303303680.0,
      "grad_norm": 1.8438026485290797,
      "language_loss": 0.81710696,
      "learning_rate": 2.3885570022357326e-06,
      "loss": 0.85573071,
      "num_input_tokens_seen": 81402915,
      "step": 3781,
      "time_per_iteration": 2.7101633548736572
    },
    {
      "auxiliary_loss_clip": 0.01150222,
      "auxiliary_loss_mlp": 0.01005678,
      "balance_loss_clip": 1.01582134,
      "balance_loss_mlp": 1.00457513,
      "epoch": 0.454758612397042,
      "flos": 64242755694720.0,
      "grad_norm": 0.8204444831447317,
      "language_loss": 0.6084919,
      "learning_rate": 2.38779284334037e-06,
      "loss": 0.6300509,
      "num_input_tokens_seen": 81467890,
      "step": 3782,
      "time_per_iteration": 3.3596444129943848
    },
    {
      "auxiliary_loss_clip": 0.01377125,
      "auxiliary_loss_mlp": 0.01031706,
      "balance_loss_clip": 1.04406738,
      "balance_loss_mlp": 1.02365947,
      "epoch": 0.4548788552876811,
      "flos": 27304175485440.0,
      "grad_norm": 2.0444942985253163,
      "language_loss": 0.7899406,
      "learning_rate": 2.387028625617141e-06,
      "loss": 0.81402898,
      "num_input_tokens_seen": 81487105,
      "step": 3783,
      "time_per_iteration": 3.693798780441284
    },
    {
      "auxiliary_loss_clip": 0.01278999,
      "auxiliary_loss_mlp": 0.01027947,
      "balance_loss_clip": 1.04688954,
      "balance_loss_mlp": 1.02040982,
      "epoch": 0.4549990981783202,
      "flos": 22857142222080.0,
      "grad_norm": 2.0653990066218424,
      "language_loss": 0.8466332,
      "learning_rate": 2.3862643491819766e-06,
      "loss": 0.8697027,
      "num_input_tokens_seen": 81505670,
      "step": 3784,
      "time_per_iteration": 2.6826469898223877
    },
    {
      "auxiliary_loss_clip": 0.01236119,
      "auxiliary_loss_mlp": 0.01032077,
      "balance_loss_clip": 1.04859519,
      "balance_loss_mlp": 1.02412546,
      "epoch": 0.4551193410689593,
      "flos": 23258587599360.0,
      "grad_norm": 2.415012622731512,
      "language_loss": 0.84360826,
      "learning_rate": 2.3855000141508186e-06,
      "loss": 0.86629021,
      "num_input_tokens_seen": 81525825,
      "step": 3785,
      "time_per_iteration": 2.720057725906372
    },
    {
      "auxiliary_loss_clip": 0.01302892,
      "auxiliary_loss_mlp": 0.01027134,
      "balance_loss_clip": 1.05982351,
      "balance_loss_mlp": 1.01896811,
      "epoch": 0.4552395839595984,
      "flos": 20777519473920.0,
      "grad_norm": 2.2518827185403874,
      "language_loss": 0.84140873,
      "learning_rate": 2.3847356206396143e-06,
      "loss": 0.86470902,
      "num_input_tokens_seen": 81543135,
      "step": 3786,
      "time_per_iteration": 3.640944242477417
    },
    {
      "auxiliary_loss_clip": 0.0119059,
      "auxiliary_loss_mlp": 0.01032681,
      "balance_loss_clip": 1.05601668,
      "balance_loss_mlp": 1.02519441,
      "epoch": 0.45535982685023746,
      "flos": 23257510191360.0,
      "grad_norm": 1.8948190942857028,
      "language_loss": 0.78547728,
      "learning_rate": 2.3839711687643227e-06,
      "loss": 0.80771005,
      "num_input_tokens_seen": 81564360,
      "step": 3787,
      "time_per_iteration": 2.6233291625976562
    },
    {
      "auxiliary_loss_clip": 0.01240711,
      "auxiliary_loss_mlp": 0.01033166,
      "balance_loss_clip": 1.05318618,
      "balance_loss_mlp": 1.02493143,
      "epoch": 0.45548006974087657,
      "flos": 19646117907840.0,
      "grad_norm": 2.0201608715337005,
      "language_loss": 0.73555052,
      "learning_rate": 2.38320665864091e-06,
      "loss": 0.75828928,
      "num_input_tokens_seen": 81583710,
      "step": 3788,
      "time_per_iteration": 2.6667556762695312
    },
    {
      "auxiliary_loss_clip": 0.01438266,
      "auxiliary_loss_mlp": 0.01029151,
      "balance_loss_clip": 1.0442121,
      "balance_loss_mlp": 1.02117634,
      "epoch": 0.4556003126315157,
      "flos": 20047778766720.0,
      "grad_norm": 1.9653944428299872,
      "language_loss": 0.82559812,
      "learning_rate": 2.3824420903853516e-06,
      "loss": 0.85027224,
      "num_input_tokens_seen": 81602175,
      "step": 3789,
      "time_per_iteration": 2.7948105335235596
    },
    {
      "auxiliary_loss_clip": 0.01239507,
      "auxiliary_loss_mlp": 0.01030792,
      "balance_loss_clip": 1.05474234,
      "balance_loss_mlp": 1.02293599,
      "epoch": 0.45572055552215474,
      "flos": 22959738443520.0,
      "grad_norm": 2.2121125141569506,
      "language_loss": 0.81801242,
      "learning_rate": 2.3816774641136324e-06,
      "loss": 0.84071541,
      "num_input_tokens_seen": 81619430,
      "step": 3790,
      "time_per_iteration": 2.7031588554382324
    },
    {
      "auxiliary_loss_clip": 0.01239656,
      "auxiliary_loss_mlp": 0.02572735,
      "balance_loss_clip": 1.05416226,
      "balance_loss_mlp": 1.00019193,
      "epoch": 0.45584079841279385,
      "flos": 33109925535360.0,
      "grad_norm": 1.6245423809058133,
      "language_loss": 0.71372426,
      "learning_rate": 2.380912779941745e-06,
      "loss": 0.7518481,
      "num_input_tokens_seen": 81642550,
      "step": 3791,
      "time_per_iteration": 2.7770488262176514
    },
    {
      "auxiliary_loss_clip": 0.01242131,
      "auxiliary_loss_mlp": 0.01030578,
      "balance_loss_clip": 1.04954028,
      "balance_loss_mlp": 1.02223337,
      "epoch": 0.45596104130343296,
      "flos": 27272179445760.0,
      "grad_norm": 3.610844004318401,
      "language_loss": 0.8348943,
      "learning_rate": 2.3801480379856918e-06,
      "loss": 0.85762137,
      "num_input_tokens_seen": 81664260,
      "step": 3792,
      "time_per_iteration": 2.7507071495056152
    },
    {
      "auxiliary_loss_clip": 0.01288446,
      "auxiliary_loss_mlp": 0.01025813,
      "balance_loss_clip": 1.05233777,
      "balance_loss_mlp": 1.01806152,
      "epoch": 0.456081284194072,
      "flos": 21579799697280.0,
      "grad_norm": 2.647962130462285,
      "language_loss": 0.83737952,
      "learning_rate": 2.379383238361484e-06,
      "loss": 0.86052209,
      "num_input_tokens_seen": 81683620,
      "step": 3793,
      "time_per_iteration": 2.668191432952881
    },
    {
      "auxiliary_loss_clip": 0.0123489,
      "auxiliary_loss_mlp": 0.01022063,
      "balance_loss_clip": 1.05227828,
      "balance_loss_mlp": 1.01454079,
      "epoch": 0.4562015270847111,
      "flos": 35918822113920.0,
      "grad_norm": 1.7806433561384112,
      "language_loss": 0.79418778,
      "learning_rate": 2.3786183811851407e-06,
      "loss": 0.81675732,
      "num_input_tokens_seen": 81704325,
      "step": 3794,
      "time_per_iteration": 2.7531847953796387
    },
    {
      "auxiliary_loss_clip": 0.01194804,
      "auxiliary_loss_mlp": 0.01034277,
      "balance_loss_clip": 1.05826211,
      "balance_loss_mlp": 1.02596831,
      "epoch": 0.45632176997535023,
      "flos": 13589783602560.0,
      "grad_norm": 2.1113601466581042,
      "language_loss": 0.80348104,
      "learning_rate": 2.3778534665726892e-06,
      "loss": 0.82577187,
      "num_input_tokens_seen": 81721155,
      "step": 3795,
      "time_per_iteration": 2.56135630607605
    },
    {
      "auxiliary_loss_clip": 0.01233451,
      "auxiliary_loss_mlp": 0.01029862,
      "balance_loss_clip": 1.05144179,
      "balance_loss_mlp": 1.02203596,
      "epoch": 0.4564420128659893,
      "flos": 32635401937920.0,
      "grad_norm": 2.2324512098020604,
      "language_loss": 0.72790921,
      "learning_rate": 2.377088494640168e-06,
      "loss": 0.7505424,
      "num_input_tokens_seen": 81742905,
      "step": 3796,
      "time_per_iteration": 2.7323715686798096
    },
    {
      "auxiliary_loss_clip": 0.01237506,
      "auxiliary_loss_mlp": 0.01031243,
      "balance_loss_clip": 1.05449665,
      "balance_loss_mlp": 1.02391791,
      "epoch": 0.4565622557566284,
      "flos": 20377690208640.0,
      "grad_norm": 1.9540346161211581,
      "language_loss": 0.78030717,
      "learning_rate": 2.3763234655036216e-06,
      "loss": 0.80299473,
      "num_input_tokens_seen": 81762105,
      "step": 3797,
      "time_per_iteration": 2.683030128479004
    },
    {
      "auxiliary_loss_clip": 0.01341368,
      "auxiliary_loss_mlp": 0.01026846,
      "balance_loss_clip": 1.04600585,
      "balance_loss_mlp": 1.01935339,
      "epoch": 0.45668249864726745,
      "flos": 25374372364800.0,
      "grad_norm": 2.239520564353719,
      "language_loss": 0.87056434,
      "learning_rate": 2.3755583792791046e-06,
      "loss": 0.89424646,
      "num_input_tokens_seen": 81781975,
      "step": 3798,
      "time_per_iteration": 2.7555291652679443
    },
    {
      "auxiliary_loss_clip": 0.0124031,
      "auxiliary_loss_mlp": 0.01029923,
      "balance_loss_clip": 1.05193353,
      "balance_loss_mlp": 1.02197742,
      "epoch": 0.45680274153790656,
      "flos": 15559806977280.0,
      "grad_norm": 2.0792173679350903,
      "language_loss": 0.74881482,
      "learning_rate": 2.3747932360826803e-06,
      "loss": 0.77151716,
      "num_input_tokens_seen": 81798905,
      "step": 3799,
      "time_per_iteration": 2.650646924972534
    },
    {
      "auxiliary_loss_clip": 0.01238254,
      "auxiliary_loss_mlp": 0.01036126,
      "balance_loss_clip": 1.05324721,
      "balance_loss_mlp": 1.02773929,
      "epoch": 0.4569229844285457,
      "flos": 19792884879360.0,
      "grad_norm": 3.2662488937391143,
      "language_loss": 0.81705034,
      "learning_rate": 2.3740280360304205e-06,
      "loss": 0.8397941,
      "num_input_tokens_seen": 81816630,
      "step": 3800,
      "time_per_iteration": 2.6398158073425293
    },
    {
      "auxiliary_loss_clip": 0.01334947,
      "auxiliary_loss_mlp": 0.01029002,
      "balance_loss_clip": 1.05160809,
      "balance_loss_mlp": 1.02150416,
      "epoch": 0.45704322731918473,
      "flos": 24093941270400.0,
      "grad_norm": 7.082715003444192,
      "language_loss": 0.68177891,
      "learning_rate": 2.3732627792384038e-06,
      "loss": 0.70541841,
      "num_input_tokens_seen": 81837700,
      "step": 3801,
      "time_per_iteration": 2.74052095413208
    },
    {
      "auxiliary_loss_clip": 0.01192128,
      "auxiliary_loss_mlp": 0.01029416,
      "balance_loss_clip": 1.05408585,
      "balance_loss_mlp": 1.02112484,
      "epoch": 0.45716347020982384,
      "flos": 31317803245440.0,
      "grad_norm": 1.8609238395939316,
      "language_loss": 0.75474596,
      "learning_rate": 2.3724974658227207e-06,
      "loss": 0.77696133,
      "num_input_tokens_seen": 81858490,
      "step": 3802,
      "time_per_iteration": 2.6296122074127197
    },
    {
      "auxiliary_loss_clip": 0.01288963,
      "auxiliary_loss_mlp": 0.02572513,
      "balance_loss_clip": 1.05369163,
      "balance_loss_mlp": 1.00024033,
      "epoch": 0.45728371310046295,
      "flos": 26501392471680.0,
      "grad_norm": 2.0080662262903344,
      "language_loss": 0.71537262,
      "learning_rate": 2.3717320958994687e-06,
      "loss": 0.75398737,
      "num_input_tokens_seen": 81876050,
      "step": 3803,
      "time_per_iteration": 3.5845093727111816
    },
    {
      "auxiliary_loss_clip": 0.01341068,
      "auxiliary_loss_mlp": 0.01025639,
      "balance_loss_clip": 1.0442735,
      "balance_loss_mlp": 1.01806927,
      "epoch": 0.457403955991102,
      "flos": 17929408222080.0,
      "grad_norm": 2.648083261223988,
      "language_loss": 0.70697004,
      "learning_rate": 2.3709666695847534e-06,
      "loss": 0.73063707,
      "num_input_tokens_seen": 81894230,
      "step": 3804,
      "time_per_iteration": 2.668367385864258
    },
    {
      "auxiliary_loss_clip": 0.01384037,
      "auxiliary_loss_mlp": 0.01025961,
      "balance_loss_clip": 1.04565036,
      "balance_loss_mlp": 1.01827455,
      "epoch": 0.4575241988817411,
      "flos": 42230660837760.0,
      "grad_norm": 1.8976836174292617,
      "language_loss": 0.70081341,
      "learning_rate": 2.370201186994689e-06,
      "loss": 0.72491336,
      "num_input_tokens_seen": 81917915,
      "step": 3805,
      "time_per_iteration": 2.9069762229919434
    },
    {
      "auxiliary_loss_clip": 0.01282162,
      "auxiliary_loss_mlp": 0.01029195,
      "balance_loss_clip": 1.05015039,
      "balance_loss_mlp": 1.02154768,
      "epoch": 0.45764444177238023,
      "flos": 30117309868800.0,
      "grad_norm": 2.0271877913603875,
      "language_loss": 0.69914877,
      "learning_rate": 2.369435648245399e-06,
      "loss": 0.72226232,
      "num_input_tokens_seen": 81938130,
      "step": 3806,
      "time_per_iteration": 3.6675643920898438
    },
    {
      "auxiliary_loss_clip": 0.01289891,
      "auxiliary_loss_mlp": 0.01028367,
      "balance_loss_clip": 1.05225801,
      "balance_loss_mlp": 1.02069592,
      "epoch": 0.4577646846630193,
      "flos": 24060293205120.0,
      "grad_norm": 1.9119481201830448,
      "language_loss": 0.85318375,
      "learning_rate": 2.368670053453015e-06,
      "loss": 0.87636638,
      "num_input_tokens_seen": 81959820,
      "step": 3807,
      "time_per_iteration": 2.6565229892730713
    },
    {
      "auxiliary_loss_clip": 0.01246728,
      "auxiliary_loss_mlp": 0.01033751,
      "balance_loss_clip": 1.05562401,
      "balance_loss_mlp": 1.02545381,
      "epoch": 0.4578849275536584,
      "flos": 17418578952960.0,
      "grad_norm": 11.217891514699565,
      "language_loss": 0.74100953,
      "learning_rate": 2.3679044027336757e-06,
      "loss": 0.76381433,
      "num_input_tokens_seen": 81975710,
      "step": 3808,
      "time_per_iteration": 2.651259660720825
    },
    {
      "auxiliary_loss_clip": 0.01192039,
      "auxiliary_loss_mlp": 0.01033443,
      "balance_loss_clip": 1.05394518,
      "balance_loss_mlp": 1.02532446,
      "epoch": 0.4580051704442975,
      "flos": 13510169107200.0,
      "grad_norm": 2.530207605006795,
      "language_loss": 0.69434786,
      "learning_rate": 2.3671386962035326e-06,
      "loss": 0.71660268,
      "num_input_tokens_seen": 81993180,
      "step": 3809,
      "time_per_iteration": 2.605445146560669
    },
    {
      "auxiliary_loss_clip": 0.0124379,
      "auxiliary_loss_mlp": 0.0102653,
      "balance_loss_clip": 1.05482984,
      "balance_loss_mlp": 1.01863241,
      "epoch": 0.45812541333493656,
      "flos": 18037606965120.0,
      "grad_norm": 2.251672658932789,
      "language_loss": 0.68758178,
      "learning_rate": 2.3663729339787405e-06,
      "loss": 0.71028495,
      "num_input_tokens_seen": 82010115,
      "step": 3810,
      "time_per_iteration": 3.5178382396698
    },
    {
      "auxiliary_loss_clip": 0.01191454,
      "auxiliary_loss_mlp": 0.01034856,
      "balance_loss_clip": 1.05425,
      "balance_loss_mlp": 1.02710724,
      "epoch": 0.45824565622557567,
      "flos": 20222196232320.0,
      "grad_norm": 2.9737099364518675,
      "language_loss": 0.73558927,
      "learning_rate": 2.365607116175466e-06,
      "loss": 0.75785232,
      "num_input_tokens_seen": 82025540,
      "step": 3811,
      "time_per_iteration": 2.6345810890197754
    },
    {
      "auxiliary_loss_clip": 0.0118936,
      "auxiliary_loss_mlp": 0.01027827,
      "balance_loss_clip": 1.05395615,
      "balance_loss_mlp": 1.02044511,
      "epoch": 0.4583658991162148,
      "flos": 19864885691520.0,
      "grad_norm": 2.4001923545394446,
      "language_loss": 0.66926008,
      "learning_rate": 2.3648412429098825e-06,
      "loss": 0.691432,
      "num_input_tokens_seen": 82043890,
      "step": 3812,
      "time_per_iteration": 2.585380792617798
    },
    {
      "auxiliary_loss_clip": 0.01337742,
      "auxiliary_loss_mlp": 0.01028219,
      "balance_loss_clip": 1.04953599,
      "balance_loss_mlp": 1.01955211,
      "epoch": 0.45848614200685384,
      "flos": 21029935322880.0,
      "grad_norm": 1.9943718656519582,
      "language_loss": 0.81728518,
      "learning_rate": 2.364075314298172e-06,
      "loss": 0.84094483,
      "num_input_tokens_seen": 82061345,
      "step": 3813,
      "time_per_iteration": 3.64162015914917
    },
    {
      "auxiliary_loss_clip": 0.01244038,
      "auxiliary_loss_mlp": 0.02572066,
      "balance_loss_clip": 1.0525043,
      "balance_loss_mlp": 1.00015593,
      "epoch": 0.45860638489749295,
      "flos": 21069293650560.0,
      "grad_norm": 1.9566767249174601,
      "language_loss": 0.70559084,
      "learning_rate": 2.3633093304565267e-06,
      "loss": 0.74375188,
      "num_input_tokens_seen": 82080400,
      "step": 3814,
      "time_per_iteration": 2.6926941871643066
    },
    {
      "auxiliary_loss_clip": 0.0119563,
      "auxiliary_loss_mlp": 0.01029397,
      "balance_loss_clip": 1.05601478,
      "balance_loss_mlp": 1.02121913,
      "epoch": 0.458726627788132,
      "flos": 26833889692800.0,
      "grad_norm": 1.971035727839775,
      "language_loss": 0.62916529,
      "learning_rate": 2.3625432915011443e-06,
      "loss": 0.65141559,
      "num_input_tokens_seen": 82102310,
      "step": 3815,
      "time_per_iteration": 2.6404056549072266
    },
    {
      "auxiliary_loss_clip": 0.01284168,
      "auxiliary_loss_mlp": 0.01027928,
      "balance_loss_clip": 1.05022264,
      "balance_loss_mlp": 1.01932669,
      "epoch": 0.4588468706787711,
      "flos": 24097927680000.0,
      "grad_norm": 1.7610226979676078,
      "language_loss": 0.65285987,
      "learning_rate": 2.3617771975482334e-06,
      "loss": 0.67598081,
      "num_input_tokens_seen": 82121140,
      "step": 3816,
      "time_per_iteration": 2.680150032043457
    },
    {
      "auxiliary_loss_clip": 0.01385535,
      "auxiliary_loss_mlp": 0.01031582,
      "balance_loss_clip": 1.04764128,
      "balance_loss_mlp": 1.02426219,
      "epoch": 0.4589671135694102,
      "flos": 17889331622400.0,
      "grad_norm": 2.3390824718578656,
      "language_loss": 0.74648345,
      "learning_rate": 2.3610110487140083e-06,
      "loss": 0.77065462,
      "num_input_tokens_seen": 82139575,
      "step": 3817,
      "time_per_iteration": 2.879978656768799
    },
    {
      "auxiliary_loss_clip": 0.01294908,
      "auxiliary_loss_mlp": 0.01029884,
      "balance_loss_clip": 1.05451012,
      "balance_loss_mlp": 1.02167034,
      "epoch": 0.4590873564600493,
      "flos": 25626967781760.0,
      "grad_norm": 1.738336404596747,
      "language_loss": 0.80646324,
      "learning_rate": 2.360244845114695e-06,
      "loss": 0.8297112,
      "num_input_tokens_seen": 82159195,
      "step": 3818,
      "time_per_iteration": 2.844585418701172
    },
    {
      "auxiliary_loss_clip": 0.01290455,
      "auxiliary_loss_mlp": 0.01026389,
      "balance_loss_clip": 1.05674374,
      "balance_loss_mlp": 1.01845527,
      "epoch": 0.4592075993506884,
      "flos": 18514788168960.0,
      "grad_norm": 2.3622480972939677,
      "language_loss": 0.68030834,
      "learning_rate": 2.3594785868665245e-06,
      "loss": 0.70347679,
      "num_input_tokens_seen": 82175500,
      "step": 3819,
      "time_per_iteration": 2.6979055404663086
    },
    {
      "auxiliary_loss_clip": 0.01335443,
      "auxiliary_loss_mlp": 0.02571774,
      "balance_loss_clip": 1.04770648,
      "balance_loss_mlp": 1.00018692,
      "epoch": 0.4593278422413275,
      "flos": 20631111638400.0,
      "grad_norm": 2.1946609926955007,
      "language_loss": 0.80950904,
      "learning_rate": 2.3587122740857386e-06,
      "loss": 0.84858119,
      "num_input_tokens_seen": 82192600,
      "step": 3820,
      "time_per_iteration": 2.7742745876312256
    },
    {
      "auxiliary_loss_clip": 0.01236732,
      "auxiliary_loss_mlp": 0.01031149,
      "balance_loss_clip": 1.05138326,
      "balance_loss_mlp": 1.02316761,
      "epoch": 0.45944808513196655,
      "flos": 21358517961600.0,
      "grad_norm": 1.6766761926432985,
      "language_loss": 0.78185833,
      "learning_rate": 2.357945906888586e-06,
      "loss": 0.80453706,
      "num_input_tokens_seen": 82212040,
      "step": 3821,
      "time_per_iteration": 2.6121435165405273
    },
    {
      "auxiliary_loss_clip": 0.01241127,
      "auxiliary_loss_mlp": 0.01029232,
      "balance_loss_clip": 1.0538342,
      "balance_loss_mlp": 1.02045858,
      "epoch": 0.45956832802260567,
      "flos": 21427789340160.0,
      "grad_norm": 2.4003641969147163,
      "language_loss": 0.800978,
      "learning_rate": 2.357179485391324e-06,
      "loss": 0.82368159,
      "num_input_tokens_seen": 82229895,
      "step": 3822,
      "time_per_iteration": 2.6502604484558105
    },
    {
      "auxiliary_loss_clip": 0.01185681,
      "auxiliary_loss_mlp": 0.0102745,
      "balance_loss_clip": 1.0523448,
      "balance_loss_mlp": 1.02010357,
      "epoch": 0.4596885709132448,
      "flos": 22382654538240.0,
      "grad_norm": 1.899972200241331,
      "language_loss": 0.86098677,
      "learning_rate": 2.3564130097102173e-06,
      "loss": 0.88311803,
      "num_input_tokens_seen": 82249550,
      "step": 3823,
      "time_per_iteration": 2.5731513500213623
    },
    {
      "auxiliary_loss_clip": 0.01293767,
      "auxiliary_loss_mlp": 0.01035174,
      "balance_loss_clip": 1.05716705,
      "balance_loss_mlp": 1.02694845,
      "epoch": 0.45980881380388383,
      "flos": 28981957806720.0,
      "grad_norm": 1.7492497690841724,
      "language_loss": 0.75084633,
      "learning_rate": 2.355646479961541e-06,
      "loss": 0.77413571,
      "num_input_tokens_seen": 82268860,
      "step": 3824,
      "time_per_iteration": 2.7721428871154785
    },
    {
      "auxiliary_loss_clip": 0.01189392,
      "auxiliary_loss_mlp": 0.01031132,
      "balance_loss_clip": 1.05443263,
      "balance_loss_mlp": 1.02322865,
      "epoch": 0.45992905669452294,
      "flos": 33396599980800.0,
      "grad_norm": 1.9579369219528553,
      "language_loss": 0.71623933,
      "learning_rate": 2.354879896261576e-06,
      "loss": 0.73844457,
      "num_input_tokens_seen": 82289070,
      "step": 3825,
      "time_per_iteration": 2.6812431812286377
    },
    {
      "auxiliary_loss_clip": 0.01331431,
      "auxiliary_loss_mlp": 0.01032729,
      "balance_loss_clip": 1.05123973,
      "balance_loss_mlp": 1.02483165,
      "epoch": 0.46004929958516205,
      "flos": 36318184502400.0,
      "grad_norm": 2.676217698682711,
      "language_loss": 0.56904352,
      "learning_rate": 2.3541132587266133e-06,
      "loss": 0.5926851,
      "num_input_tokens_seen": 82311790,
      "step": 3826,
      "time_per_iteration": 2.8692526817321777
    },
    {
      "auxiliary_loss_clip": 0.01343162,
      "auxiliary_loss_mlp": 0.0102837,
      "balance_loss_clip": 1.05146766,
      "balance_loss_mlp": 1.02062702,
      "epoch": 0.4601695424758011,
      "flos": 17238451224960.0,
      "grad_norm": 2.6788369998359505,
      "language_loss": 0.69467956,
      "learning_rate": 2.3533465674729515e-06,
      "loss": 0.71839488,
      "num_input_tokens_seen": 82329020,
      "step": 3827,
      "time_per_iteration": 2.745424747467041
    },
    {
      "auxiliary_loss_clip": 0.01192441,
      "auxiliary_loss_mlp": 0.01027789,
      "balance_loss_clip": 1.05639625,
      "balance_loss_mlp": 1.01941478,
      "epoch": 0.4602897853664402,
      "flos": 15888425529600.0,
      "grad_norm": 1.894337244581451,
      "language_loss": 0.7312004,
      "learning_rate": 2.352579822616895e-06,
      "loss": 0.75340271,
      "num_input_tokens_seen": 82346455,
      "step": 3828,
      "time_per_iteration": 3.534074068069458
    },
    {
      "auxiliary_loss_clip": 0.01192552,
      "auxiliary_loss_mlp": 0.01027362,
      "balance_loss_clip": 1.05332279,
      "balance_loss_mlp": 1.01979828,
      "epoch": 0.4604100282570793,
      "flos": 25412617370880.0,
      "grad_norm": 1.8920544572995615,
      "language_loss": 0.777246,
      "learning_rate": 2.351813024274761e-06,
      "loss": 0.79944515,
      "num_input_tokens_seen": 82367810,
      "step": 3829,
      "time_per_iteration": 2.7319180965423584
    },
    {
      "auxiliary_loss_clip": 0.01341443,
      "auxiliary_loss_mlp": 0.01020772,
      "balance_loss_clip": 1.05235386,
      "balance_loss_mlp": 1.01285672,
      "epoch": 0.4605302711477184,
      "flos": 27630711048960.0,
      "grad_norm": 1.920987220726425,
      "language_loss": 0.73696518,
      "learning_rate": 2.3510461725628693e-06,
      "loss": 0.76058733,
      "num_input_tokens_seen": 82388275,
      "step": 3830,
      "time_per_iteration": 2.7346620559692383
    },
    {
      "auxiliary_loss_clip": 0.01338609,
      "auxiliary_loss_mlp": 0.01027695,
      "balance_loss_clip": 1.0490036,
      "balance_loss_mlp": 1.02019119,
      "epoch": 0.4606505140383575,
      "flos": 23839657914240.0,
      "grad_norm": 3.3960501003335546,
      "language_loss": 0.71244437,
      "learning_rate": 2.350279267597554e-06,
      "loss": 0.73610741,
      "num_input_tokens_seen": 82408915,
      "step": 3831,
      "time_per_iteration": 2.7440078258514404
    },
    {
      "auxiliary_loss_clip": 0.012409,
      "auxiliary_loss_mlp": 0.01029263,
      "balance_loss_clip": 1.05416906,
      "balance_loss_mlp": 1.02061987,
      "epoch": 0.46077075692899655,
      "flos": 16107013745280.0,
      "grad_norm": 2.5420020654804603,
      "language_loss": 0.82981455,
      "learning_rate": 2.3495123094951515e-06,
      "loss": 0.85251617,
      "num_input_tokens_seen": 82427260,
      "step": 3832,
      "time_per_iteration": 3.604879856109619
    },
    {
      "auxiliary_loss_clip": 0.0128251,
      "auxiliary_loss_mlp": 0.010304,
      "balance_loss_clip": 1.05017579,
      "balance_loss_mlp": 1.02262127,
      "epoch": 0.46089099981963566,
      "flos": 48798147634560.0,
      "grad_norm": 2.6648086171973686,
      "language_loss": 0.76215899,
      "learning_rate": 2.34874529837201e-06,
      "loss": 0.7852881,
      "num_input_tokens_seen": 82450805,
      "step": 3833,
      "time_per_iteration": 2.9201557636260986
    },
    {
      "auxiliary_loss_clip": 0.0143032,
      "auxiliary_loss_mlp": 0.01030507,
      "balance_loss_clip": 1.0441618,
      "balance_loss_mlp": 1.02197194,
      "epoch": 0.46101124271027477,
      "flos": 19099234362240.0,
      "grad_norm": 1.8672540102189963,
      "language_loss": 0.79092324,
      "learning_rate": 2.347978234344483e-06,
      "loss": 0.81553155,
      "num_input_tokens_seen": 82467010,
      "step": 3834,
      "time_per_iteration": 2.803262233734131
    },
    {
      "auxiliary_loss_clip": 0.01245801,
      "auxiliary_loss_mlp": 0.01031434,
      "balance_loss_clip": 1.05458593,
      "balance_loss_mlp": 1.02363205,
      "epoch": 0.4611314856009138,
      "flos": 39347931853440.0,
      "grad_norm": 1.9659964585638647,
      "language_loss": 0.68971902,
      "learning_rate": 2.347211117528935e-06,
      "loss": 0.71249139,
      "num_input_tokens_seen": 82489310,
      "step": 3835,
      "time_per_iteration": 3.6241583824157715
    },
    {
      "auxiliary_loss_clip": 0.0134866,
      "auxiliary_loss_mlp": 0.01029825,
      "balance_loss_clip": 1.05534399,
      "balance_loss_mlp": 1.02226686,
      "epoch": 0.46125172849155294,
      "flos": 20810772489600.0,
      "grad_norm": 2.8829025999502917,
      "language_loss": 0.7159692,
      "learning_rate": 2.3464439480417374e-06,
      "loss": 0.73975402,
      "num_input_tokens_seen": 82508830,
      "step": 3836,
      "time_per_iteration": 2.703158378601074
    },
    {
      "auxiliary_loss_clip": 0.0124189,
      "auxiliary_loss_mlp": 0.01024461,
      "balance_loss_clip": 1.05371499,
      "balance_loss_mlp": 1.01630092,
      "epoch": 0.46137197138219205,
      "flos": 17930808852480.0,
      "grad_norm": 15.635150617582998,
      "language_loss": 0.77053308,
      "learning_rate": 2.3456767259992676e-06,
      "loss": 0.79319656,
      "num_input_tokens_seen": 82526475,
      "step": 3837,
      "time_per_iteration": 2.6110827922821045
    },
    {
      "auxiliary_loss_clip": 0.01193844,
      "auxiliary_loss_mlp": 0.02568455,
      "balance_loss_clip": 1.05594635,
      "balance_loss_mlp": 1.00020957,
      "epoch": 0.4614922142728311,
      "flos": 16836610798080.0,
      "grad_norm": 2.3173641939712284,
      "language_loss": 0.8901853,
      "learning_rate": 2.3449094515179135e-06,
      "loss": 0.92780834,
      "num_input_tokens_seen": 82543935,
      "step": 3838,
      "time_per_iteration": 3.4762136936187744
    },
    {
      "auxiliary_loss_clip": 0.01293739,
      "auxiliary_loss_mlp": 0.0102611,
      "balance_loss_clip": 1.05033576,
      "balance_loss_mlp": 1.01845098,
      "epoch": 0.4616124571634702,
      "flos": 26614906427520.0,
      "grad_norm": 1.5480555633087338,
      "language_loss": 0.81836867,
      "learning_rate": 2.34414212471407e-06,
      "loss": 0.84156716,
      "num_input_tokens_seen": 82563730,
      "step": 3839,
      "time_per_iteration": 2.8071155548095703
    },
    {
      "auxiliary_loss_clip": 0.01247657,
      "auxiliary_loss_mlp": 0.01025039,
      "balance_loss_clip": 1.05414832,
      "balance_loss_mlp": 1.01689684,
      "epoch": 0.4617327000541093,
      "flos": 20340127560960.0,
      "grad_norm": 2.9636621252214206,
      "language_loss": 0.73047394,
      "learning_rate": 2.3433747457041394e-06,
      "loss": 0.75320089,
      "num_input_tokens_seen": 82582435,
      "step": 3840,
      "time_per_iteration": 2.6484642028808594
    },
    {
      "auxiliary_loss_clip": 0.01342393,
      "auxiliary_loss_mlp": 0.01033308,
      "balance_loss_clip": 1.05313444,
      "balance_loss_mlp": 1.02456975,
      "epoch": 0.4618529429447484,
      "flos": 29570749545600.0,
      "grad_norm": 6.568252629198071,
      "language_loss": 0.84850806,
      "learning_rate": 2.342607314604533e-06,
      "loss": 0.87226504,
      "num_input_tokens_seen": 82602185,
      "step": 3841,
      "time_per_iteration": 2.8291783332824707
    },
    {
      "auxiliary_loss_clip": 0.01245938,
      "auxiliary_loss_mlp": 0.01029858,
      "balance_loss_clip": 1.0587728,
      "balance_loss_mlp": 1.02181768,
      "epoch": 0.4619731858353875,
      "flos": 19787030962560.0,
      "grad_norm": 1.8522557388596232,
      "language_loss": 0.84383571,
      "learning_rate": 2.3418398315316694e-06,
      "loss": 0.86659372,
      "num_input_tokens_seen": 82620005,
      "step": 3842,
      "time_per_iteration": 2.6408703327178955
    },
    {
      "auxiliary_loss_clip": 0.01191556,
      "auxiliary_loss_mlp": 0.010346,
      "balance_loss_clip": 1.05556786,
      "balance_loss_mlp": 1.02697086,
      "epoch": 0.4620934287260266,
      "flos": 18951138587520.0,
      "grad_norm": 2.35199553352666,
      "language_loss": 0.78359401,
      "learning_rate": 2.3410722966019755e-06,
      "loss": 0.80585563,
      "num_input_tokens_seen": 82635120,
      "step": 3843,
      "time_per_iteration": 2.649674415588379
    },
    {
      "auxiliary_loss_clip": 0.01238289,
      "auxiliary_loss_mlp": 0.01030488,
      "balance_loss_clip": 1.05266333,
      "balance_loss_mlp": 1.02316308,
      "epoch": 0.46221367161666566,
      "flos": 37341674634240.0,
      "grad_norm": 1.8048458367479314,
      "language_loss": 0.65530419,
      "learning_rate": 2.3403047099318848e-06,
      "loss": 0.67799199,
      "num_input_tokens_seen": 82659190,
      "step": 3844,
      "time_per_iteration": 2.7783634662628174
    },
    {
      "auxiliary_loss_clip": 0.0137982,
      "auxiliary_loss_mlp": 0.01032544,
      "balance_loss_clip": 1.04550195,
      "balance_loss_mlp": 1.02492619,
      "epoch": 0.46233391450730477,
      "flos": 14428549065600.0,
      "grad_norm": 2.9047903771485366,
      "language_loss": 0.75847697,
      "learning_rate": 2.3395370716378405e-06,
      "loss": 0.78260064,
      "num_input_tokens_seen": 82676635,
      "step": 3845,
      "time_per_iteration": 2.7800769805908203
    },
    {
      "auxiliary_loss_clip": 0.01243597,
      "auxiliary_loss_mlp": 0.01031303,
      "balance_loss_clip": 1.05370629,
      "balance_loss_mlp": 1.02418351,
      "epoch": 0.4624541573979438,
      "flos": 22493044010880.0,
      "grad_norm": 2.0965437585827775,
      "language_loss": 0.7221607,
      "learning_rate": 2.338769381836292e-06,
      "loss": 0.74490964,
      "num_input_tokens_seen": 82696245,
      "step": 3846,
      "time_per_iteration": 2.6813509464263916
    },
    {
      "auxiliary_loss_clip": 0.01342958,
      "auxiliary_loss_mlp": 0.01030042,
      "balance_loss_clip": 1.05467081,
      "balance_loss_mlp": 1.02202797,
      "epoch": 0.46257440028858293,
      "flos": 14465070218880.0,
      "grad_norm": 1.984260830798692,
      "language_loss": 0.73034894,
      "learning_rate": 2.3380016406436984e-06,
      "loss": 0.75407898,
      "num_input_tokens_seen": 82713725,
      "step": 3847,
      "time_per_iteration": 2.6821343898773193
    },
    {
      "auxiliary_loss_clip": 0.01390271,
      "auxiliary_loss_mlp": 0.01031649,
      "balance_loss_clip": 1.05113637,
      "balance_loss_mlp": 1.02381134,
      "epoch": 0.46269464317922204,
      "flos": 23332204523520.0,
      "grad_norm": 2.01460061179832,
      "language_loss": 0.81514555,
      "learning_rate": 2.337233848176524e-06,
      "loss": 0.83936477,
      "num_input_tokens_seen": 82731495,
      "step": 3848,
      "time_per_iteration": 2.879979372024536
    },
    {
      "auxiliary_loss_clip": 0.01383794,
      "auxiliary_loss_mlp": 0.01031887,
      "balance_loss_clip": 1.04712367,
      "balance_loss_mlp": 1.02354813,
      "epoch": 0.4628148860698611,
      "flos": 18552027594240.0,
      "grad_norm": 2.0361011978829855,
      "language_loss": 0.83033037,
      "learning_rate": 2.3364660045512435e-06,
      "loss": 0.85448718,
      "num_input_tokens_seen": 82750255,
      "step": 3849,
      "time_per_iteration": 2.664414405822754
    },
    {
      "auxiliary_loss_clip": 0.01185461,
      "auxiliary_loss_mlp": 0.0100237,
      "balance_loss_clip": 1.02154899,
      "balance_loss_mlp": 1.0013566,
      "epoch": 0.4629351289605002,
      "flos": 70667569670400.0,
      "grad_norm": 0.7486996006766862,
      "language_loss": 0.58196676,
      "learning_rate": 2.335698109884337e-06,
      "loss": 0.60384512,
      "num_input_tokens_seen": 82815460,
      "step": 3850,
      "time_per_iteration": 3.4197018146514893
    },
    {
      "auxiliary_loss_clip": 0.01313312,
      "auxiliary_loss_mlp": 0.01019517,
      "balance_loss_clip": 1.03424931,
      "balance_loss_mlp": 1.01840806,
      "epoch": 0.4630553718511393,
      "flos": 59687200465920.0,
      "grad_norm": 0.7892008360447,
      "language_loss": 0.59867829,
      "learning_rate": 2.334930164292294e-06,
      "loss": 0.6220066,
      "num_input_tokens_seen": 82878010,
      "step": 3851,
      "time_per_iteration": 3.5182158946990967
    },
    {
      "auxiliary_loss_clip": 0.01385191,
      "auxiliary_loss_mlp": 0.01025816,
      "balance_loss_clip": 1.04520607,
      "balance_loss_mlp": 1.01804948,
      "epoch": 0.4631756147417784,
      "flos": 15960605909760.0,
      "grad_norm": 1.8486022245140306,
      "language_loss": 0.79701912,
      "learning_rate": 2.334162167891612e-06,
      "loss": 0.8211292,
      "num_input_tokens_seen": 82895275,
      "step": 3852,
      "time_per_iteration": 2.920813798904419
    },
    {
      "auxiliary_loss_clip": 0.01295922,
      "auxiliary_loss_mlp": 0.0102781,
      "balance_loss_clip": 1.05124497,
      "balance_loss_mlp": 1.01997828,
      "epoch": 0.4632958576324175,
      "flos": 16472907636480.0,
      "grad_norm": 2.0389089370066094,
      "language_loss": 0.74874353,
      "learning_rate": 2.333394120798795e-06,
      "loss": 0.77198076,
      "num_input_tokens_seen": 82914010,
      "step": 3853,
      "time_per_iteration": 2.6729819774627686
    },
    {
      "auxiliary_loss_clip": 0.01289979,
      "auxiliary_loss_mlp": 0.01033267,
      "balance_loss_clip": 1.05015373,
      "balance_loss_mlp": 1.02538109,
      "epoch": 0.4634161005230566,
      "flos": 22346492520960.0,
      "grad_norm": 2.1931798317197866,
      "language_loss": 0.71918738,
      "learning_rate": 2.3326260231303545e-06,
      "loss": 0.74241984,
      "num_input_tokens_seen": 82932610,
      "step": 3854,
      "time_per_iteration": 3.8385283946990967
    },
    {
      "auxiliary_loss_clip": 0.01188669,
      "auxiliary_loss_mlp": 0.01023279,
      "balance_loss_clip": 1.05504,
      "balance_loss_mlp": 1.01595962,
      "epoch": 0.46353634341369565,
      "flos": 15742233175680.0,
      "grad_norm": 1.667287444767265,
      "language_loss": 0.87012988,
      "learning_rate": 2.331857875002811e-06,
      "loss": 0.89224935,
      "num_input_tokens_seen": 82951210,
      "step": 3855,
      "time_per_iteration": 2.603156805038452
    },
    {
      "auxiliary_loss_clip": 0.0128935,
      "auxiliary_loss_mlp": 0.0102733,
      "balance_loss_clip": 1.0536921,
      "balance_loss_mlp": 1.01982856,
      "epoch": 0.46365658630433476,
      "flos": 28329820433280.0,
      "grad_norm": 2.7357952467480193,
      "language_loss": 0.76256466,
      "learning_rate": 2.3310896765326916e-06,
      "loss": 0.78573143,
      "num_input_tokens_seen": 82972210,
      "step": 3856,
      "time_per_iteration": 2.7322206497192383
    },
    {
      "auxiliary_loss_clip": 0.0134273,
      "auxiliary_loss_mlp": 0.01030812,
      "balance_loss_clip": 1.05188942,
      "balance_loss_mlp": 1.02295017,
      "epoch": 0.46377682919497387,
      "flos": 24608074590720.0,
      "grad_norm": 1.628071144377028,
      "language_loss": 0.84261227,
      "learning_rate": 2.330321427836531e-06,
      "loss": 0.86634767,
      "num_input_tokens_seen": 82994080,
      "step": 3857,
      "time_per_iteration": 2.7455902099609375
    },
    {
      "auxiliary_loss_clip": 0.01238758,
      "auxiliary_loss_mlp": 0.01030552,
      "balance_loss_clip": 1.05294752,
      "balance_loss_mlp": 1.0230478,
      "epoch": 0.4638970720856129,
      "flos": 19060953442560.0,
      "grad_norm": 4.937000546848872,
      "language_loss": 0.82667112,
      "learning_rate": 2.3295531290308733e-06,
      "loss": 0.84936416,
      "num_input_tokens_seen": 83012230,
      "step": 3858,
      "time_per_iteration": 3.5424699783325195
    },
    {
      "auxiliary_loss_clip": 0.01195501,
      "auxiliary_loss_mlp": 0.02571163,
      "balance_loss_clip": 1.05774343,
      "balance_loss_mlp": 1.00020623,
      "epoch": 0.46401731497625204,
      "flos": 18471012468480.0,
      "grad_norm": 3.1475662327111453,
      "language_loss": 0.75693989,
      "learning_rate": 2.3287847802322678e-06,
      "loss": 0.79460651,
      "num_input_tokens_seen": 83027800,
      "step": 3859,
      "time_per_iteration": 2.5762007236480713
    },
    {
      "auxiliary_loss_clip": 0.01207859,
      "auxiliary_loss_mlp": 0.01028746,
      "balance_loss_clip": 1.05828667,
      "balance_loss_mlp": 1.0202527,
      "epoch": 0.4641375578668911,
      "flos": 26067053214720.0,
      "grad_norm": 2.0983025961162025,
      "language_loss": 0.83909732,
      "learning_rate": 2.3280163815572723e-06,
      "loss": 0.86146343,
      "num_input_tokens_seen": 83048395,
      "step": 3860,
      "time_per_iteration": 2.719430923461914
    },
    {
      "auxiliary_loss_clip": 0.01284345,
      "auxiliary_loss_mlp": 0.01033321,
      "balance_loss_clip": 1.05091989,
      "balance_loss_mlp": 1.0253582,
      "epoch": 0.4642578007575302,
      "flos": 19570382081280.0,
      "grad_norm": 2.9309296954607698,
      "language_loss": 0.76878822,
      "learning_rate": 2.3272479331224522e-06,
      "loss": 0.79196483,
      "num_input_tokens_seen": 83065825,
      "step": 3861,
      "time_per_iteration": 3.6008126735687256
    },
    {
      "auxiliary_loss_clip": 0.01192302,
      "auxiliary_loss_mlp": 0.01032666,
      "balance_loss_clip": 1.05543804,
      "balance_loss_mlp": 1.02458394,
      "epoch": 0.4643780436481693,
      "flos": 28186249772160.0,
      "grad_norm": 1.8485893133633902,
      "language_loss": 0.78107607,
      "learning_rate": 2.3264794350443817e-06,
      "loss": 0.80332571,
      "num_input_tokens_seen": 83087920,
      "step": 3862,
      "time_per_iteration": 2.6886115074157715
    },
    {
      "auxiliary_loss_clip": 0.01238715,
      "auxiliary_loss_mlp": 0.01031053,
      "balance_loss_clip": 1.05025005,
      "balance_loss_mlp": 1.02311397,
      "epoch": 0.46449828653880837,
      "flos": 25375270204800.0,
      "grad_norm": 1.85439789290648,
      "language_loss": 0.7869975,
      "learning_rate": 2.3257108874396396e-06,
      "loss": 0.80969518,
      "num_input_tokens_seen": 83109015,
      "step": 3863,
      "time_per_iteration": 2.6694624423980713
    },
    {
      "auxiliary_loss_clip": 0.01288811,
      "auxiliary_loss_mlp": 0.01032088,
      "balance_loss_clip": 1.05049086,
      "balance_loss_mlp": 1.02510548,
      "epoch": 0.4646185294294475,
      "flos": 16034330574720.0,
      "grad_norm": 2.171391523733925,
      "language_loss": 0.73928106,
      "learning_rate": 2.3249422904248152e-06,
      "loss": 0.76249003,
      "num_input_tokens_seen": 83127450,
      "step": 3864,
      "time_per_iteration": 3.5090813636779785
    },
    {
      "auxiliary_loss_clip": 0.01243684,
      "auxiliary_loss_mlp": 0.0102784,
      "balance_loss_clip": 1.05290926,
      "balance_loss_mlp": 1.01982927,
      "epoch": 0.4647387723200866,
      "flos": 26363101109760.0,
      "grad_norm": 1.5285880639697718,
      "language_loss": 0.87379962,
      "learning_rate": 2.324173644116504e-06,
      "loss": 0.89651489,
      "num_input_tokens_seen": 83150300,
      "step": 3865,
      "time_per_iteration": 2.758568286895752
    },
    {
      "auxiliary_loss_clip": 0.01237388,
      "auxiliary_loss_mlp": 0.0102636,
      "balance_loss_clip": 1.05519736,
      "balance_loss_mlp": 1.01827729,
      "epoch": 0.46485901521072565,
      "flos": 27160209774720.0,
      "grad_norm": 1.9221556911853304,
      "language_loss": 0.81545585,
      "learning_rate": 2.3234049486313087e-06,
      "loss": 0.83809328,
      "num_input_tokens_seen": 83171750,
      "step": 3866,
      "time_per_iteration": 2.6738998889923096
    },
    {
      "auxiliary_loss_clip": 0.01238708,
      "auxiliary_loss_mlp": 0.01029949,
      "balance_loss_clip": 1.05396533,
      "balance_loss_mlp": 1.02279902,
      "epoch": 0.46497925810136476,
      "flos": 24279851088000.0,
      "grad_norm": 1.7699653322686977,
      "language_loss": 0.76188815,
      "learning_rate": 2.322636204085839e-06,
      "loss": 0.78457475,
      "num_input_tokens_seen": 83191820,
      "step": 3867,
      "time_per_iteration": 2.6841983795166016
    },
    {
      "auxiliary_loss_clip": 0.0128318,
      "auxiliary_loss_mlp": 0.010329,
      "balance_loss_clip": 1.04678071,
      "balance_loss_mlp": 1.02472484,
      "epoch": 0.46509950099200387,
      "flos": 16253134272000.0,
      "grad_norm": 3.3241189315631283,
      "language_loss": 0.78545845,
      "learning_rate": 2.3218674105967143e-06,
      "loss": 0.8086192,
      "num_input_tokens_seen": 83210085,
      "step": 3868,
      "time_per_iteration": 2.6614644527435303
    },
    {
      "auxiliary_loss_clip": 0.01285479,
      "auxiliary_loss_mlp": 0.0103073,
      "balance_loss_clip": 1.05003321,
      "balance_loss_mlp": 1.02368522,
      "epoch": 0.4652197438826429,
      "flos": 23442270773760.0,
      "grad_norm": 2.7076702555854495,
      "language_loss": 0.83645105,
      "learning_rate": 2.3210985682805593e-06,
      "loss": 0.85961318,
      "num_input_tokens_seen": 83231865,
      "step": 3869,
      "time_per_iteration": 2.78041934967041
    },
    {
      "auxiliary_loss_clip": 0.01195687,
      "auxiliary_loss_mlp": 0.01029542,
      "balance_loss_clip": 1.05923581,
      "balance_loss_mlp": 1.02192175,
      "epoch": 0.46533998677328203,
      "flos": 16216397637120.0,
      "grad_norm": 4.629587961674402,
      "language_loss": 0.68059623,
      "learning_rate": 2.320329677254007e-06,
      "loss": 0.70284849,
      "num_input_tokens_seen": 83249195,
      "step": 3870,
      "time_per_iteration": 2.6413164138793945
    },
    {
      "auxiliary_loss_clip": 0.01190632,
      "auxiliary_loss_mlp": 0.01028145,
      "balance_loss_clip": 1.05544472,
      "balance_loss_mlp": 1.02018523,
      "epoch": 0.46546022966392114,
      "flos": 21141869080320.0,
      "grad_norm": 2.4077953343932315,
      "language_loss": 0.73006856,
      "learning_rate": 2.319560737633697e-06,
      "loss": 0.75225633,
      "num_input_tokens_seen": 83267915,
      "step": 3871,
      "time_per_iteration": 2.608645439147949
    },
    {
      "auxiliary_loss_clip": 0.01341945,
      "auxiliary_loss_mlp": 0.01027503,
      "balance_loss_clip": 1.04761922,
      "balance_loss_mlp": 1.01966524,
      "epoch": 0.4655804725545602,
      "flos": 41171942442240.0,
      "grad_norm": 1.5565961237058026,
      "language_loss": 0.68239713,
      "learning_rate": 2.3187917495362775e-06,
      "loss": 0.70609164,
      "num_input_tokens_seen": 83292325,
      "step": 3872,
      "time_per_iteration": 2.9852609634399414
    },
    {
      "auxiliary_loss_clip": 0.01383008,
      "auxiliary_loss_mlp": 0.01032713,
      "balance_loss_clip": 1.04716206,
      "balance_loss_mlp": 1.02509844,
      "epoch": 0.4657007154451993,
      "flos": 19570956698880.0,
      "grad_norm": 2.5247329692906884,
      "language_loss": 0.77067667,
      "learning_rate": 2.318022713078403e-06,
      "loss": 0.79483384,
      "num_input_tokens_seen": 83306905,
      "step": 3873,
      "time_per_iteration": 2.702501058578491
    },
    {
      "auxiliary_loss_clip": 0.01288365,
      "auxiliary_loss_mlp": 0.0102671,
      "balance_loss_clip": 1.05135345,
      "balance_loss_mlp": 1.01914024,
      "epoch": 0.4658209583358384,
      "flos": 15517826956800.0,
      "grad_norm": 4.107848349146468,
      "language_loss": 0.85276395,
      "learning_rate": 2.3172536283767354e-06,
      "loss": 0.87591469,
      "num_input_tokens_seen": 83320665,
      "step": 3874,
      "time_per_iteration": 2.598076581954956
    },
    {
      "auxiliary_loss_clip": 0.01336223,
      "auxiliary_loss_mlp": 0.01029097,
      "balance_loss_clip": 1.05101824,
      "balance_loss_mlp": 1.02191448,
      "epoch": 0.4659412012264775,
      "flos": 14903180403840.0,
      "grad_norm": 2.1561596720107854,
      "language_loss": 0.80348647,
      "learning_rate": 2.3164844955479447e-06,
      "loss": 0.82713968,
      "num_input_tokens_seen": 83336475,
      "step": 3875,
      "time_per_iteration": 2.616000175476074
    },
    {
      "auxiliary_loss_clip": 0.01299993,
      "auxiliary_loss_mlp": 0.01031975,
      "balance_loss_clip": 1.04999661,
      "balance_loss_mlp": 1.02444065,
      "epoch": 0.4660614441171166,
      "flos": 24425612478720.0,
      "grad_norm": 2.0230906597644274,
      "language_loss": 0.70597959,
      "learning_rate": 2.3157153147087082e-06,
      "loss": 0.72929925,
      "num_input_tokens_seen": 83358365,
      "step": 3876,
      "time_per_iteration": 2.8357338905334473
    },
    {
      "auxiliary_loss_clip": 0.01295985,
      "auxiliary_loss_mlp": 0.01025678,
      "balance_loss_clip": 1.05066943,
      "balance_loss_mlp": 1.01872826,
      "epoch": 0.46618168700775564,
      "flos": 22091095843200.0,
      "grad_norm": 1.743677653916453,
      "language_loss": 0.83377421,
      "learning_rate": 2.314946085975709e-06,
      "loss": 0.85699081,
      "num_input_tokens_seen": 83377345,
      "step": 3877,
      "time_per_iteration": 2.7015841007232666
    },
    {
      "auxiliary_loss_clip": 0.01327354,
      "auxiliary_loss_mlp": 0.01025883,
      "balance_loss_clip": 1.04807866,
      "balance_loss_mlp": 1.01890326,
      "epoch": 0.46630192989839475,
      "flos": 26176975810560.0,
      "grad_norm": 6.416994234632226,
      "language_loss": 0.82155848,
      "learning_rate": 2.3141768094656393e-06,
      "loss": 0.84509087,
      "num_input_tokens_seen": 83395920,
      "step": 3878,
      "time_per_iteration": 2.721611976623535
    },
    {
      "auxiliary_loss_clip": 0.01491748,
      "auxiliary_loss_mlp": 0.01028202,
      "balance_loss_clip": 1.04491878,
      "balance_loss_mlp": 1.02118099,
      "epoch": 0.46642217278903386,
      "flos": 11509622150400.0,
      "grad_norm": 5.994106248219784,
      "language_loss": 0.82807565,
      "learning_rate": 2.3134074852951966e-06,
      "loss": 0.85327512,
      "num_input_tokens_seen": 83412510,
      "step": 3879,
      "time_per_iteration": 2.8420121669769287
    },
    {
      "auxiliary_loss_clip": 0.01384264,
      "auxiliary_loss_mlp": 0.01034234,
      "balance_loss_clip": 1.04529953,
      "balance_loss_mlp": 1.02653313,
      "epoch": 0.4665424156796729,
      "flos": 32306819299200.0,
      "grad_norm": 2.306896696275707,
      "language_loss": 0.78073144,
      "learning_rate": 2.312638113581088e-06,
      "loss": 0.80491644,
      "num_input_tokens_seen": 83432995,
      "step": 3880,
      "time_per_iteration": 3.2542166709899902
    },
    {
      "auxiliary_loss_clip": 0.01237385,
      "auxiliary_loss_mlp": 0.01026751,
      "balance_loss_clip": 1.05006003,
      "balance_loss_mlp": 1.01860917,
      "epoch": 0.46666265857031203,
      "flos": 18436179254400.0,
      "grad_norm": 2.898512923291977,
      "language_loss": 0.78067297,
      "learning_rate": 2.311868694440027e-06,
      "loss": 0.80331433,
      "num_input_tokens_seen": 83447415,
      "step": 3881,
      "time_per_iteration": 3.666517496109009
    },
    {
      "auxiliary_loss_clip": 0.01084922,
      "auxiliary_loss_mlp": 0.01004907,
      "balance_loss_clip": 1.02116275,
      "balance_loss_mlp": 1.00382257,
      "epoch": 0.46678290146095114,
      "flos": 68438989221120.0,
      "grad_norm": 0.7478051995290669,
      "language_loss": 0.62397528,
      "learning_rate": 2.3110992279887323e-06,
      "loss": 0.6448735,
      "num_input_tokens_seen": 83519340,
      "step": 3882,
      "time_per_iteration": 3.356684684753418
    },
    {
      "auxiliary_loss_clip": 0.01340633,
      "auxiliary_loss_mlp": 0.01025102,
      "balance_loss_clip": 1.05046868,
      "balance_loss_mlp": 1.01678109,
      "epoch": 0.4669031443515902,
      "flos": 17712507945600.0,
      "grad_norm": 2.5107467784915443,
      "language_loss": 0.85553062,
      "learning_rate": 2.310329714343932e-06,
      "loss": 0.879188,
      "num_input_tokens_seen": 83535490,
      "step": 3883,
      "time_per_iteration": 2.6964762210845947
    },
    {
      "auxiliary_loss_clip": 0.01286131,
      "auxiliary_loss_mlp": 0.01032072,
      "balance_loss_clip": 1.05197501,
      "balance_loss_mlp": 1.0245558,
      "epoch": 0.4670233872422293,
      "flos": 23947748916480.0,
      "grad_norm": 1.9243652318358317,
      "language_loss": 0.81725645,
      "learning_rate": 2.309560153622361e-06,
      "loss": 0.84043849,
      "num_input_tokens_seen": 83552400,
      "step": 3884,
      "time_per_iteration": 3.6289138793945312
    },
    {
      "auxiliary_loss_clip": 0.01334686,
      "auxiliary_loss_mlp": 0.01036254,
      "balance_loss_clip": 1.05118823,
      "balance_loss_mlp": 1.02808785,
      "epoch": 0.4671436301328684,
      "flos": 28111268131200.0,
      "grad_norm": 2.483906238663339,
      "language_loss": 0.74869001,
      "learning_rate": 2.3087905459407602e-06,
      "loss": 0.77239943,
      "num_input_tokens_seen": 83571340,
      "step": 3885,
      "time_per_iteration": 2.728893995285034
    },
    {
      "auxiliary_loss_clip": 0.01039317,
      "auxiliary_loss_mlp": 0.00999144,
      "balance_loss_clip": 1.01983309,
      "balance_loss_mlp": 0.9981184,
      "epoch": 0.46726387302350747,
      "flos": 69369684566400.0,
      "grad_norm": 0.7873649697721482,
      "language_loss": 0.62854487,
      "learning_rate": 2.3080208914158795e-06,
      "loss": 0.64892942,
      "num_input_tokens_seen": 83634340,
      "step": 3886,
      "time_per_iteration": 4.114591121673584
    },
    {
      "auxiliary_loss_clip": 0.01284294,
      "auxiliary_loss_mlp": 0.01020969,
      "balance_loss_clip": 1.0538063,
      "balance_loss_mlp": 1.0134083,
      "epoch": 0.4673841159141466,
      "flos": 25519666878720.0,
      "grad_norm": 2.252387317682799,
      "language_loss": 0.71942103,
      "learning_rate": 2.3072511901644753e-06,
      "loss": 0.7424736,
      "num_input_tokens_seen": 83653410,
      "step": 3887,
      "time_per_iteration": 2.638237714767456
    },
    {
      "auxiliary_loss_clip": 0.01187938,
      "auxiliary_loss_mlp": 0.01028839,
      "balance_loss_clip": 1.05504048,
      "balance_loss_mlp": 1.02141237,
      "epoch": 0.4675043588047857,
      "flos": 24499265316480.0,
      "grad_norm": 4.248376903970199,
      "language_loss": 0.80380124,
      "learning_rate": 2.306481442303309e-06,
      "loss": 0.82596898,
      "num_input_tokens_seen": 83672985,
      "step": 3888,
      "time_per_iteration": 2.528179883956909
    },
    {
      "auxiliary_loss_clip": 0.01239482,
      "auxiliary_loss_mlp": 0.01030722,
      "balance_loss_clip": 1.05370271,
      "balance_loss_mlp": 1.02265763,
      "epoch": 0.46762460169542475,
      "flos": 20960771685120.0,
      "grad_norm": 2.1150221966962386,
      "language_loss": 0.73359996,
      "learning_rate": 2.3057116479491515e-06,
      "loss": 0.756302,
      "num_input_tokens_seen": 83692395,
      "step": 3889,
      "time_per_iteration": 2.530686616897583
    },
    {
      "auxiliary_loss_clip": 0.01235788,
      "auxiliary_loss_mlp": 0.01022515,
      "balance_loss_clip": 1.04960752,
      "balance_loss_mlp": 1.01477242,
      "epoch": 0.46774484458606386,
      "flos": 19171666137600.0,
      "grad_norm": 2.1420045730883506,
      "language_loss": 0.75860715,
      "learning_rate": 2.30494180721878e-06,
      "loss": 0.78119016,
      "num_input_tokens_seen": 83709735,
      "step": 3890,
      "time_per_iteration": 3.4020655155181885
    },
    {
      "auxiliary_loss_clip": 0.01235641,
      "auxiliary_loss_mlp": 0.01024363,
      "balance_loss_clip": 1.05013847,
      "balance_loss_mlp": 1.01726127,
      "epoch": 0.4678650874767029,
      "flos": 17967689141760.0,
      "grad_norm": 2.375298980528182,
      "language_loss": 0.90315241,
      "learning_rate": 2.3041719202289794e-06,
      "loss": 0.9257524,
      "num_input_tokens_seen": 83725910,
      "step": 3891,
      "time_per_iteration": 2.6203625202178955
    },
    {
      "auxiliary_loss_clip": 0.01242271,
      "auxiliary_loss_mlp": 0.01024031,
      "balance_loss_clip": 1.05561066,
      "balance_loss_mlp": 1.01689923,
      "epoch": 0.467985330367342,
      "flos": 21360816432000.0,
      "grad_norm": 1.9127098147427022,
      "language_loss": 0.80433381,
      "learning_rate": 2.30340198709654e-06,
      "loss": 0.8269968,
      "num_input_tokens_seen": 83745745,
      "step": 3892,
      "time_per_iteration": 2.6290643215179443
    },
    {
      "auxiliary_loss_clip": 0.01292987,
      "auxiliary_loss_mlp": 0.01028502,
      "balance_loss_clip": 1.04982877,
      "balance_loss_mlp": 1.0212903,
      "epoch": 0.46810557325798113,
      "flos": 20521835487360.0,
      "grad_norm": 2.1037873169842927,
      "language_loss": 0.74612689,
      "learning_rate": 2.3026320079382605e-06,
      "loss": 0.76934183,
      "num_input_tokens_seen": 83762680,
      "step": 3893,
      "time_per_iteration": 2.7236270904541016
    },
    {
      "auxiliary_loss_clip": 0.01188937,
      "auxiliary_loss_mlp": 0.01029152,
      "balance_loss_clip": 1.055004,
      "balance_loss_mlp": 1.0217011,
      "epoch": 0.4682258161486202,
      "flos": 30117848572800.0,
      "grad_norm": 2.990807685843572,
      "language_loss": 0.76568913,
      "learning_rate": 2.3018619828709454e-06,
      "loss": 0.78787005,
      "num_input_tokens_seen": 83784220,
      "step": 3894,
      "time_per_iteration": 2.758061647415161
    },
    {
      "auxiliary_loss_clip": 0.01233408,
      "auxiliary_loss_mlp": 0.02567574,
      "balance_loss_clip": 1.05220079,
      "balance_loss_mlp": 1.00017691,
      "epoch": 0.4683460590392593,
      "flos": 25293357239040.0,
      "grad_norm": 1.8623998526309142,
      "language_loss": 0.81980181,
      "learning_rate": 2.3010919120114084e-06,
      "loss": 0.85781163,
      "num_input_tokens_seen": 83800750,
      "step": 3895,
      "time_per_iteration": 2.665712356567383
    },
    {
      "auxiliary_loss_clip": 0.01236521,
      "auxiliary_loss_mlp": 0.01029241,
      "balance_loss_clip": 1.04951477,
      "balance_loss_mlp": 1.02144432,
      "epoch": 0.4684663019298984,
      "flos": 15368330551680.0,
      "grad_norm": 8.108567294012968,
      "language_loss": 0.65924871,
      "learning_rate": 2.3003217954764672e-06,
      "loss": 0.68190634,
      "num_input_tokens_seen": 83815455,
      "step": 3896,
      "time_per_iteration": 2.6742501258850098
    },
    {
      "auxiliary_loss_clip": 0.0123974,
      "auxiliary_loss_mlp": 0.01023258,
      "balance_loss_clip": 1.04980767,
      "balance_loss_mlp": 1.0161413,
      "epoch": 0.46858654482053747,
      "flos": 27778842737280.0,
      "grad_norm": 1.9334657421026111,
      "language_loss": 0.7889837,
      "learning_rate": 2.299551633382949e-06,
      "loss": 0.81161368,
      "num_input_tokens_seen": 83835765,
      "step": 3897,
      "time_per_iteration": 2.7105257511138916
    },
    {
      "auxiliary_loss_clip": 0.01284432,
      "auxiliary_loss_mlp": 0.01029945,
      "balance_loss_clip": 1.04896653,
      "balance_loss_mlp": 1.02200592,
      "epoch": 0.4687067877111766,
      "flos": 18040623707520.0,
      "grad_norm": 1.8932394859409858,
      "language_loss": 0.85997438,
      "learning_rate": 2.2987814258476854e-06,
      "loss": 0.88311815,
      "num_input_tokens_seen": 83853565,
      "step": 3898,
      "time_per_iteration": 2.651214122772217
    },
    {
      "auxiliary_loss_clip": 0.01392173,
      "auxiliary_loss_mlp": 0.01031916,
      "balance_loss_clip": 1.04541063,
      "balance_loss_mlp": 1.02360129,
      "epoch": 0.4688270306018157,
      "flos": 16977380198400.0,
      "grad_norm": 2.400775946285375,
      "language_loss": 0.67831135,
      "learning_rate": 2.2980111729875177e-06,
      "loss": 0.7025522,
      "num_input_tokens_seen": 83869815,
      "step": 3899,
      "time_per_iteration": 2.7033140659332275
    },
    {
      "auxiliary_loss_clip": 0.0128202,
      "auxiliary_loss_mlp": 0.01026376,
      "balance_loss_clip": 1.05202007,
      "balance_loss_mlp": 1.01851463,
      "epoch": 0.46894727349245474,
      "flos": 17821640442240.0,
      "grad_norm": 1.671405268958826,
      "language_loss": 0.8216784,
      "learning_rate": 2.2972408749192917e-06,
      "loss": 0.84476233,
      "num_input_tokens_seen": 83887545,
      "step": 3900,
      "time_per_iteration": 2.6845109462738037
    },
    {
      "auxiliary_loss_clip": 0.0123695,
      "auxiliary_loss_mlp": 0.02565252,
      "balance_loss_clip": 1.05410206,
      "balance_loss_mlp": 1.00013745,
      "epoch": 0.46906751638309385,
      "flos": 21471349559040.0,
      "grad_norm": 2.295814603021938,
      "language_loss": 0.67158353,
      "learning_rate": 2.296470531759861e-06,
      "loss": 0.70960546,
      "num_input_tokens_seen": 83905645,
      "step": 3901,
      "time_per_iteration": 2.6139473915100098
    },
    {
      "auxiliary_loss_clip": 0.01329459,
      "auxiliary_loss_mlp": 0.01031749,
      "balance_loss_clip": 1.04692769,
      "balance_loss_mlp": 1.02370214,
      "epoch": 0.46918775927373296,
      "flos": 20337829090560.0,
      "grad_norm": 2.4372271378378425,
      "language_loss": 0.7959249,
      "learning_rate": 2.2957001436260866e-06,
      "loss": 0.81953704,
      "num_input_tokens_seen": 83922705,
      "step": 3902,
      "time_per_iteration": 2.7673256397247314
    },
    {
      "auxiliary_loss_clip": 0.01287674,
      "auxiliary_loss_mlp": 0.01028281,
      "balance_loss_clip": 1.05063128,
      "balance_loss_mlp": 1.02066898,
      "epoch": 0.469308002164372,
      "flos": 18403249461120.0,
      "grad_norm": 1.6557719706940468,
      "language_loss": 0.73064095,
      "learning_rate": 2.294929710634836e-06,
      "loss": 0.75380045,
      "num_input_tokens_seen": 83940795,
      "step": 3903,
      "time_per_iteration": 2.6921041011810303
    },
    {
      "auxiliary_loss_clip": 0.01237465,
      "auxiliary_loss_mlp": 0.01026341,
      "balance_loss_clip": 1.04958045,
      "balance_loss_mlp": 1.01854467,
      "epoch": 0.46942824505501113,
      "flos": 37962067363200.0,
      "grad_norm": 1.7840306153228374,
      "language_loss": 0.61354971,
      "learning_rate": 2.2941592329029823e-06,
      "loss": 0.63618779,
      "num_input_tokens_seen": 83961900,
      "step": 3904,
      "time_per_iteration": 2.805628776550293
    },
    {
      "auxiliary_loss_clip": 0.0123443,
      "auxiliary_loss_mlp": 0.01032033,
      "balance_loss_clip": 1.05044317,
      "balance_loss_mlp": 1.02430868,
      "epoch": 0.46954848794565024,
      "flos": 21872507627520.0,
      "grad_norm": 1.8704253847915475,
      "language_loss": 0.78853863,
      "learning_rate": 2.2933887105474067e-06,
      "loss": 0.81120324,
      "num_input_tokens_seen": 83980075,
      "step": 3905,
      "time_per_iteration": 2.6583614349365234
    },
    {
      "auxiliary_loss_clip": 0.01233571,
      "auxiliary_loss_mlp": 0.01025541,
      "balance_loss_clip": 1.05239332,
      "balance_loss_mlp": 1.01844251,
      "epoch": 0.4696687308362893,
      "flos": 22016545165440.0,
      "grad_norm": 1.9940618922678748,
      "language_loss": 0.81432635,
      "learning_rate": 2.2926181436849974e-06,
      "loss": 0.8369174,
      "num_input_tokens_seen": 83999430,
      "step": 3906,
      "time_per_iteration": 2.6614344120025635
    },
    {
      "auxiliary_loss_clip": 0.01234559,
      "auxiliary_loss_mlp": 0.01033211,
      "balance_loss_clip": 1.05116057,
      "balance_loss_mlp": 1.02503872,
      "epoch": 0.4697889737269284,
      "flos": 21613663244160.0,
      "grad_norm": 1.681803990448591,
      "language_loss": 0.72715849,
      "learning_rate": 2.2918475324326478e-06,
      "loss": 0.74983621,
      "num_input_tokens_seen": 84019150,
      "step": 3907,
      "time_per_iteration": 4.0589306354522705
    },
    {
      "auxiliary_loss_clip": 0.01241157,
      "auxiliary_loss_mlp": 0.02573023,
      "balance_loss_clip": 1.05415344,
      "balance_loss_mlp": 1.00013995,
      "epoch": 0.46990921661756746,
      "flos": 25228323665280.0,
      "grad_norm": 7.554441628052622,
      "language_loss": 0.90862489,
      "learning_rate": 2.2910768769072603e-06,
      "loss": 0.94676673,
      "num_input_tokens_seen": 84037930,
      "step": 3908,
      "time_per_iteration": 2.736666679382324
    },
    {
      "auxiliary_loss_clip": 0.01235044,
      "auxiliary_loss_mlp": 0.01027043,
      "balance_loss_clip": 1.05088413,
      "balance_loss_mlp": 1.01953268,
      "epoch": 0.47002945950820657,
      "flos": 13844031045120.0,
      "grad_norm": 1.9065051119870893,
      "language_loss": 0.75851679,
      "learning_rate": 2.2903061772257417e-06,
      "loss": 0.78113765,
      "num_input_tokens_seen": 84055915,
      "step": 3909,
      "time_per_iteration": 3.7427358627319336
    },
    {
      "auxiliary_loss_clip": 0.01233858,
      "auxiliary_loss_mlp": 0.01023904,
      "balance_loss_clip": 1.05204117,
      "balance_loss_mlp": 1.01616764,
      "epoch": 0.4701497023988457,
      "flos": 26247001374720.0,
      "grad_norm": 1.6738935323694455,
      "language_loss": 0.78383553,
      "learning_rate": 2.289535433505007e-06,
      "loss": 0.80641311,
      "num_input_tokens_seen": 84077270,
      "step": 3910,
      "time_per_iteration": 2.9135754108428955
    },
    {
      "auxiliary_loss_clip": 0.0119042,
      "auxiliary_loss_mlp": 0.01027652,
      "balance_loss_clip": 1.04832363,
      "balance_loss_mlp": 1.02013016,
      "epoch": 0.47026994528948474,
      "flos": 25629517647360.0,
      "grad_norm": 2.3431891396469195,
      "language_loss": 0.63479877,
      "learning_rate": 2.2887646458619767e-06,
      "loss": 0.65697944,
      "num_input_tokens_seen": 84098635,
      "step": 3911,
      "time_per_iteration": 2.802206516265869
    },
    {
      "auxiliary_loss_clip": 0.01342178,
      "auxiliary_loss_mlp": 0.01034646,
      "balance_loss_clip": 1.04953861,
      "balance_loss_mlp": 1.0262711,
      "epoch": 0.47039018818012385,
      "flos": 20554406144640.0,
      "grad_norm": 3.4738316532290523,
      "language_loss": 0.76671708,
      "learning_rate": 2.2879938144135797e-06,
      "loss": 0.79048526,
      "num_input_tokens_seen": 84114740,
      "step": 3912,
      "time_per_iteration": 3.668818235397339
    },
    {
      "auxiliary_loss_clip": 0.01332811,
      "auxiliary_loss_mlp": 0.02565897,
      "balance_loss_clip": 1.04567683,
      "balance_loss_mlp": 1.0002048,
      "epoch": 0.47051043107076296,
      "flos": 21577249831680.0,
      "grad_norm": 1.6527090530638666,
      "language_loss": 0.75087452,
      "learning_rate": 2.2872229392767496e-06,
      "loss": 0.78986162,
      "num_input_tokens_seen": 84134845,
      "step": 3913,
      "time_per_iteration": 2.8628573417663574
    },
    {
      "auxiliary_loss_clip": 0.01242193,
      "auxiliary_loss_mlp": 0.01029499,
      "balance_loss_clip": 1.05422044,
      "balance_loss_mlp": 1.02184021,
      "epoch": 0.470630673961402,
      "flos": 18953185662720.0,
      "grad_norm": 1.6113938928246734,
      "language_loss": 0.74825269,
      "learning_rate": 2.286452020568428e-06,
      "loss": 0.77096963,
      "num_input_tokens_seen": 84152920,
      "step": 3914,
      "time_per_iteration": 2.764180898666382
    },
    {
      "auxiliary_loss_clip": 0.01191615,
      "auxiliary_loss_mlp": 0.01032873,
      "balance_loss_clip": 1.0528394,
      "balance_loss_mlp": 1.02503812,
      "epoch": 0.4707509168520411,
      "flos": 19938969492480.0,
      "grad_norm": 1.8337817391145739,
      "language_loss": 0.73128062,
      "learning_rate": 2.2856810584055637e-06,
      "loss": 0.75352556,
      "num_input_tokens_seen": 84170455,
      "step": 3915,
      "time_per_iteration": 2.701190710067749
    },
    {
      "auxiliary_loss_clip": 0.01235778,
      "auxiliary_loss_mlp": 0.01026352,
      "balance_loss_clip": 1.04959679,
      "balance_loss_mlp": 1.01889277,
      "epoch": 0.47087115974268023,
      "flos": 40118754741120.0,
      "grad_norm": 1.5506755319063072,
      "language_loss": 0.67993486,
      "learning_rate": 2.2849100529051085e-06,
      "loss": 0.70255613,
      "num_input_tokens_seen": 84197390,
      "step": 3916,
      "time_per_iteration": 3.8579235076904297
    },
    {
      "auxiliary_loss_clip": 0.01186462,
      "auxiliary_loss_mlp": 0.01025168,
      "balance_loss_clip": 1.05357242,
      "balance_loss_mlp": 1.01772046,
      "epoch": 0.4709914026333193,
      "flos": 13552723745280.0,
      "grad_norm": 2.975055790916585,
      "language_loss": 0.8044709,
      "learning_rate": 2.284139004184026e-06,
      "loss": 0.8265872,
      "num_input_tokens_seen": 84214620,
      "step": 3917,
      "time_per_iteration": 2.660277843475342
    },
    {
      "auxiliary_loss_clip": 0.01187203,
      "auxiliary_loss_mlp": 0.01032052,
      "balance_loss_clip": 1.05367887,
      "balance_loss_mlp": 1.02452993,
      "epoch": 0.4711116455239584,
      "flos": 19974628719360.0,
      "grad_norm": 2.43849666289561,
      "language_loss": 0.74714828,
      "learning_rate": 2.2833679123592814e-06,
      "loss": 0.76934075,
      "num_input_tokens_seen": 84231880,
      "step": 3918,
      "time_per_iteration": 2.673644781112671
    },
    {
      "auxiliary_loss_clip": 0.0128288,
      "auxiliary_loss_mlp": 0.01027055,
      "balance_loss_clip": 1.05097151,
      "balance_loss_mlp": 1.01941931,
      "epoch": 0.4712318884145975,
      "flos": 32124824064000.0,
      "grad_norm": 1.833962628451763,
      "language_loss": 0.63441914,
      "learning_rate": 2.2825967775478508e-06,
      "loss": 0.65751845,
      "num_input_tokens_seen": 84252980,
      "step": 3919,
      "time_per_iteration": 2.816105842590332
    },
    {
      "auxiliary_loss_clip": 0.0118566,
      "auxiliary_loss_mlp": 0.01032923,
      "balance_loss_clip": 1.05080593,
      "balance_loss_mlp": 1.02520466,
      "epoch": 0.47135213130523657,
      "flos": 20047850593920.0,
      "grad_norm": 2.363930401147122,
      "language_loss": 0.83542126,
      "learning_rate": 2.2818255998667135e-06,
      "loss": 0.85760713,
      "num_input_tokens_seen": 84271490,
      "step": 3920,
      "time_per_iteration": 2.725762128829956
    },
    {
      "auxiliary_loss_clip": 0.01234836,
      "auxiliary_loss_mlp": 0.0102426,
      "balance_loss_clip": 1.052495,
      "balance_loss_mlp": 1.01680398,
      "epoch": 0.4714723741958757,
      "flos": 19426990988160.0,
      "grad_norm": 1.7958809792003754,
      "language_loss": 0.79170597,
      "learning_rate": 2.2810543794328566e-06,
      "loss": 0.81429696,
      "num_input_tokens_seen": 84290525,
      "step": 3921,
      "time_per_iteration": 2.8254036903381348
    },
    {
      "auxiliary_loss_clip": 0.01239801,
      "auxiliary_loss_mlp": 0.01025675,
      "balance_loss_clip": 1.05076838,
      "balance_loss_mlp": 1.01815319,
      "epoch": 0.4715926170865148,
      "flos": 20373883367040.0,
      "grad_norm": 1.705236295464475,
      "language_loss": 0.82401478,
      "learning_rate": 2.2802831163632735e-06,
      "loss": 0.8466695,
      "num_input_tokens_seen": 84309245,
      "step": 3922,
      "time_per_iteration": 2.715057373046875
    },
    {
      "auxiliary_loss_clip": 0.01435909,
      "auxiliary_loss_mlp": 0.01025209,
      "balance_loss_clip": 1.0472188,
      "balance_loss_mlp": 1.01721597,
      "epoch": 0.47171285997715384,
      "flos": 22672884430080.0,
      "grad_norm": 1.8173060506324743,
      "language_loss": 0.74362874,
      "learning_rate": 2.279511810774965e-06,
      "loss": 0.76823986,
      "num_input_tokens_seen": 84330775,
      "step": 3923,
      "time_per_iteration": 2.966808319091797
    },
    {
      "auxiliary_loss_clip": 0.01189151,
      "auxiliary_loss_mlp": 0.01031648,
      "balance_loss_clip": 1.05330479,
      "balance_loss_mlp": 1.02382791,
      "epoch": 0.47183310286779295,
      "flos": 21105419754240.0,
      "grad_norm": 1.8949857578185092,
      "language_loss": 0.71778643,
      "learning_rate": 2.2787404627849364e-06,
      "loss": 0.73999441,
      "num_input_tokens_seen": 84349985,
      "step": 3924,
      "time_per_iteration": 2.8080499172210693
    },
    {
      "auxiliary_loss_clip": 0.01286847,
      "auxiliary_loss_mlp": 0.01019657,
      "balance_loss_clip": 1.04826212,
      "balance_loss_mlp": 1.01232874,
      "epoch": 0.471953345758432,
      "flos": 21726566668800.0,
      "grad_norm": 1.8122090142116205,
      "language_loss": 0.79063654,
      "learning_rate": 2.277969072510202e-06,
      "loss": 0.81370163,
      "num_input_tokens_seen": 84368965,
      "step": 3925,
      "time_per_iteration": 2.67380690574646
    },
    {
      "auxiliary_loss_clip": 0.01288044,
      "auxiliary_loss_mlp": 0.01029117,
      "balance_loss_clip": 1.05140662,
      "balance_loss_mlp": 1.02185118,
      "epoch": 0.4720735886490711,
      "flos": 19861078849920.0,
      "grad_norm": 1.613905506677496,
      "language_loss": 0.81380552,
      "learning_rate": 2.2771976400677803e-06,
      "loss": 0.83697712,
      "num_input_tokens_seen": 84387795,
      "step": 3926,
      "time_per_iteration": 2.6948089599609375
    },
    {
      "auxiliary_loss_clip": 0.0137358,
      "auxiliary_loss_mlp": 0.01026088,
      "balance_loss_clip": 1.04292881,
      "balance_loss_mlp": 1.01845813,
      "epoch": 0.47219383153971023,
      "flos": 19171809792000.0,
      "grad_norm": 1.892630437250918,
      "language_loss": 0.78943223,
      "learning_rate": 2.2764261655746965e-06,
      "loss": 0.81342894,
      "num_input_tokens_seen": 84405290,
      "step": 3927,
      "time_per_iteration": 2.818697929382324
    },
    {
      "auxiliary_loss_clip": 0.01334178,
      "auxiliary_loss_mlp": 0.01030857,
      "balance_loss_clip": 1.04755116,
      "balance_loss_mlp": 1.02312589,
      "epoch": 0.4723140744303493,
      "flos": 23224005780480.0,
      "grad_norm": 1.8359433197767585,
      "language_loss": 0.75925875,
      "learning_rate": 2.2756546491479832e-06,
      "loss": 0.78290904,
      "num_input_tokens_seen": 84426205,
      "step": 3928,
      "time_per_iteration": 2.7749733924865723
    },
    {
      "auxiliary_loss_clip": 0.01188793,
      "auxiliary_loss_mlp": 0.02567814,
      "balance_loss_clip": 1.0528332,
      "balance_loss_mlp": 1.00021577,
      "epoch": 0.4724343173209884,
      "flos": 18223265387520.0,
      "grad_norm": 3.6964995985432156,
      "language_loss": 0.79862845,
      "learning_rate": 2.274883090904679e-06,
      "loss": 0.83619452,
      "num_input_tokens_seen": 84443970,
      "step": 3929,
      "time_per_iteration": 2.71064829826355
    },
    {
      "auxiliary_loss_clip": 0.01190962,
      "auxiliary_loss_mlp": 0.01032534,
      "balance_loss_clip": 1.05483794,
      "balance_loss_mlp": 1.0253098,
      "epoch": 0.4725545602116275,
      "flos": 21251037490560.0,
      "grad_norm": 2.39846411248549,
      "language_loss": 0.67616296,
      "learning_rate": 2.2741114909618283e-06,
      "loss": 0.69839787,
      "num_input_tokens_seen": 84459865,
      "step": 3930,
      "time_per_iteration": 2.675497531890869
    },
    {
      "auxiliary_loss_clip": 0.01332995,
      "auxiliary_loss_mlp": 0.01026089,
      "balance_loss_clip": 1.0483408,
      "balance_loss_mlp": 1.01854324,
      "epoch": 0.47267480310226656,
      "flos": 21434002392960.0,
      "grad_norm": 1.6638324099100241,
      "language_loss": 0.72121102,
      "learning_rate": 2.2733398494364828e-06,
      "loss": 0.74480188,
      "num_input_tokens_seen": 84479110,
      "step": 3931,
      "time_per_iteration": 2.791306734085083
    },
    {
      "auxiliary_loss_clip": 0.01287824,
      "auxiliary_loss_mlp": 0.01030485,
      "balance_loss_clip": 1.05499268,
      "balance_loss_mlp": 1.02270615,
      "epoch": 0.47279504599290567,
      "flos": 18770508069120.0,
      "grad_norm": 2.7699453221914396,
      "language_loss": 0.84438396,
      "learning_rate": 2.272568166445699e-06,
      "loss": 0.86756706,
      "num_input_tokens_seen": 84497675,
      "step": 3932,
      "time_per_iteration": 2.6907429695129395
    },
    {
      "auxiliary_loss_clip": 0.01236563,
      "auxiliary_loss_mlp": 0.01024644,
      "balance_loss_clip": 1.05134284,
      "balance_loss_mlp": 1.0172528,
      "epoch": 0.4729152888835448,
      "flos": 21105742976640.0,
      "grad_norm": 2.11056153672006,
      "language_loss": 0.64665806,
      "learning_rate": 2.271796442106541e-06,
      "loss": 0.6692701,
      "num_input_tokens_seen": 84517030,
      "step": 3933,
      "time_per_iteration": 4.231297254562378
    },
    {
      "auxiliary_loss_clip": 0.01236998,
      "auxiliary_loss_mlp": 0.01000255,
      "balance_loss_clip": 1.01772606,
      "balance_loss_mlp": 0.99903917,
      "epoch": 0.47303553177418384,
      "flos": 70201877840640.0,
      "grad_norm": 0.7974589522435358,
      "language_loss": 0.56440485,
      "learning_rate": 2.271024676536079e-06,
      "loss": 0.58677745,
      "num_input_tokens_seen": 84577290,
      "step": 3934,
      "time_per_iteration": 3.2665016651153564
    },
    {
      "auxiliary_loss_clip": 0.01292692,
      "auxiliary_loss_mlp": 0.01033764,
      "balance_loss_clip": 1.05306506,
      "balance_loss_mlp": 1.02556801,
      "epoch": 0.47315577466482295,
      "flos": 22455122227200.0,
      "grad_norm": 2.1421724363186407,
      "language_loss": 0.73627174,
      "learning_rate": 2.2702528698513894e-06,
      "loss": 0.75953627,
      "num_input_tokens_seen": 84598415,
      "step": 3935,
      "time_per_iteration": 3.633065938949585
    },
    {
      "auxiliary_loss_clip": 0.01286365,
      "auxiliary_loss_mlp": 0.01026082,
      "balance_loss_clip": 1.0466001,
      "balance_loss_mlp": 1.01815164,
      "epoch": 0.47327601755546206,
      "flos": 24352857480960.0,
      "grad_norm": 1.896690986200204,
      "language_loss": 0.78522491,
      "learning_rate": 2.269481022169554e-06,
      "loss": 0.80834943,
      "num_input_tokens_seen": 84617010,
      "step": 3936,
      "time_per_iteration": 2.7973756790161133
    },
    {
      "auxiliary_loss_clip": 0.01293496,
      "auxiliary_loss_mlp": 0.01029545,
      "balance_loss_clip": 1.05051684,
      "balance_loss_mlp": 1.02184963,
      "epoch": 0.4733962604461011,
      "flos": 22926772736640.0,
      "grad_norm": 3.0629398692096808,
      "language_loss": 0.80557334,
      "learning_rate": 2.2687091336076614e-06,
      "loss": 0.82880366,
      "num_input_tokens_seen": 84636350,
      "step": 3937,
      "time_per_iteration": 2.8156309127807617
    },
    {
      "auxiliary_loss_clip": 0.01234401,
      "auxiliary_loss_mlp": 0.01027518,
      "balance_loss_clip": 1.05138195,
      "balance_loss_mlp": 1.0199182,
      "epoch": 0.4735165033367402,
      "flos": 18327369980160.0,
      "grad_norm": 2.2958506021619662,
      "language_loss": 0.80162716,
      "learning_rate": 2.267937204282807e-06,
      "loss": 0.82424635,
      "num_input_tokens_seen": 84653490,
      "step": 3938,
      "time_per_iteration": 3.5652012825012207
    },
    {
      "auxiliary_loss_clip": 0.01246335,
      "auxiliary_loss_mlp": 0.01028833,
      "balance_loss_clip": 1.0558207,
      "balance_loss_mlp": 1.0196836,
      "epoch": 0.4736367462273793,
      "flos": 23037018554880.0,
      "grad_norm": 2.4874566271943945,
      "language_loss": 0.79096305,
      "learning_rate": 2.2671652343120926e-06,
      "loss": 0.8137148,
      "num_input_tokens_seen": 84673965,
      "step": 3939,
      "time_per_iteration": 2.766378879547119
    },
    {
      "auxiliary_loss_clip": 0.01188449,
      "auxiliary_loss_mlp": 0.01024794,
      "balance_loss_clip": 1.05473793,
      "balance_loss_mlp": 1.0175643,
      "epoch": 0.4737569891180184,
      "flos": 25374336451200.0,
      "grad_norm": 1.762274535106121,
      "language_loss": 0.80377519,
      "learning_rate": 2.2663932238126236e-06,
      "loss": 0.82590765,
      "num_input_tokens_seen": 84692525,
      "step": 3940,
      "time_per_iteration": 2.706486940383911
    },
    {
      "auxiliary_loss_clip": 0.01233078,
      "auxiliary_loss_mlp": 0.0102588,
      "balance_loss_clip": 1.0484035,
      "balance_loss_mlp": 1.0183574,
      "epoch": 0.4738772320086575,
      "flos": 25849326925440.0,
      "grad_norm": 2.18132645301693,
      "language_loss": 0.80035657,
      "learning_rate": 2.265621172901515e-06,
      "loss": 0.82294619,
      "num_input_tokens_seen": 84715640,
      "step": 3941,
      "time_per_iteration": 2.762964963912964
    },
    {
      "auxiliary_loss_clip": 0.01190749,
      "auxiliary_loss_mlp": 0.01030535,
      "balance_loss_clip": 1.05653024,
      "balance_loss_mlp": 1.02254462,
      "epoch": 0.47399747489929656,
      "flos": 27564420499200.0,
      "grad_norm": 4.0417218395442704,
      "language_loss": 0.71655589,
      "learning_rate": 2.2648490816958854e-06,
      "loss": 0.7387687,
      "num_input_tokens_seen": 84736635,
      "step": 3942,
      "time_per_iteration": 3.7417593002319336
    },
    {
      "auxiliary_loss_clip": 0.01236401,
      "auxiliary_loss_mlp": 0.01027215,
      "balance_loss_clip": 1.04952419,
      "balance_loss_mlp": 1.01953197,
      "epoch": 0.47411771778993567,
      "flos": 24863650836480.0,
      "grad_norm": 2.2557851228150065,
      "language_loss": 0.73376787,
      "learning_rate": 2.264076950312861e-06,
      "loss": 0.75640404,
      "num_input_tokens_seen": 84755445,
      "step": 3943,
      "time_per_iteration": 2.6926236152648926
    },
    {
      "auxiliary_loss_clip": 0.01291465,
      "auxiliary_loss_mlp": 0.01024201,
      "balance_loss_clip": 1.05187368,
      "balance_loss_mlp": 1.01658976,
      "epoch": 0.4742379606805748,
      "flos": 22748009725440.0,
      "grad_norm": 2.249737052717874,
      "language_loss": 0.82289582,
      "learning_rate": 2.2633047788695727e-06,
      "loss": 0.84605253,
      "num_input_tokens_seen": 84775750,
      "step": 3944,
      "time_per_iteration": 2.782825231552124
    },
    {
      "auxiliary_loss_clip": 0.01284332,
      "auxiliary_loss_mlp": 0.01026,
      "balance_loss_clip": 1.05080104,
      "balance_loss_mlp": 1.01895213,
      "epoch": 0.47435820357121383,
      "flos": 19681130689920.0,
      "grad_norm": 1.927246801049443,
      "language_loss": 0.64257133,
      "learning_rate": 2.262532567483159e-06,
      "loss": 0.66567469,
      "num_input_tokens_seen": 84794310,
      "step": 3945,
      "time_per_iteration": 2.668940782546997
    },
    {
      "auxiliary_loss_clip": 0.01190014,
      "auxiliary_loss_mlp": 0.02566846,
      "balance_loss_clip": 1.05527449,
      "balance_loss_mlp": 1.00031841,
      "epoch": 0.47447844646185294,
      "flos": 25228718714880.0,
      "grad_norm": 4.739263912519275,
      "language_loss": 0.80288005,
      "learning_rate": 2.2617603162707635e-06,
      "loss": 0.84044856,
      "num_input_tokens_seen": 84814720,
      "step": 3946,
      "time_per_iteration": 2.736337184906006
    },
    {
      "auxiliary_loss_clip": 0.01187654,
      "auxiliary_loss_mlp": 0.01029158,
      "balance_loss_clip": 1.05384421,
      "balance_loss_mlp": 1.02118635,
      "epoch": 0.47459868935249205,
      "flos": 24570619683840.0,
      "grad_norm": 2.1389869362654697,
      "language_loss": 0.82371402,
      "learning_rate": 2.2609880253495363e-06,
      "loss": 0.84588218,
      "num_input_tokens_seen": 84834355,
      "step": 3947,
      "time_per_iteration": 2.6820895671844482
    },
    {
      "auxiliary_loss_clip": 0.01247716,
      "auxiliary_loss_mlp": 0.01025103,
      "balance_loss_clip": 1.04741144,
      "balance_loss_mlp": 1.01769376,
      "epoch": 0.4747189322431311,
      "flos": 20558500295040.0,
      "grad_norm": 2.239605966113392,
      "language_loss": 0.86783105,
      "learning_rate": 2.260215694836633e-06,
      "loss": 0.89055926,
      "num_input_tokens_seen": 84853530,
      "step": 3948,
      "time_per_iteration": 2.79030704498291
    },
    {
      "auxiliary_loss_clip": 0.01387552,
      "auxiliary_loss_mlp": 0.02565579,
      "balance_loss_clip": 1.04549289,
      "balance_loss_mlp": 1.00024807,
      "epoch": 0.4748391751337702,
      "flos": 25995231970560.0,
      "grad_norm": 2.1805686669858333,
      "language_loss": 0.64921361,
      "learning_rate": 2.2594433248492157e-06,
      "loss": 0.68874496,
      "num_input_tokens_seen": 84872505,
      "step": 3949,
      "time_per_iteration": 2.885796308517456
    },
    {
      "auxiliary_loss_clip": 0.01243497,
      "auxiliary_loss_mlp": 0.01027942,
      "balance_loss_clip": 1.05204296,
      "balance_loss_mlp": 1.01956737,
      "epoch": 0.47495941802440933,
      "flos": 22821052032000.0,
      "grad_norm": 1.6710842657326714,
      "language_loss": 0.80548239,
      "learning_rate": 2.2586709155044527e-06,
      "loss": 0.82819688,
      "num_input_tokens_seen": 84893105,
      "step": 3950,
      "time_per_iteration": 2.791947603225708
    },
    {
      "auxiliary_loss_clip": 0.01187589,
      "auxiliary_loss_mlp": 0.01026322,
      "balance_loss_clip": 1.05318379,
      "balance_loss_mlp": 1.01894593,
      "epoch": 0.4750796609150484,
      "flos": 27891782075520.0,
      "grad_norm": 1.5025861796852586,
      "language_loss": 0.75974464,
      "learning_rate": 2.2578984669195167e-06,
      "loss": 0.78188384,
      "num_input_tokens_seen": 84914070,
      "step": 3951,
      "time_per_iteration": 2.7215468883514404
    },
    {
      "auxiliary_loss_clip": 0.01230782,
      "auxiliary_loss_mlp": 0.01024449,
      "balance_loss_clip": 1.04672122,
      "balance_loss_mlp": 1.01730275,
      "epoch": 0.4751999038056875,
      "flos": 35660085471360.0,
      "grad_norm": 1.8412168329249965,
      "language_loss": 0.67742682,
      "learning_rate": 2.2571259792115887e-06,
      "loss": 0.69997919,
      "num_input_tokens_seen": 84935290,
      "step": 3952,
      "time_per_iteration": 2.750565528869629
    },
    {
      "auxiliary_loss_clip": 0.01232454,
      "auxiliary_loss_mlp": 0.01027539,
      "balance_loss_clip": 1.05063593,
      "balance_loss_mlp": 1.02052045,
      "epoch": 0.4753201466963266,
      "flos": 22090880361600.0,
      "grad_norm": 1.6735227537574664,
      "language_loss": 0.7958945,
      "learning_rate": 2.2563534524978544e-06,
      "loss": 0.81849444,
      "num_input_tokens_seen": 84952760,
      "step": 3953,
      "time_per_iteration": 2.729879856109619
    },
    {
      "auxiliary_loss_clip": 0.01339083,
      "auxiliary_loss_mlp": 0.01026631,
      "balance_loss_clip": 1.05397201,
      "balance_loss_mlp": 1.01920474,
      "epoch": 0.47544038958696566,
      "flos": 30190854965760.0,
      "grad_norm": 1.6204710135236071,
      "language_loss": 0.70550346,
      "learning_rate": 2.2555808868955052e-06,
      "loss": 0.72916067,
      "num_input_tokens_seen": 84974890,
      "step": 3954,
      "time_per_iteration": 2.8586926460266113
    },
    {
      "auxiliary_loss_clip": 0.01381629,
      "auxiliary_loss_mlp": 0.01023773,
      "balance_loss_clip": 1.04702067,
      "balance_loss_mlp": 1.01631606,
      "epoch": 0.47556063247760477,
      "flos": 23472219738240.0,
      "grad_norm": 3.1507245964209707,
      "language_loss": 0.73641163,
      "learning_rate": 2.254808282521738e-06,
      "loss": 0.76046568,
      "num_input_tokens_seen": 84993640,
      "step": 3955,
      "time_per_iteration": 2.849005937576294
    },
    {
      "auxiliary_loss_clip": 0.01341423,
      "auxiliary_loss_mlp": 0.02566301,
      "balance_loss_clip": 1.04908919,
      "balance_loss_mlp": 1.00026488,
      "epoch": 0.4756808753682438,
      "flos": 25155209531520.0,
      "grad_norm": 6.851139644165576,
      "language_loss": 0.81267321,
      "learning_rate": 2.2540356394937573e-06,
      "loss": 0.85175037,
      "num_input_tokens_seen": 85012340,
      "step": 3956,
      "time_per_iteration": 2.798614978790283
    },
    {
      "auxiliary_loss_clip": 0.0133892,
      "auxiliary_loss_mlp": 0.0102838,
      "balance_loss_clip": 1.0477376,
      "balance_loss_mlp": 1.01981449,
      "epoch": 0.47580111825888294,
      "flos": 15669729573120.0,
      "grad_norm": 2.4476015448329025,
      "language_loss": 0.83901536,
      "learning_rate": 2.253262957928772e-06,
      "loss": 0.86268836,
      "num_input_tokens_seen": 85029225,
      "step": 3957,
      "time_per_iteration": 2.7657079696655273
    },
    {
      "auxiliary_loss_clip": 0.01284287,
      "auxiliary_loss_mlp": 0.01033326,
      "balance_loss_clip": 1.04775882,
      "balance_loss_mlp": 1.02575648,
      "epoch": 0.47592136114952205,
      "flos": 17636556637440.0,
      "grad_norm": 1.7039129765766372,
      "language_loss": 0.71943408,
      "learning_rate": 2.2524902379439976e-06,
      "loss": 0.74261022,
      "num_input_tokens_seen": 85047895,
      "step": 3958,
      "time_per_iteration": 2.7573227882385254
    },
    {
      "auxiliary_loss_clip": 0.01360223,
      "auxiliary_loss_mlp": 0.01014767,
      "balance_loss_clip": 1.02938914,
      "balance_loss_mlp": 1.01376581,
      "epoch": 0.4760416040401611,
      "flos": 61417159292160.0,
      "grad_norm": 0.7441109632420955,
      "language_loss": 0.63659275,
      "learning_rate": 2.251717479656655e-06,
      "loss": 0.66034269,
      "num_input_tokens_seen": 85112690,
      "step": 3959,
      "time_per_iteration": 4.5547401905059814
    },
    {
      "auxiliary_loss_clip": 0.01190654,
      "auxiliary_loss_mlp": 0.01031635,
      "balance_loss_clip": 1.0541631,
      "balance_loss_mlp": 1.0245657,
      "epoch": 0.4761618469308002,
      "flos": 18405871153920.0,
      "grad_norm": 2.3418153878350645,
      "language_loss": 0.76518071,
      "learning_rate": 2.2509446831839704e-06,
      "loss": 0.78740358,
      "num_input_tokens_seen": 85132130,
      "step": 3960,
      "time_per_iteration": 2.8758199214935303
    },
    {
      "auxiliary_loss_clip": 0.01285627,
      "auxiliary_loss_mlp": 0.01029408,
      "balance_loss_clip": 1.047768,
      "balance_loss_mlp": 1.02195752,
      "epoch": 0.4762820898214393,
      "flos": 18040911016320.0,
      "grad_norm": 3.0476294303179765,
      "language_loss": 0.82148439,
      "learning_rate": 2.250171848643177e-06,
      "loss": 0.84463471,
      "num_input_tokens_seen": 85149420,
      "step": 3961,
      "time_per_iteration": 4.0129101276397705
    },
    {
      "auxiliary_loss_clip": 0.01278668,
      "auxiliary_loss_mlp": 0.01030721,
      "balance_loss_clip": 1.05009365,
      "balance_loss_mlp": 1.02330291,
      "epoch": 0.4764023327120784,
      "flos": 19318253541120.0,
      "grad_norm": 1.8549003156613535,
      "language_loss": 0.85983694,
      "learning_rate": 2.249398976151513e-06,
      "loss": 0.88293087,
      "num_input_tokens_seen": 85166970,
      "step": 3962,
      "time_per_iteration": 2.7475900650024414
    },
    {
      "auxiliary_loss_clip": 0.01186669,
      "auxiliary_loss_mlp": 0.01026176,
      "balance_loss_clip": 1.05422711,
      "balance_loss_mlp": 1.01897013,
      "epoch": 0.4765225756027175,
      "flos": 22747255539840.0,
      "grad_norm": 2.4345417609650157,
      "language_loss": 0.78772306,
      "learning_rate": 2.248626065826223e-06,
      "loss": 0.80985147,
      "num_input_tokens_seen": 85185175,
      "step": 3963,
      "time_per_iteration": 2.7157766819000244
    },
    {
      "auxiliary_loss_clip": 0.01083894,
      "auxiliary_loss_mlp": 0.01004806,
      "balance_loss_clip": 1.02077532,
      "balance_loss_mlp": 1.00370932,
      "epoch": 0.4766428184933566,
      "flos": 65933392106880.0,
      "grad_norm": 0.7852950522344894,
      "language_loss": 0.62578619,
      "learning_rate": 2.2478531177845564e-06,
      "loss": 0.64667308,
      "num_input_tokens_seen": 85246170,
      "step": 3964,
      "time_per_iteration": 4.031198263168335
    },
    {
      "auxiliary_loss_clip": 0.01290135,
      "auxiliary_loss_mlp": 0.01032674,
      "balance_loss_clip": 1.0540688,
      "balance_loss_mlp": 1.02500272,
      "epoch": 0.47676306138399566,
      "flos": 24136495908480.0,
      "grad_norm": 1.78785069531535,
      "language_loss": 0.85008192,
      "learning_rate": 2.247080132143769e-06,
      "loss": 0.87330997,
      "num_input_tokens_seen": 85268525,
      "step": 3965,
      "time_per_iteration": 2.7172610759735107
    },
    {
      "auxiliary_loss_clip": 0.0133418,
      "auxiliary_loss_mlp": 0.01027052,
      "balance_loss_clip": 1.0442946,
      "balance_loss_mlp": 1.01935077,
      "epoch": 0.47688330427463477,
      "flos": 12604322995200.0,
      "grad_norm": 2.572148498916762,
      "language_loss": 0.68935776,
      "learning_rate": 2.246307109021121e-06,
      "loss": 0.71297008,
      "num_input_tokens_seen": 85285930,
      "step": 3966,
      "time_per_iteration": 2.787047863006592
    },
    {
      "auxiliary_loss_clip": 0.01285031,
      "auxiliary_loss_mlp": 0.01026456,
      "balance_loss_clip": 1.04786777,
      "balance_loss_mlp": 1.01924658,
      "epoch": 0.4770035471652739,
      "flos": 21390585828480.0,
      "grad_norm": 1.627610406016541,
      "language_loss": 0.82048762,
      "learning_rate": 2.2455340485338817e-06,
      "loss": 0.84360254,
      "num_input_tokens_seen": 85303565,
      "step": 3967,
      "time_per_iteration": 2.643749952316284
    },
    {
      "auxiliary_loss_clip": 0.01236945,
      "auxiliary_loss_mlp": 0.01033621,
      "balance_loss_clip": 1.0509541,
      "balance_loss_mlp": 1.02667713,
      "epoch": 0.47712379005591293,
      "flos": 25156251025920.0,
      "grad_norm": 3.269736722692732,
      "language_loss": 0.6760323,
      "learning_rate": 2.244760950799322e-06,
      "loss": 0.69873798,
      "num_input_tokens_seen": 85321835,
      "step": 3968,
      "time_per_iteration": 3.6445305347442627
    },
    {
      "auxiliary_loss_clip": 0.01323409,
      "auxiliary_loss_mlp": 0.01027644,
      "balance_loss_clip": 1.0492034,
      "balance_loss_mlp": 1.02062893,
      "epoch": 0.47724403294655204,
      "flos": 22054323294720.0,
      "grad_norm": 2.1609394316323054,
      "language_loss": 0.72725427,
      "learning_rate": 2.2439878159347203e-06,
      "loss": 0.75076485,
      "num_input_tokens_seen": 85341260,
      "step": 3969,
      "time_per_iteration": 2.958996057510376
    },
    {
      "auxiliary_loss_clip": 0.01081325,
      "auxiliary_loss_mlp": 0.01004782,
      "balance_loss_clip": 1.01822031,
      "balance_loss_mlp": 1.003703,
      "epoch": 0.4773642758371911,
      "flos": 70229387658240.0,
      "grad_norm": 0.7286375436426726,
      "language_loss": 0.55233574,
      "learning_rate": 2.2432146440573616e-06,
      "loss": 0.57319677,
      "num_input_tokens_seen": 85407220,
      "step": 3970,
      "time_per_iteration": 3.334859848022461
    },
    {
      "auxiliary_loss_clip": 0.01188662,
      "auxiliary_loss_mlp": 0.01024769,
      "balance_loss_clip": 1.05181313,
      "balance_loss_mlp": 1.0172739,
      "epoch": 0.4774845187278302,
      "flos": 23548602009600.0,
      "grad_norm": 2.208211564161935,
      "language_loss": 0.66675234,
      "learning_rate": 2.242441435284534e-06,
      "loss": 0.68888664,
      "num_input_tokens_seen": 85426095,
      "step": 3971,
      "time_per_iteration": 2.7501134872436523
    },
    {
      "auxiliary_loss_clip": 0.01233546,
      "auxiliary_loss_mlp": 0.01029786,
      "balance_loss_clip": 1.05142403,
      "balance_loss_mlp": 1.02179933,
      "epoch": 0.4776047616184693,
      "flos": 23075371301760.0,
      "grad_norm": 2.720810644724066,
      "language_loss": 0.85459173,
      "learning_rate": 2.2416681897335337e-06,
      "loss": 0.87722498,
      "num_input_tokens_seen": 85444245,
      "step": 3972,
      "time_per_iteration": 2.687241315841675
    },
    {
      "auxiliary_loss_clip": 0.0137802,
      "auxiliary_loss_mlp": 0.01025991,
      "balance_loss_clip": 1.04877186,
      "balance_loss_mlp": 1.01840901,
      "epoch": 0.4777250045091084,
      "flos": 31898119374720.0,
      "grad_norm": 2.2456607616726956,
      "language_loss": 0.67007506,
      "learning_rate": 2.240894907521661e-06,
      "loss": 0.69411522,
      "num_input_tokens_seen": 85463325,
      "step": 3973,
      "time_per_iteration": 2.899890184402466
    },
    {
      "auxiliary_loss_clip": 0.01283774,
      "auxiliary_loss_mlp": 0.01027415,
      "balance_loss_clip": 1.04765558,
      "balance_loss_mlp": 1.01977956,
      "epoch": 0.4778452473997475,
      "flos": 24278163148800.0,
      "grad_norm": 2.2306348616346727,
      "language_loss": 0.63761413,
      "learning_rate": 2.240121588766223e-06,
      "loss": 0.66072607,
      "num_input_tokens_seen": 85483375,
      "step": 3974,
      "time_per_iteration": 2.763699531555176
    },
    {
      "auxiliary_loss_clip": 0.01276002,
      "auxiliary_loss_mlp": 0.01026766,
      "balance_loss_clip": 1.04719388,
      "balance_loss_mlp": 1.01958346,
      "epoch": 0.4779654902903866,
      "flos": 31575031516800.0,
      "grad_norm": 1.8546795276002148,
      "language_loss": 0.71640593,
      "learning_rate": 2.239348233584531e-06,
      "loss": 0.73943365,
      "num_input_tokens_seen": 85504230,
      "step": 3975,
      "time_per_iteration": 2.7645697593688965
    },
    {
      "auxiliary_loss_clip": 0.01238331,
      "auxiliary_loss_mlp": 0.01024291,
      "balance_loss_clip": 1.05183995,
      "balance_loss_mlp": 1.01682806,
      "epoch": 0.47808573318102565,
      "flos": 19500428344320.0,
      "grad_norm": 1.8359139387174148,
      "language_loss": 0.809044,
      "learning_rate": 2.2385748420939013e-06,
      "loss": 0.83167022,
      "num_input_tokens_seen": 85523425,
      "step": 3976,
      "time_per_iteration": 2.699110507965088
    },
    {
      "auxiliary_loss_clip": 0.01185931,
      "auxiliary_loss_mlp": 0.01029795,
      "balance_loss_clip": 1.05499864,
      "balance_loss_mlp": 1.02292848,
      "epoch": 0.47820597607166476,
      "flos": 22601135013120.0,
      "grad_norm": 1.732870255702771,
      "language_loss": 0.72781718,
      "learning_rate": 2.2378014144116583e-06,
      "loss": 0.74997443,
      "num_input_tokens_seen": 85542235,
      "step": 3977,
      "time_per_iteration": 2.634321451187134
    },
    {
      "auxiliary_loss_clip": 0.01187781,
      "auxiliary_loss_mlp": 0.01030561,
      "balance_loss_clip": 1.05303824,
      "balance_loss_mlp": 1.0232414,
      "epoch": 0.4783262189623039,
      "flos": 23003011353600.0,
      "grad_norm": 2.470771129311594,
      "language_loss": 0.79700643,
      "learning_rate": 2.23702795065513e-06,
      "loss": 0.81918991,
      "num_input_tokens_seen": 85561815,
      "step": 3978,
      "time_per_iteration": 2.701625347137451
    },
    {
      "auxiliary_loss_clip": 0.01133142,
      "auxiliary_loss_mlp": 0.01001718,
      "balance_loss_clip": 1.01624656,
      "balance_loss_mlp": 1.00072312,
      "epoch": 0.47844646185294293,
      "flos": 49772801226240.0,
      "grad_norm": 0.9826473853767607,
      "language_loss": 0.67395473,
      "learning_rate": 2.2362544509416493e-06,
      "loss": 0.69530332,
      "num_input_tokens_seen": 85613930,
      "step": 3979,
      "time_per_iteration": 3.0818591117858887
    },
    {
      "auxiliary_loss_clip": 0.01276178,
      "auxiliary_loss_mlp": 0.01024332,
      "balance_loss_clip": 1.04636824,
      "balance_loss_mlp": 1.01741219,
      "epoch": 0.47856670474358204,
      "flos": 20229558520320.0,
      "grad_norm": 2.4093009860714574,
      "language_loss": 0.82856303,
      "learning_rate": 2.2354809153885572e-06,
      "loss": 0.8515681,
      "num_input_tokens_seen": 85631000,
      "step": 3980,
      "time_per_iteration": 2.736797332763672
    },
    {
      "auxiliary_loss_clip": 0.01234592,
      "auxiliary_loss_mlp": 0.01026413,
      "balance_loss_clip": 1.05011404,
      "balance_loss_mlp": 1.018646,
      "epoch": 0.47868694763422115,
      "flos": 20990936131200.0,
      "grad_norm": 2.719277404021331,
      "language_loss": 0.82800543,
      "learning_rate": 2.234707344113197e-06,
      "loss": 0.8506155,
      "num_input_tokens_seen": 85649095,
      "step": 3981,
      "time_per_iteration": 2.6978394985198975
    },
    {
      "auxiliary_loss_clip": 0.01182608,
      "auxiliary_loss_mlp": 0.01023522,
      "balance_loss_clip": 1.05162776,
      "balance_loss_mlp": 1.01624155,
      "epoch": 0.4788071905248602,
      "flos": 19026551191680.0,
      "grad_norm": 1.6641792501909178,
      "language_loss": 0.77446038,
      "learning_rate": 2.233933737232919e-06,
      "loss": 0.79652166,
      "num_input_tokens_seen": 85666875,
      "step": 3982,
      "time_per_iteration": 2.6217119693756104
    },
    {
      "auxiliary_loss_clip": 0.01379545,
      "auxiliary_loss_mlp": 0.02562996,
      "balance_loss_clip": 1.04569101,
      "balance_loss_mlp": 1.00035119,
      "epoch": 0.4789274334154993,
      "flos": 23002221254400.0,
      "grad_norm": 2.0204986668225327,
      "language_loss": 0.78319514,
      "learning_rate": 2.2331600948650793e-06,
      "loss": 0.82262057,
      "num_input_tokens_seen": 85687020,
      "step": 3983,
      "time_per_iteration": 2.873868465423584
    },
    {
      "auxiliary_loss_clip": 0.01335084,
      "auxiliary_loss_mlp": 0.02571538,
      "balance_loss_clip": 1.0519942,
      "balance_loss_mlp": 1.00033474,
      "epoch": 0.4790476763061384,
      "flos": 23075586783360.0,
      "grad_norm": 1.7715765350156634,
      "language_loss": 0.80095851,
      "learning_rate": 2.2323864171270386e-06,
      "loss": 0.84002477,
      "num_input_tokens_seen": 85708290,
      "step": 3984,
      "time_per_iteration": 2.754855155944824
    },
    {
      "auxiliary_loss_clip": 0.01336303,
      "auxiliary_loss_mlp": 0.01025142,
      "balance_loss_clip": 1.04648423,
      "balance_loss_mlp": 1.01765847,
      "epoch": 0.4791679191967775,
      "flos": 21179288073600.0,
      "grad_norm": 1.930216739826783,
      "language_loss": 0.72416496,
      "learning_rate": 2.231612704136164e-06,
      "loss": 0.74777949,
      "num_input_tokens_seen": 85728660,
      "step": 3985,
      "time_per_iteration": 3.6903622150421143
    },
    {
      "auxiliary_loss_clip": 0.01231419,
      "auxiliary_loss_mlp": 0.01026632,
      "balance_loss_clip": 1.04853737,
      "balance_loss_mlp": 1.01912189,
      "epoch": 0.4792881620874166,
      "flos": 22301495758080.0,
      "grad_norm": 2.491800884423435,
      "language_loss": 0.75007451,
      "learning_rate": 2.2308389560098253e-06,
      "loss": 0.77265501,
      "num_input_tokens_seen": 85745035,
      "step": 3986,
      "time_per_iteration": 2.6690518856048584
    },
    {
      "auxiliary_loss_clip": 0.01248974,
      "auxiliary_loss_mlp": 0.01031561,
      "balance_loss_clip": 1.05420542,
      "balance_loss_mlp": 1.0233357,
      "epoch": 0.47940840497805565,
      "flos": 17420877423360.0,
      "grad_norm": 4.479306698256506,
      "language_loss": 0.77117801,
      "learning_rate": 2.2300651728654008e-06,
      "loss": 0.79398328,
      "num_input_tokens_seen": 85760295,
      "step": 3987,
      "time_per_iteration": 3.5420398712158203
    },
    {
      "auxiliary_loss_clip": 0.0112577,
      "auxiliary_loss_mlp": 0.02505656,
      "balance_loss_clip": 1.01565099,
      "balance_loss_mlp": 1.00013447,
      "epoch": 0.47952864786869476,
      "flos": 65358175708800.0,
      "grad_norm": 0.7220814177731106,
      "language_loss": 0.60164118,
      "learning_rate": 2.229291354820272e-06,
      "loss": 0.63795543,
      "num_input_tokens_seen": 85821305,
      "step": 3988,
      "time_per_iteration": 3.2551169395446777
    },
    {
      "auxiliary_loss_clip": 0.01235458,
      "auxiliary_loss_mlp": 0.0102678,
      "balance_loss_clip": 1.04915416,
      "balance_loss_mlp": 1.01864433,
      "epoch": 0.47964889075933387,
      "flos": 16799802336000.0,
      "grad_norm": 2.3409474135061856,
      "language_loss": 0.75594026,
      "learning_rate": 2.228517501991828e-06,
      "loss": 0.77856261,
      "num_input_tokens_seen": 85840105,
      "step": 3989,
      "time_per_iteration": 2.6436736583709717
    },
    {
      "auxiliary_loss_clip": 0.01178879,
      "auxiliary_loss_mlp": 0.01002444,
      "balance_loss_clip": 1.0155282,
      "balance_loss_mlp": 1.00121617,
      "epoch": 0.4797691336499729,
      "flos": 70079244808320.0,
      "grad_norm": 0.8071989060090927,
      "language_loss": 0.61009216,
      "learning_rate": 2.22774361449746e-06,
      "loss": 0.63190532,
      "num_input_tokens_seen": 85896585,
      "step": 3990,
      "time_per_iteration": 4.235157489776611
    },
    {
      "auxiliary_loss_clip": 0.01438982,
      "auxiliary_loss_mlp": 0.01029005,
      "balance_loss_clip": 1.048491,
      "balance_loss_mlp": 1.02076149,
      "epoch": 0.47988937654061203,
      "flos": 18953329317120.0,
      "grad_norm": 2.642432576211445,
      "language_loss": 0.70379174,
      "learning_rate": 2.2269696924545668e-06,
      "loss": 0.72847164,
      "num_input_tokens_seen": 85914415,
      "step": 3991,
      "time_per_iteration": 2.854766368865967
    },
    {
      "auxiliary_loss_clip": 0.01331106,
      "auxiliary_loss_mlp": 0.01026927,
      "balance_loss_clip": 1.04929352,
      "balance_loss_mlp": 1.02030158,
      "epoch": 0.48000961943125114,
      "flos": 14461981649280.0,
      "grad_norm": 2.3437818584386,
      "language_loss": 0.78236985,
      "learning_rate": 2.2261957359805523e-06,
      "loss": 0.80595016,
      "num_input_tokens_seen": 85931650,
      "step": 3992,
      "time_per_iteration": 2.6875789165496826
    },
    {
      "auxiliary_loss_clip": 0.0118451,
      "auxiliary_loss_mlp": 0.01027323,
      "balance_loss_clip": 1.0512861,
      "balance_loss_mlp": 1.01927066,
      "epoch": 0.4801298623218902,
      "flos": 27051149105280.0,
      "grad_norm": 2.1023363148315455,
      "language_loss": 0.7421025,
      "learning_rate": 2.225421745192823e-06,
      "loss": 0.76422083,
      "num_input_tokens_seen": 85951805,
      "step": 3993,
      "time_per_iteration": 2.704941987991333
    },
    {
      "auxiliary_loss_clip": 0.01233092,
      "auxiliary_loss_mlp": 0.01029955,
      "balance_loss_clip": 1.05104232,
      "balance_loss_mlp": 1.02112722,
      "epoch": 0.4802501052125293,
      "flos": 26355236031360.0,
      "grad_norm": 3.249295590247212,
      "language_loss": 0.78285152,
      "learning_rate": 2.2246477202087955e-06,
      "loss": 0.80548203,
      "num_input_tokens_seen": 85972485,
      "step": 3994,
      "time_per_iteration": 3.5701184272766113
    },
    {
      "auxiliary_loss_clip": 0.01285908,
      "auxiliary_loss_mlp": 0.01024938,
      "balance_loss_clip": 1.04760158,
      "balance_loss_mlp": 1.01790452,
      "epoch": 0.4803703481031684,
      "flos": 20993916960000.0,
      "grad_norm": 1.8445970586096467,
      "language_loss": 0.83014798,
      "learning_rate": 2.223873661145887e-06,
      "loss": 0.85325646,
      "num_input_tokens_seen": 85992540,
      "step": 3995,
      "time_per_iteration": 2.6784660816192627
    },
    {
      "auxiliary_loss_clip": 0.01290519,
      "auxiliary_loss_mlp": 0.02565776,
      "balance_loss_clip": 1.05639935,
      "balance_loss_mlp": 1.00037444,
      "epoch": 0.4804905909938075,
      "flos": 20703722981760.0,
      "grad_norm": 1.6338998398076803,
      "language_loss": 0.7158978,
      "learning_rate": 2.2230995681215226e-06,
      "loss": 0.75446075,
      "num_input_tokens_seen": 86012065,
      "step": 3996,
      "time_per_iteration": 2.781186819076538
    },
    {
      "auxiliary_loss_clip": 0.01330022,
      "auxiliary_loss_mlp": 0.01021669,
      "balance_loss_clip": 1.04665768,
      "balance_loss_mlp": 1.01404512,
      "epoch": 0.4806108338844466,
      "flos": 16654831044480.0,
      "grad_norm": 3.924458019417071,
      "language_loss": 0.77876991,
      "learning_rate": 2.2223254412531305e-06,
      "loss": 0.80228674,
      "num_input_tokens_seen": 86029435,
      "step": 3997,
      "time_per_iteration": 2.7020909786224365
    },
    {
      "auxiliary_loss_clip": 0.01275562,
      "auxiliary_loss_mlp": 0.01023281,
      "balance_loss_clip": 1.04579413,
      "balance_loss_mlp": 1.01683223,
      "epoch": 0.4807310767750857,
      "flos": 20011329440640.0,
      "grad_norm": 2.832967154021592,
      "language_loss": 0.8174113,
      "learning_rate": 2.221551280658146e-06,
      "loss": 0.84039974,
      "num_input_tokens_seen": 86048495,
      "step": 3998,
      "time_per_iteration": 2.7537178993225098
    },
    {
      "auxiliary_loss_clip": 0.0138136,
      "auxiliary_loss_mlp": 0.01039797,
      "balance_loss_clip": 1.04673147,
      "balance_loss_mlp": 1.03219116,
      "epoch": 0.48085131966572475,
      "flos": 23185257984000.0,
      "grad_norm": 5.20948616848376,
      "language_loss": 0.74222827,
      "learning_rate": 2.2207770864540085e-06,
      "loss": 0.7664398,
      "num_input_tokens_seen": 86067470,
      "step": 3999,
      "time_per_iteration": 2.8132967948913574
    },
    {
      "auxiliary_loss_clip": 0.01281633,
      "auxiliary_loss_mlp": 0.01024725,
      "balance_loss_clip": 1.04824853,
      "balance_loss_mlp": 1.01746511,
      "epoch": 0.48097156255636386,
      "flos": 20558643949440.0,
      "grad_norm": 1.8048584761020843,
      "language_loss": 0.73304325,
      "learning_rate": 2.220002858758162e-06,
      "loss": 0.75610685,
      "num_input_tokens_seen": 86085460,
      "step": 4000,
      "time_per_iteration": 2.692286968231201
    },
    {
      "auxiliary_loss_clip": 0.01133084,
      "auxiliary_loss_mlp": 0.00999427,
      "balance_loss_clip": 1.0148586,
      "balance_loss_mlp": 0.99842566,
      "epoch": 0.481091805447003,
      "flos": 70511608817280.0,
      "grad_norm": 0.8806941133959448,
      "language_loss": 0.60813951,
      "learning_rate": 2.2192285976880573e-06,
      "loss": 0.62946463,
      "num_input_tokens_seen": 86149715,
      "step": 4001,
      "time_per_iteration": 3.2378737926483154
    },
    {
      "auxiliary_loss_clip": 0.01334106,
      "auxiliary_loss_mlp": 0.02561533,
      "balance_loss_clip": 1.04567647,
      "balance_loss_mlp": 1.00034976,
      "epoch": 0.48121204833764203,
      "flos": 36428214839040.0,
      "grad_norm": 1.668726532589476,
      "language_loss": 0.81013978,
      "learning_rate": 2.2184543033611485e-06,
      "loss": 0.84909618,
      "num_input_tokens_seen": 86170795,
      "step": 4002,
      "time_per_iteration": 2.8525373935699463
    },
    {
      "auxiliary_loss_clip": 0.01236075,
      "auxiliary_loss_mlp": 0.01025706,
      "balance_loss_clip": 1.049793,
      "balance_loss_mlp": 1.01844954,
      "epoch": 0.48133229122828114,
      "flos": 27490264871040.0,
      "grad_norm": 2.0979533724727446,
      "language_loss": 0.81868732,
      "learning_rate": 2.2176799758948957e-06,
      "loss": 0.84130514,
      "num_input_tokens_seen": 86190955,
      "step": 4003,
      "time_per_iteration": 2.750671625137329
    },
    {
      "auxiliary_loss_clip": 0.0128175,
      "auxiliary_loss_mlp": 0.01027302,
      "balance_loss_clip": 1.04811752,
      "balance_loss_mlp": 1.01981533,
      "epoch": 0.4814525341189202,
      "flos": 43072802179200.0,
      "grad_norm": 2.9896715897573762,
      "language_loss": 0.73624015,
      "learning_rate": 2.2169056154067635e-06,
      "loss": 0.75933069,
      "num_input_tokens_seen": 86214875,
      "step": 4004,
      "time_per_iteration": 2.9134700298309326
    },
    {
      "auxiliary_loss_clip": 0.01234935,
      "auxiliary_loss_mlp": 0.02566943,
      "balance_loss_clip": 1.05186069,
      "balance_loss_mlp": 1.00026333,
      "epoch": 0.4815727770095593,
      "flos": 24236901400320.0,
      "grad_norm": 2.289772599992823,
      "language_loss": 0.82553023,
      "learning_rate": 2.216131222014222e-06,
      "loss": 0.86354899,
      "num_input_tokens_seen": 86232950,
      "step": 4005,
      "time_per_iteration": 2.7134668827056885
    },
    {
      "auxiliary_loss_clip": 0.0133278,
      "auxiliary_loss_mlp": 0.01029657,
      "balance_loss_clip": 1.05023789,
      "balance_loss_mlp": 1.02145505,
      "epoch": 0.4816930199001984,
      "flos": 18113630100480.0,
      "grad_norm": 3.1721657204080995,
      "language_loss": 0.80695355,
      "learning_rate": 2.2153567958347455e-06,
      "loss": 0.83057791,
      "num_input_tokens_seen": 86249160,
      "step": 4006,
      "time_per_iteration": 2.7942960262298584
    },
    {
      "auxiliary_loss_clip": 0.0128484,
      "auxiliary_loss_mlp": 0.01031954,
      "balance_loss_clip": 1.05151486,
      "balance_loss_mlp": 1.02489972,
      "epoch": 0.48181326279083747,
      "flos": 17274720983040.0,
      "grad_norm": 12.734515583041603,
      "language_loss": 0.79268873,
      "learning_rate": 2.214582336985815e-06,
      "loss": 0.81585664,
      "num_input_tokens_seen": 86267060,
      "step": 4007,
      "time_per_iteration": 2.6797971725463867
    },
    {
      "auxiliary_loss_clip": 0.01280774,
      "auxiliary_loss_mlp": 0.01028701,
      "balance_loss_clip": 1.04815674,
      "balance_loss_mlp": 1.02074397,
      "epoch": 0.4819335056814766,
      "flos": 14903252231040.0,
      "grad_norm": 2.5296339281498943,
      "language_loss": 0.66682529,
      "learning_rate": 2.2138078455849142e-06,
      "loss": 0.68992001,
      "num_input_tokens_seen": 86285055,
      "step": 4008,
      "time_per_iteration": 2.626786947250366
    },
    {
      "auxiliary_loss_clip": 0.01140781,
      "auxiliary_loss_mlp": 0.01029705,
      "balance_loss_clip": 1.04978538,
      "balance_loss_mlp": 1.02199197,
      "epoch": 0.4820537485721157,
      "flos": 19244888012160.0,
      "grad_norm": 2.204931484147545,
      "language_loss": 0.78783667,
      "learning_rate": 2.2130333217495334e-06,
      "loss": 0.80954146,
      "num_input_tokens_seen": 86304225,
      "step": 4009,
      "time_per_iteration": 2.7325327396392822
    },
    {
      "auxiliary_loss_clip": 0.01280951,
      "auxiliary_loss_mlp": 0.01028394,
      "balance_loss_clip": 1.04807091,
      "balance_loss_mlp": 1.02033544,
      "epoch": 0.48217399146275475,
      "flos": 16033791870720.0,
      "grad_norm": 2.4978340762360394,
      "language_loss": 0.67973709,
      "learning_rate": 2.2122587655971665e-06,
      "loss": 0.70283055,
      "num_input_tokens_seen": 86319170,
      "step": 4010,
      "time_per_iteration": 2.6405539512634277
    },
    {
      "auxiliary_loss_clip": 0.01287087,
      "auxiliary_loss_mlp": 0.01029657,
      "balance_loss_clip": 1.04897749,
      "balance_loss_mlp": 1.02197409,
      "epoch": 0.48229423435339386,
      "flos": 24134197438080.0,
      "grad_norm": 1.7625858856840173,
      "language_loss": 0.64221549,
      "learning_rate": 2.211484177245314e-06,
      "loss": 0.66538286,
      "num_input_tokens_seen": 86338760,
      "step": 4011,
      "time_per_iteration": 3.7310776710510254
    },
    {
      "auxiliary_loss_clip": 0.01190579,
      "auxiliary_loss_mlp": 0.01025552,
      "balance_loss_clip": 1.05561256,
      "balance_loss_mlp": 1.01775277,
      "epoch": 0.48241447724403297,
      "flos": 23805435231360.0,
      "grad_norm": 2.546835394600579,
      "language_loss": 0.72585553,
      "learning_rate": 2.21070955681148e-06,
      "loss": 0.74801683,
      "num_input_tokens_seen": 86357865,
      "step": 4012,
      "time_per_iteration": 2.6638896465301514
    },
    {
      "auxiliary_loss_clip": 0.0132702,
      "auxiliary_loss_mlp": 0.01030468,
      "balance_loss_clip": 1.04797268,
      "balance_loss_mlp": 1.02283275,
      "epoch": 0.482534720134672,
      "flos": 23110312256640.0,
      "grad_norm": 1.808249246711387,
      "language_loss": 0.78404933,
      "learning_rate": 2.209934904413174e-06,
      "loss": 0.80762422,
      "num_input_tokens_seen": 86379470,
      "step": 4013,
      "time_per_iteration": 3.5591137409210205
    },
    {
      "auxiliary_loss_clip": 0.01432751,
      "auxiliary_loss_mlp": 0.01032718,
      "balance_loss_clip": 1.03826082,
      "balance_loss_mlp": 1.02521992,
      "epoch": 0.48265496302531113,
      "flos": 20923819568640.0,
      "grad_norm": 2.35181230164141,
      "language_loss": 0.71708703,
      "learning_rate": 2.2091602201679095e-06,
      "loss": 0.74174178,
      "num_input_tokens_seen": 86399080,
      "step": 4014,
      "time_per_iteration": 2.859541893005371
    },
    {
      "auxiliary_loss_clip": 0.0134077,
      "auxiliary_loss_mlp": 0.01034544,
      "balance_loss_clip": 1.04958916,
      "balance_loss_mlp": 1.02718258,
      "epoch": 0.48277520591595025,
      "flos": 15231152511360.0,
      "grad_norm": 2.0658825405787575,
      "language_loss": 0.83400148,
      "learning_rate": 2.208385504193206e-06,
      "loss": 0.85775459,
      "num_input_tokens_seen": 86416580,
      "step": 4015,
      "time_per_iteration": 2.6651864051818848
    },
    {
      "auxiliary_loss_clip": 0.01184312,
      "auxiliary_loss_mlp": 0.0102836,
      "balance_loss_clip": 1.05036795,
      "balance_loss_mlp": 1.02045095,
      "epoch": 0.4828954488065893,
      "flos": 17858664385920.0,
      "grad_norm": 2.0870787882864157,
      "language_loss": 0.81161821,
      "learning_rate": 2.2076107566065873e-06,
      "loss": 0.833745,
      "num_input_tokens_seen": 86434365,
      "step": 4016,
      "time_per_iteration": 2.6630756855010986
    },
    {
      "auxiliary_loss_clip": 0.01143179,
      "auxiliary_loss_mlp": 0.01026558,
      "balance_loss_clip": 1.05296922,
      "balance_loss_mlp": 1.01896417,
      "epoch": 0.4830156916972284,
      "flos": 32087405070720.0,
      "grad_norm": 2.254671188865721,
      "language_loss": 0.75408506,
      "learning_rate": 2.2068359775255816e-06,
      "loss": 0.77578247,
      "num_input_tokens_seen": 86452675,
      "step": 4017,
      "time_per_iteration": 3.613736391067505
    },
    {
      "auxiliary_loss_clip": 0.01378043,
      "auxiliary_loss_mlp": 0.01028318,
      "balance_loss_clip": 1.04404855,
      "balance_loss_mlp": 1.02101636,
      "epoch": 0.48313593458786747,
      "flos": 21871717528320.0,
      "grad_norm": 3.017668213162362,
      "language_loss": 0.7915355,
      "learning_rate": 2.206061167067723e-06,
      "loss": 0.81559914,
      "num_input_tokens_seen": 86470785,
      "step": 4018,
      "time_per_iteration": 2.759812116622925
    },
    {
      "auxiliary_loss_clip": 0.01333589,
      "auxiliary_loss_mlp": 0.01027893,
      "balance_loss_clip": 1.04325008,
      "balance_loss_mlp": 1.01995361,
      "epoch": 0.4832561774785066,
      "flos": 22601206840320.0,
      "grad_norm": 2.794478330075195,
      "language_loss": 0.79228622,
      "learning_rate": 2.205286325350549e-06,
      "loss": 0.81590104,
      "num_input_tokens_seen": 86489850,
      "step": 4019,
      "time_per_iteration": 2.7434096336364746
    },
    {
      "auxiliary_loss_clip": 0.01384046,
      "auxiliary_loss_mlp": 0.01028023,
      "balance_loss_clip": 1.04696274,
      "balance_loss_mlp": 1.02078676,
      "epoch": 0.4833764203691457,
      "flos": 13437342282240.0,
      "grad_norm": 2.179825280001189,
      "language_loss": 0.72474223,
      "learning_rate": 2.204511452491603e-06,
      "loss": 0.74886298,
      "num_input_tokens_seen": 86506475,
      "step": 4020,
      "time_per_iteration": 3.6907846927642822
    },
    {
      "auxiliary_loss_clip": 0.01182805,
      "auxiliary_loss_mlp": 0.01024547,
      "balance_loss_clip": 1.05269265,
      "balance_loss_mlp": 1.01737642,
      "epoch": 0.48349666325978474,
      "flos": 44128036955520.0,
      "grad_norm": 1.9534178265892506,
      "language_loss": 0.7496835,
      "learning_rate": 2.2037365486084316e-06,
      "loss": 0.77175701,
      "num_input_tokens_seen": 86529715,
      "step": 4021,
      "time_per_iteration": 2.786121368408203
    },
    {
      "auxiliary_loss_clip": 0.01242848,
      "auxiliary_loss_mlp": 0.01023392,
      "balance_loss_clip": 1.04332495,
      "balance_loss_mlp": 1.01570344,
      "epoch": 0.48361690615042385,
      "flos": 26028377245440.0,
      "grad_norm": 2.316750200538825,
      "language_loss": 0.78068733,
      "learning_rate": 2.2029616138185886e-06,
      "loss": 0.80334967,
      "num_input_tokens_seen": 86548715,
      "step": 4022,
      "time_per_iteration": 2.7821168899536133
    },
    {
      "auxiliary_loss_clip": 0.01328097,
      "auxiliary_loss_mlp": 0.01026439,
      "balance_loss_clip": 1.0503819,
      "balance_loss_mlp": 1.01867247,
      "epoch": 0.48373714904106296,
      "flos": 22273306560000.0,
      "grad_norm": 2.314021514105893,
      "language_loss": 0.82735658,
      "learning_rate": 2.202186648239629e-06,
      "loss": 0.85090196,
      "num_input_tokens_seen": 86568650,
      "step": 4023,
      "time_per_iteration": 2.693026065826416
    },
    {
      "auxiliary_loss_clip": 0.01229845,
      "auxiliary_loss_mlp": 0.0102313,
      "balance_loss_clip": 1.04870367,
      "balance_loss_mlp": 1.01590335,
      "epoch": 0.483857391931702,
      "flos": 28292293699200.0,
      "grad_norm": 3.9824183091694962,
      "language_loss": 0.71751487,
      "learning_rate": 2.201411651989117e-06,
      "loss": 0.74004459,
      "num_input_tokens_seen": 86590630,
      "step": 4024,
      "time_per_iteration": 2.6909263134002686
    },
    {
      "auxiliary_loss_clip": 0.01278185,
      "auxiliary_loss_mlp": 0.02563724,
      "balance_loss_clip": 1.0497973,
      "balance_loss_mlp": 1.00021887,
      "epoch": 0.48397763482234113,
      "flos": 27418048577280.0,
      "grad_norm": 1.847446337958562,
      "language_loss": 0.78327632,
      "learning_rate": 2.2006366251846167e-06,
      "loss": 0.82169539,
      "num_input_tokens_seen": 86611270,
      "step": 4025,
      "time_per_iteration": 2.748288869857788
    },
    {
      "auxiliary_loss_clip": 0.01285236,
      "auxiliary_loss_mlp": 0.01030569,
      "balance_loss_clip": 1.05182195,
      "balance_loss_mlp": 1.02333307,
      "epoch": 0.48409787771298024,
      "flos": 16797252470400.0,
      "grad_norm": 2.1116509048645384,
      "language_loss": 0.75521237,
      "learning_rate": 2.1998615679436997e-06,
      "loss": 0.77837038,
      "num_input_tokens_seen": 86628810,
      "step": 4026,
      "time_per_iteration": 2.6406500339508057
    },
    {
      "auxiliary_loss_clip": 0.01289314,
      "auxiliary_loss_mlp": 0.01029219,
      "balance_loss_clip": 1.04746127,
      "balance_loss_mlp": 1.02141047,
      "epoch": 0.4842181206036193,
      "flos": 25083496028160.0,
      "grad_norm": 2.093244695413779,
      "language_loss": 0.77205276,
      "learning_rate": 2.199086480383942e-06,
      "loss": 0.79523814,
      "num_input_tokens_seen": 86648185,
      "step": 4027,
      "time_per_iteration": 2.6515297889709473
    },
    {
      "auxiliary_loss_clip": 0.01299997,
      "auxiliary_loss_mlp": 0.01025923,
      "balance_loss_clip": 1.05233049,
      "balance_loss_mlp": 1.01729178,
      "epoch": 0.4843383634942584,
      "flos": 30372311496960.0,
      "grad_norm": 8.199483773780651,
      "language_loss": 0.68539035,
      "learning_rate": 2.1983113626229234e-06,
      "loss": 0.70864952,
      "num_input_tokens_seen": 86667435,
      "step": 4028,
      "time_per_iteration": 2.785170555114746
    },
    {
      "auxiliary_loss_clip": 0.01331325,
      "auxiliary_loss_mlp": 0.02564501,
      "balance_loss_clip": 1.04338896,
      "balance_loss_mlp": 1.00017452,
      "epoch": 0.4844586063848975,
      "flos": 20413564917120.0,
      "grad_norm": 1.9139298197525163,
      "language_loss": 0.7877171,
      "learning_rate": 2.1975362147782293e-06,
      "loss": 0.82667542,
      "num_input_tokens_seen": 86686630,
      "step": 4029,
      "time_per_iteration": 2.7041947841644287
    },
    {
      "auxiliary_loss_clip": 0.0120626,
      "auxiliary_loss_mlp": 0.01011912,
      "balance_loss_clip": 1.03012466,
      "balance_loss_mlp": 1.01089275,
      "epoch": 0.48457884927553657,
      "flos": 70303722854400.0,
      "grad_norm": 0.6906493740568422,
      "language_loss": 0.54150295,
      "learning_rate": 2.196761036967448e-06,
      "loss": 0.5636847,
      "num_input_tokens_seen": 86754595,
      "step": 4030,
      "time_per_iteration": 3.3886163234710693
    },
    {
      "auxiliary_loss_clip": 0.01226314,
      "auxiliary_loss_mlp": 0.01024526,
      "balance_loss_clip": 1.04674876,
      "balance_loss_mlp": 1.01746273,
      "epoch": 0.4846990921661757,
      "flos": 19934516206080.0,
      "grad_norm": 2.151617822072996,
      "language_loss": 0.77638751,
      "learning_rate": 2.1959858293081743e-06,
      "loss": 0.79889596,
      "num_input_tokens_seen": 86773730,
      "step": 4031,
      "time_per_iteration": 2.6911773681640625
    },
    {
      "auxiliary_loss_clip": 0.01329995,
      "auxiliary_loss_mlp": 0.01025285,
      "balance_loss_clip": 1.04743338,
      "balance_loss_mlp": 1.01800132,
      "epoch": 0.4848193350568148,
      "flos": 23075945919360.0,
      "grad_norm": 2.038446735143694,
      "language_loss": 0.762555,
      "learning_rate": 2.1952105919180056e-06,
      "loss": 0.78610772,
      "num_input_tokens_seen": 86792985,
      "step": 4032,
      "time_per_iteration": 2.6715898513793945
    },
    {
      "auxiliary_loss_clip": 0.01280335,
      "auxiliary_loss_mlp": 0.01028043,
      "balance_loss_clip": 1.04965508,
      "balance_loss_mlp": 1.02075028,
      "epoch": 0.48493957794745385,
      "flos": 22455481363200.0,
      "grad_norm": 4.519225891287855,
      "language_loss": 0.68804157,
      "learning_rate": 2.1944353249145456e-06,
      "loss": 0.71112537,
      "num_input_tokens_seen": 86812095,
      "step": 4033,
      "time_per_iteration": 2.7250940799713135
    },
    {
      "auxiliary_loss_clip": 0.011865,
      "auxiliary_loss_mlp": 0.01024442,
      "balance_loss_clip": 1.05450082,
      "balance_loss_mlp": 1.01705956,
      "epoch": 0.48505982083809296,
      "flos": 25046112948480.0,
      "grad_norm": 1.857713479995933,
      "language_loss": 0.74371058,
      "learning_rate": 2.193660028415401e-06,
      "loss": 0.76581997,
      "num_input_tokens_seen": 86832875,
      "step": 4034,
      "time_per_iteration": 2.6284420490264893
    },
    {
      "auxiliary_loss_clip": 0.0127758,
      "auxiliary_loss_mlp": 0.01030051,
      "balance_loss_clip": 1.04645395,
      "balance_loss_mlp": 1.02214766,
      "epoch": 0.485180063728732,
      "flos": 26761386090240.0,
      "grad_norm": 1.7166658704649305,
      "language_loss": 0.81545728,
      "learning_rate": 2.1928847025381852e-06,
      "loss": 0.83853358,
      "num_input_tokens_seen": 86853480,
      "step": 4035,
      "time_per_iteration": 2.7271432876586914
    },
    {
      "auxiliary_loss_clip": 0.01232982,
      "auxiliary_loss_mlp": 0.01027563,
      "balance_loss_clip": 1.04640174,
      "balance_loss_mlp": 1.01926029,
      "epoch": 0.4853003066193711,
      "flos": 24059143969920.0,
      "grad_norm": 1.6816150033570587,
      "language_loss": 0.83917463,
      "learning_rate": 2.192109347400512e-06,
      "loss": 0.86177999,
      "num_input_tokens_seen": 86873695,
      "step": 4036,
      "time_per_iteration": 2.70835542678833
    },
    {
      "auxiliary_loss_clip": 0.01285842,
      "auxiliary_loss_mlp": 0.01029779,
      "balance_loss_clip": 1.04885614,
      "balance_loss_mlp": 1.02204776,
      "epoch": 0.48542054951001024,
      "flos": 23076376882560.0,
      "grad_norm": 1.8787355625076705,
      "language_loss": 0.79001677,
      "learning_rate": 2.191333963120004e-06,
      "loss": 0.81317306,
      "num_input_tokens_seen": 86892675,
      "step": 4037,
      "time_per_iteration": 3.6955370903015137
    },
    {
      "auxiliary_loss_clip": 0.01280155,
      "auxiliary_loss_mlp": 0.01020454,
      "balance_loss_clip": 1.04826057,
      "balance_loss_mlp": 1.01301503,
      "epoch": 0.4855407924006493,
      "flos": 25664889565440.0,
      "grad_norm": 2.577322850743525,
      "language_loss": 0.70110315,
      "learning_rate": 2.190558549814286e-06,
      "loss": 0.72410923,
      "num_input_tokens_seen": 86912835,
      "step": 4038,
      "time_per_iteration": 2.740154981613159
    },
    {
      "auxiliary_loss_clip": 0.01284973,
      "auxiliary_loss_mlp": 0.01029972,
      "balance_loss_clip": 1.04680979,
      "balance_loss_mlp": 1.02272403,
      "epoch": 0.4856610352912884,
      "flos": 23987933256960.0,
      "grad_norm": 1.8724534510300423,
      "language_loss": 0.79464209,
      "learning_rate": 2.1897831076009872e-06,
      "loss": 0.81779152,
      "num_input_tokens_seen": 86932475,
      "step": 4039,
      "time_per_iteration": 3.563478708267212
    },
    {
      "auxiliary_loss_clip": 0.01233513,
      "auxiliary_loss_mlp": 0.01029334,
      "balance_loss_clip": 1.04965317,
      "balance_loss_mlp": 1.02213359,
      "epoch": 0.4857812781819275,
      "flos": 24096814358400.0,
      "grad_norm": 5.536514802884875,
      "language_loss": 0.79977751,
      "learning_rate": 2.1890076365977426e-06,
      "loss": 0.82240593,
      "num_input_tokens_seen": 86952300,
      "step": 4040,
      "time_per_iteration": 2.6622445583343506
    },
    {
      "auxiliary_loss_clip": 0.01180111,
      "auxiliary_loss_mlp": 0.01002073,
      "balance_loss_clip": 1.01513648,
      "balance_loss_mlp": 1.00098825,
      "epoch": 0.48590152107256657,
      "flos": 56266635185280.0,
      "grad_norm": 0.8513363102495279,
      "language_loss": 0.52733511,
      "learning_rate": 2.188232136922189e-06,
      "loss": 0.54915696,
      "num_input_tokens_seen": 87010420,
      "step": 4041,
      "time_per_iteration": 3.1942598819732666
    },
    {
      "auxiliary_loss_clip": 0.01330134,
      "auxiliary_loss_mlp": 0.0102513,
      "balance_loss_clip": 1.04313731,
      "balance_loss_mlp": 1.01740265,
      "epoch": 0.4860217639632057,
      "flos": 20046988667520.0,
      "grad_norm": 2.934324086980654,
      "language_loss": 0.76255238,
      "learning_rate": 2.187456608691971e-06,
      "loss": 0.78610504,
      "num_input_tokens_seen": 87029295,
      "step": 4042,
      "time_per_iteration": 2.764496088027954
    },
    {
      "auxiliary_loss_clip": 0.01340569,
      "auxiliary_loss_mlp": 0.01027429,
      "balance_loss_clip": 1.05077398,
      "balance_loss_mlp": 1.01974607,
      "epoch": 0.4861420068538448,
      "flos": 17822143232640.0,
      "grad_norm": 2.3989080757998797,
      "language_loss": 0.87708789,
      "learning_rate": 2.1866810520247334e-06,
      "loss": 0.90076786,
      "num_input_tokens_seen": 87048165,
      "step": 4043,
      "time_per_iteration": 3.6264841556549072
    },
    {
      "auxiliary_loss_clip": 0.01237753,
      "auxiliary_loss_mlp": 0.01027381,
      "balance_loss_clip": 1.04941964,
      "balance_loss_mlp": 1.01903009,
      "epoch": 0.48626224974448384,
      "flos": 26250125857920.0,
      "grad_norm": 2.2298147493003944,
      "language_loss": 0.65172511,
      "learning_rate": 2.185905467038129e-06,
      "loss": 0.67437643,
      "num_input_tokens_seen": 87067070,
      "step": 4044,
      "time_per_iteration": 2.6729867458343506
    },
    {
      "auxiliary_loss_clip": 0.01185837,
      "auxiliary_loss_mlp": 0.01032922,
      "balance_loss_clip": 1.05520606,
      "balance_loss_mlp": 1.02547777,
      "epoch": 0.48638249263512295,
      "flos": 22054502862720.0,
      "grad_norm": 1.7556145550608777,
      "language_loss": 0.77540231,
      "learning_rate": 2.1851298538498127e-06,
      "loss": 0.7975899,
      "num_input_tokens_seen": 87086785,
      "step": 4045,
      "time_per_iteration": 3.5811498165130615
    },
    {
      "auxiliary_loss_clip": 0.01244861,
      "auxiliary_loss_mlp": 0.02571706,
      "balance_loss_clip": 1.05551708,
      "balance_loss_mlp": 1.00021744,
      "epoch": 0.48650273552576206,
      "flos": 25119945354240.0,
      "grad_norm": 2.203234559846567,
      "language_loss": 0.79781342,
      "learning_rate": 2.184354212577446e-06,
      "loss": 0.8359791,
      "num_input_tokens_seen": 87107090,
      "step": 4046,
      "time_per_iteration": 2.722137928009033
    },
    {
      "auxiliary_loss_clip": 0.01188955,
      "auxiliary_loss_mlp": 0.01034831,
      "balance_loss_clip": 1.05210924,
      "balance_loss_mlp": 1.02721334,
      "epoch": 0.4866229784164011,
      "flos": 17456931699840.0,
      "grad_norm": 2.9383662867719096,
      "language_loss": 0.6314677,
      "learning_rate": 2.1835785433386907e-06,
      "loss": 0.6537056,
      "num_input_tokens_seen": 87125905,
      "step": 4047,
      "time_per_iteration": 2.5418028831481934
    },
    {
      "auxiliary_loss_clip": 0.01330853,
      "auxiliary_loss_mlp": 0.01027253,
      "balance_loss_clip": 1.0499692,
      "balance_loss_mlp": 1.0195998,
      "epoch": 0.48674322130704023,
      "flos": 23331127115520.0,
      "grad_norm": 1.8948497124081514,
      "language_loss": 0.6558606,
      "learning_rate": 2.182802846251216e-06,
      "loss": 0.67944163,
      "num_input_tokens_seen": 87146175,
      "step": 4048,
      "time_per_iteration": 2.7988412380218506
    },
    {
      "auxiliary_loss_clip": 0.01334119,
      "auxiliary_loss_mlp": 0.01027105,
      "balance_loss_clip": 1.04529989,
      "balance_loss_mlp": 1.01988709,
      "epoch": 0.4868634641976793,
      "flos": 28804344030720.0,
      "grad_norm": 2.151002499061319,
      "language_loss": 0.72478008,
      "learning_rate": 2.182027121432696e-06,
      "loss": 0.74839234,
      "num_input_tokens_seen": 87166800,
      "step": 4049,
      "time_per_iteration": 2.7443809509277344
    },
    {
      "auxiliary_loss_clip": 0.01190696,
      "auxiliary_loss_mlp": 0.01032898,
      "balance_loss_clip": 1.05434,
      "balance_loss_mlp": 1.02521491,
      "epoch": 0.4869837070883184,
      "flos": 19025976574080.0,
      "grad_norm": 1.7493796448457717,
      "language_loss": 0.82173431,
      "learning_rate": 2.1812513690008054e-06,
      "loss": 0.84397018,
      "num_input_tokens_seen": 87185920,
      "step": 4050,
      "time_per_iteration": 2.6153476238250732
    },
    {
      "auxiliary_loss_clip": 0.01243943,
      "auxiliary_loss_mlp": 0.01029785,
      "balance_loss_clip": 1.05330718,
      "balance_loss_mlp": 1.02187598,
      "epoch": 0.4871039499789575,
      "flos": 15121409483520.0,
      "grad_norm": 2.1583693310231253,
      "language_loss": 0.80383146,
      "learning_rate": 2.180475589073227e-06,
      "loss": 0.82656872,
      "num_input_tokens_seen": 87203620,
      "step": 4051,
      "time_per_iteration": 2.6712169647216797
    },
    {
      "auxiliary_loss_clip": 0.01228235,
      "auxiliary_loss_mlp": 0.01023953,
      "balance_loss_clip": 1.04724002,
      "balance_loss_mlp": 1.0165832,
      "epoch": 0.48722419286959656,
      "flos": 26174066808960.0,
      "grad_norm": 1.8291550169859159,
      "language_loss": 0.73254263,
      "learning_rate": 2.1796997817676456e-06,
      "loss": 0.75506455,
      "num_input_tokens_seen": 87224630,
      "step": 4052,
      "time_per_iteration": 2.6843678951263428
    },
    {
      "auxiliary_loss_clip": 0.01239788,
      "auxiliary_loss_mlp": 0.02561772,
      "balance_loss_clip": 1.0521332,
      "balance_loss_mlp": 1.00015545,
      "epoch": 0.4873444357602357,
      "flos": 24026142349440.0,
      "grad_norm": 2.143198221918596,
      "language_loss": 0.67939043,
      "learning_rate": 2.1789239472017494e-06,
      "loss": 0.71740603,
      "num_input_tokens_seen": 87246280,
      "step": 4053,
      "time_per_iteration": 2.7078354358673096
    },
    {
      "auxiliary_loss_clip": 0.01332122,
      "auxiliary_loss_mlp": 0.01026527,
      "balance_loss_clip": 1.04630566,
      "balance_loss_mlp": 1.01886177,
      "epoch": 0.4874646786508748,
      "flos": 22820441500800.0,
      "grad_norm": 2.27829822490563,
      "language_loss": 0.73057711,
      "learning_rate": 2.1781480854932326e-06,
      "loss": 0.75416362,
      "num_input_tokens_seen": 87266045,
      "step": 4054,
      "time_per_iteration": 2.7918620109558105
    },
    {
      "auxiliary_loss_clip": 0.01381631,
      "auxiliary_loss_mlp": 0.01028623,
      "balance_loss_clip": 1.05006933,
      "balance_loss_mlp": 1.02126181,
      "epoch": 0.48758492154151384,
      "flos": 21287594557440.0,
      "grad_norm": 2.1570043409678323,
      "language_loss": 0.78997982,
      "learning_rate": 2.1773721967597933e-06,
      "loss": 0.81408232,
      "num_input_tokens_seen": 87284495,
      "step": 4055,
      "time_per_iteration": 2.7391819953918457
    },
    {
      "auxiliary_loss_clip": 0.01171331,
      "auxiliary_loss_mlp": 0.0099951,
      "balance_loss_clip": 1.01212263,
      "balance_loss_mlp": 0.99848461,
      "epoch": 0.48770516443215295,
      "flos": 62244109180800.0,
      "grad_norm": 0.853015285159282,
      "language_loss": 0.57344317,
      "learning_rate": 2.1765962811191322e-06,
      "loss": 0.5951516,
      "num_input_tokens_seen": 87338960,
      "step": 4056,
      "time_per_iteration": 3.217440366744995
    },
    {
      "auxiliary_loss_clip": 0.0128205,
      "auxiliary_loss_mlp": 0.01002132,
      "balance_loss_clip": 1.01282871,
      "balance_loss_mlp": 1.00108325,
      "epoch": 0.48782540732279206,
      "flos": 66133451882880.0,
      "grad_norm": 0.8232219642285158,
      "language_loss": 0.61960328,
      "learning_rate": 2.1758203386889566e-06,
      "loss": 0.64244509,
      "num_input_tokens_seen": 87401730,
      "step": 4057,
      "time_per_iteration": 3.3421006202697754
    },
    {
      "auxiliary_loss_clip": 0.01337817,
      "auxiliary_loss_mlp": 0.02568989,
      "balance_loss_clip": 1.04965067,
      "balance_loss_mlp": 1.00015569,
      "epoch": 0.4879456502134311,
      "flos": 14607922608000.0,
      "grad_norm": 2.2858043154276695,
      "language_loss": 0.84692621,
      "learning_rate": 2.1750443695869746e-06,
      "loss": 0.88599432,
      "num_input_tokens_seen": 87417300,
      "step": 4058,
      "time_per_iteration": 2.7047276496887207
    },
    {
      "auxiliary_loss_clip": 0.01237779,
      "auxiliary_loss_mlp": 0.01031002,
      "balance_loss_clip": 1.05080652,
      "balance_loss_mlp": 1.02290797,
      "epoch": 0.4880658931040702,
      "flos": 19500464257920.0,
      "grad_norm": 1.9678700378595788,
      "language_loss": 0.85934412,
      "learning_rate": 2.174268373930901e-06,
      "loss": 0.88203198,
      "num_input_tokens_seen": 87434815,
      "step": 4059,
      "time_per_iteration": 2.699521780014038
    },
    {
      "auxiliary_loss_clip": 0.01328945,
      "auxiliary_loss_mlp": 0.02566388,
      "balance_loss_clip": 1.05094659,
      "balance_loss_mlp": 1.00017095,
      "epoch": 0.48818613599470934,
      "flos": 16723060928640.0,
      "grad_norm": 2.1011662137601284,
      "language_loss": 0.80159765,
      "learning_rate": 2.1734923518384537e-06,
      "loss": 0.84055102,
      "num_input_tokens_seen": 87451420,
      "step": 4060,
      "time_per_iteration": 2.6909468173980713
    },
    {
      "auxiliary_loss_clip": 0.0132489,
      "auxiliary_loss_mlp": 0.01029363,
      "balance_loss_clip": 1.05001557,
      "balance_loss_mlp": 1.02237201,
      "epoch": 0.4883063788853484,
      "flos": 26756932803840.0,
      "grad_norm": 2.0312641605487167,
      "language_loss": 0.82507312,
      "learning_rate": 2.1727163034273547e-06,
      "loss": 0.84861565,
      "num_input_tokens_seen": 87469585,
      "step": 4061,
      "time_per_iteration": 2.792018175125122
    },
    {
      "auxiliary_loss_clip": 0.01238178,
      "auxiliary_loss_mlp": 0.01027197,
      "balance_loss_clip": 1.05030322,
      "balance_loss_mlp": 1.01980889,
      "epoch": 0.4884266217759875,
      "flos": 16763388923520.0,
      "grad_norm": 2.1124358990082253,
      "language_loss": 0.78641951,
      "learning_rate": 2.17194022881533e-06,
      "loss": 0.80907321,
      "num_input_tokens_seen": 87485675,
      "step": 4062,
      "time_per_iteration": 2.7032992839813232
    },
    {
      "auxiliary_loss_clip": 0.01287614,
      "auxiliary_loss_mlp": 0.01031086,
      "balance_loss_clip": 1.05010986,
      "balance_loss_mlp": 1.02343893,
      "epoch": 0.4885468646666266,
      "flos": 24207132003840.0,
      "grad_norm": 1.9797445643938614,
      "language_loss": 0.67703128,
      "learning_rate": 2.1711641281201092e-06,
      "loss": 0.70021832,
      "num_input_tokens_seen": 87505605,
      "step": 4063,
      "time_per_iteration": 3.846893548965454
    },
    {
      "auxiliary_loss_clip": 0.01235949,
      "auxiliary_loss_mlp": 0.01024862,
      "balance_loss_clip": 1.05315888,
      "balance_loss_mlp": 1.01765037,
      "epoch": 0.48866710755726567,
      "flos": 14610795696000.0,
      "grad_norm": 2.041597914281744,
      "language_loss": 0.79196149,
      "learning_rate": 2.1703880014594264e-06,
      "loss": 0.81456959,
      "num_input_tokens_seen": 87523195,
      "step": 4064,
      "time_per_iteration": 2.7315523624420166
    },
    {
      "auxiliary_loss_clip": 0.01385519,
      "auxiliary_loss_mlp": 0.0102382,
      "balance_loss_clip": 1.0530839,
      "balance_loss_mlp": 1.01667666,
      "epoch": 0.4887873504479048,
      "flos": 28804451771520.0,
      "grad_norm": 2.199166503907879,
      "language_loss": 0.73894882,
      "learning_rate": 2.1696118489510182e-06,
      "loss": 0.76304221,
      "num_input_tokens_seen": 87544125,
      "step": 4065,
      "time_per_iteration": 3.713191270828247
    },
    {
      "auxiliary_loss_clip": 0.01339338,
      "auxiliary_loss_mlp": 0.025665,
      "balance_loss_clip": 1.04899788,
      "balance_loss_mlp": 1.00017607,
      "epoch": 0.48890759333854383,
      "flos": 22784387224320.0,
      "grad_norm": 2.569873835685232,
      "language_loss": 0.72806352,
      "learning_rate": 2.1688356707126286e-06,
      "loss": 0.76712191,
      "num_input_tokens_seen": 87563745,
      "step": 4066,
      "time_per_iteration": 2.7563745975494385
    },
    {
      "auxiliary_loss_clip": 0.01329647,
      "auxiliary_loss_mlp": 0.01032476,
      "balance_loss_clip": 1.04827261,
      "balance_loss_mlp": 1.02399468,
      "epoch": 0.48902783622918294,
      "flos": 17786088956160.0,
      "grad_norm": 2.2853673090312157,
      "language_loss": 0.70046687,
      "learning_rate": 2.168059466862001e-06,
      "loss": 0.72408807,
      "num_input_tokens_seen": 87581895,
      "step": 4067,
      "time_per_iteration": 2.708004951477051
    },
    {
      "auxiliary_loss_clip": 0.01282561,
      "auxiliary_loss_mlp": 0.01026205,
      "balance_loss_clip": 1.04610324,
      "balance_loss_mlp": 1.0193119,
      "epoch": 0.48914807911982205,
      "flos": 22310294590080.0,
      "grad_norm": 1.8764557716521835,
      "language_loss": 0.81804717,
      "learning_rate": 2.167283237516887e-06,
      "loss": 0.84113479,
      "num_input_tokens_seen": 87600170,
      "step": 4068,
      "time_per_iteration": 2.691319465637207
    },
    {
      "auxiliary_loss_clip": 0.01287546,
      "auxiliary_loss_mlp": 0.01026409,
      "balance_loss_clip": 1.04881167,
      "balance_loss_mlp": 1.01873171,
      "epoch": 0.4892683220104611,
      "flos": 16363020954240.0,
      "grad_norm": 1.924332755272138,
      "language_loss": 0.747159,
      "learning_rate": 2.1665069827950383e-06,
      "loss": 0.77029854,
      "num_input_tokens_seen": 87617455,
      "step": 4069,
      "time_per_iteration": 3.533473014831543
    },
    {
      "auxiliary_loss_clip": 0.01282348,
      "auxiliary_loss_mlp": 0.01021654,
      "balance_loss_clip": 1.04929638,
      "balance_loss_mlp": 1.01459718,
      "epoch": 0.4893885649011002,
      "flos": 15739144606080.0,
      "grad_norm": 2.1778333826309026,
      "language_loss": 0.86840725,
      "learning_rate": 2.1657307028142126e-06,
      "loss": 0.89144731,
      "num_input_tokens_seen": 87634995,
      "step": 4070,
      "time_per_iteration": 2.6145646572113037
    },
    {
      "auxiliary_loss_clip": 0.01284168,
      "auxiliary_loss_mlp": 0.01027815,
      "balance_loss_clip": 1.04968762,
      "balance_loss_mlp": 1.01937461,
      "epoch": 0.48950880779173933,
      "flos": 28581984887040.0,
      "grad_norm": 2.588690343760685,
      "language_loss": 0.67528522,
      "learning_rate": 2.164954397692171e-06,
      "loss": 0.69840503,
      "num_input_tokens_seen": 87654420,
      "step": 4071,
      "time_per_iteration": 3.6981096267700195
    },
    {
      "auxiliary_loss_clip": 0.01185293,
      "auxiliary_loss_mlp": 0.00999859,
      "balance_loss_clip": 1.01450849,
      "balance_loss_mlp": 0.9989174,
      "epoch": 0.4896290506823784,
      "flos": 66186310746240.0,
      "grad_norm": 1.08742452803563,
      "language_loss": 0.77302003,
      "learning_rate": 2.164178067546678e-06,
      "loss": 0.79487157,
      "num_input_tokens_seen": 87713585,
      "step": 4072,
      "time_per_iteration": 3.363568067550659
    },
    {
      "auxiliary_loss_clip": 0.0129047,
      "auxiliary_loss_mlp": 0.01026379,
      "balance_loss_clip": 1.04785371,
      "balance_loss_mlp": 1.01892281,
      "epoch": 0.4897492935730175,
      "flos": 12531065207040.0,
      "grad_norm": 2.625214398679279,
      "language_loss": 0.91025341,
      "learning_rate": 2.163401712495504e-06,
      "loss": 0.93342197,
      "num_input_tokens_seen": 87731280,
      "step": 4073,
      "time_per_iteration": 2.695124626159668
    },
    {
      "auxiliary_loss_clip": 0.0129785,
      "auxiliary_loss_mlp": 0.01034959,
      "balance_loss_clip": 1.05186009,
      "balance_loss_mlp": 1.02712131,
      "epoch": 0.4898695364636566,
      "flos": 23476816679040.0,
      "grad_norm": 1.6759869821067048,
      "language_loss": 0.79355657,
      "learning_rate": 2.1626253326564194e-06,
      "loss": 0.8168847,
      "num_input_tokens_seen": 87750230,
      "step": 4074,
      "time_per_iteration": 2.844325304031372
    },
    {
      "auxiliary_loss_clip": 0.01287172,
      "auxiliary_loss_mlp": 0.01029783,
      "balance_loss_clip": 1.05012655,
      "balance_loss_mlp": 1.02142692,
      "epoch": 0.48998977935429566,
      "flos": 27160209774720.0,
      "grad_norm": 1.8195501910112515,
      "language_loss": 0.76881802,
      "learning_rate": 2.161848928147201e-06,
      "loss": 0.7919876,
      "num_input_tokens_seen": 87770500,
      "step": 4075,
      "time_per_iteration": 2.8167190551757812
    },
    {
      "auxiliary_loss_clip": 0.01235346,
      "auxiliary_loss_mlp": 0.01031504,
      "balance_loss_clip": 1.05185914,
      "balance_loss_mlp": 1.02355301,
      "epoch": 0.4901100222449348,
      "flos": 20339588856960.0,
      "grad_norm": 2.830602692894066,
      "language_loss": 0.81023985,
      "learning_rate": 2.161072499085629e-06,
      "loss": 0.83290833,
      "num_input_tokens_seen": 87789495,
      "step": 4076,
      "time_per_iteration": 2.697890520095825
    },
    {
      "auxiliary_loss_clip": 0.0133568,
      "auxiliary_loss_mlp": 0.0103276,
      "balance_loss_clip": 1.04733777,
      "balance_loss_mlp": 1.02527928,
      "epoch": 0.4902302651355739,
      "flos": 30446359384320.0,
      "grad_norm": 1.719481161503393,
      "language_loss": 0.83239222,
      "learning_rate": 2.160296045589487e-06,
      "loss": 0.8560766,
      "num_input_tokens_seen": 87812955,
      "step": 4077,
      "time_per_iteration": 2.8560867309570312
    },
    {
      "auxiliary_loss_clip": 0.01236304,
      "auxiliary_loss_mlp": 0.01029314,
      "balance_loss_clip": 1.05264294,
      "balance_loss_mlp": 1.02179778,
      "epoch": 0.49035050802621294,
      "flos": 19174180089600.0,
      "grad_norm": 6.099918873185634,
      "language_loss": 0.70260644,
      "learning_rate": 2.159519567776562e-06,
      "loss": 0.72526264,
      "num_input_tokens_seen": 87832605,
      "step": 4078,
      "time_per_iteration": 2.7209858894348145
    },
    {
      "auxiliary_loss_clip": 0.01294442,
      "auxiliary_loss_mlp": 0.01026982,
      "balance_loss_clip": 1.04439187,
      "balance_loss_mlp": 1.01912034,
      "epoch": 0.49047075091685205,
      "flos": 22228489365120.0,
      "grad_norm": 3.0994053941026714,
      "language_loss": 0.70817995,
      "learning_rate": 2.1587430657646463e-06,
      "loss": 0.73139417,
      "num_input_tokens_seen": 87846040,
      "step": 4079,
      "time_per_iteration": 2.8434557914733887
    },
    {
      "auxiliary_loss_clip": 0.01282201,
      "auxiliary_loss_mlp": 0.01032007,
      "balance_loss_clip": 1.05185914,
      "balance_loss_mlp": 1.02458668,
      "epoch": 0.4905909938074911,
      "flos": 20156516213760.0,
      "grad_norm": 2.274945405022825,
      "language_loss": 0.78118086,
      "learning_rate": 2.157966539671533e-06,
      "loss": 0.80432296,
      "num_input_tokens_seen": 87865680,
      "step": 4080,
      "time_per_iteration": 2.7114460468292236
    },
    {
      "auxiliary_loss_clip": 0.01334718,
      "auxiliary_loss_mlp": 0.01027401,
      "balance_loss_clip": 1.04729223,
      "balance_loss_mlp": 1.02030802,
      "epoch": 0.4907112366981302,
      "flos": 17202217380480.0,
      "grad_norm": 1.916759785111905,
      "language_loss": 0.67214251,
      "learning_rate": 2.157189989615021e-06,
      "loss": 0.69576371,
      "num_input_tokens_seen": 87884270,
      "step": 4081,
      "time_per_iteration": 2.891043186187744
    },
    {
      "auxiliary_loss_clip": 0.01240931,
      "auxiliary_loss_mlp": 0.02572182,
      "balance_loss_clip": 1.05157208,
      "balance_loss_mlp": 1.00020254,
      "epoch": 0.4908314795887693,
      "flos": 21688968107520.0,
      "grad_norm": 3.6581874914413195,
      "language_loss": 0.74826384,
      "learning_rate": 2.156413415712913e-06,
      "loss": 0.78639501,
      "num_input_tokens_seen": 87906320,
      "step": 4082,
      "time_per_iteration": 2.761448621749878
    },
    {
      "auxiliary_loss_clip": 0.0129025,
      "auxiliary_loss_mlp": 0.02569688,
      "balance_loss_clip": 1.05090094,
      "balance_loss_mlp": 1.0002265,
      "epoch": 0.4909517224794084,
      "flos": 26213676531840.0,
      "grad_norm": 1.8617298862042326,
      "language_loss": 0.78987825,
      "learning_rate": 2.155636818083014e-06,
      "loss": 0.82847768,
      "num_input_tokens_seen": 87927690,
      "step": 4083,
      "time_per_iteration": 2.7744667530059814
    },
    {
      "auxiliary_loss_clip": 0.01278553,
      "auxiliary_loss_mlp": 0.01023399,
      "balance_loss_clip": 1.05027938,
      "balance_loss_mlp": 1.01618052,
      "epoch": 0.4910719653700475,
      "flos": 23148377694720.0,
      "grad_norm": 2.164436453279822,
      "language_loss": 0.83973384,
      "learning_rate": 2.154860196843134e-06,
      "loss": 0.86275333,
      "num_input_tokens_seen": 87946885,
      "step": 4084,
      "time_per_iteration": 2.7947726249694824
    },
    {
      "auxiliary_loss_clip": 0.01187373,
      "auxiliary_loss_mlp": 0.01030278,
      "balance_loss_clip": 1.05212045,
      "balance_loss_mlp": 1.02292013,
      "epoch": 0.4911922082606866,
      "flos": 23331845387520.0,
      "grad_norm": 1.9358775967418005,
      "language_loss": 0.76707506,
      "learning_rate": 2.154083552111085e-06,
      "loss": 0.78925151,
      "num_input_tokens_seen": 87966055,
      "step": 4085,
      "time_per_iteration": 2.7005770206451416
    },
    {
      "auxiliary_loss_clip": 0.01189607,
      "auxiliary_loss_mlp": 0.0102843,
      "balance_loss_clip": 1.05369461,
      "balance_loss_mlp": 1.02029991,
      "epoch": 0.49131245115132566,
      "flos": 29203239542400.0,
      "grad_norm": 2.221277937280941,
      "language_loss": 0.81707418,
      "learning_rate": 2.1533068840046834e-06,
      "loss": 0.83925462,
      "num_input_tokens_seen": 87986320,
      "step": 4086,
      "time_per_iteration": 2.7311692237854004
    },
    {
      "auxiliary_loss_clip": 0.01281672,
      "auxiliary_loss_mlp": 0.02571027,
      "balance_loss_clip": 1.0500474,
      "balance_loss_mlp": 1.00021267,
      "epoch": 0.49143269404196477,
      "flos": 20147465986560.0,
      "grad_norm": 8.024639708311575,
      "language_loss": 0.6168806,
      "learning_rate": 2.152530192641749e-06,
      "loss": 0.65540755,
      "num_input_tokens_seen": 88001230,
      "step": 4087,
      "time_per_iteration": 2.6845226287841797
    },
    {
      "auxiliary_loss_clip": 0.0114111,
      "auxiliary_loss_mlp": 0.01024815,
      "balance_loss_clip": 1.05155313,
      "balance_loss_mlp": 1.01740313,
      "epoch": 0.4915529369326039,
      "flos": 24389809597440.0,
      "grad_norm": 1.6920368708968616,
      "language_loss": 0.72350204,
      "learning_rate": 2.1517534781401068e-06,
      "loss": 0.74516129,
      "num_input_tokens_seen": 88019110,
      "step": 4088,
      "time_per_iteration": 2.7221028804779053
    },
    {
      "auxiliary_loss_clip": 0.01234108,
      "auxiliary_loss_mlp": 0.01025442,
      "balance_loss_clip": 1.04994154,
      "balance_loss_mlp": 1.01781535,
      "epoch": 0.49167317982324293,
      "flos": 10524305197440.0,
      "grad_norm": 3.3991941057863,
      "language_loss": 0.69562048,
      "learning_rate": 2.150976740617581e-06,
      "loss": 0.71821594,
      "num_input_tokens_seen": 88035670,
      "step": 4089,
      "time_per_iteration": 3.6098384857177734
    },
    {
      "auxiliary_loss_clip": 0.01289601,
      "auxiliary_loss_mlp": 0.01028228,
      "balance_loss_clip": 1.05084586,
      "balance_loss_mlp": 1.02088785,
      "epoch": 0.49179342271388204,
      "flos": 25593427457280.0,
      "grad_norm": 1.8711344686404499,
      "language_loss": 0.71469527,
      "learning_rate": 2.150199980192006e-06,
      "loss": 0.73787355,
      "num_input_tokens_seen": 88054790,
      "step": 4090,
      "time_per_iteration": 2.732696771621704
    },
    {
      "auxiliary_loss_clip": 0.01279554,
      "auxiliary_loss_mlp": 0.01027998,
      "balance_loss_clip": 1.04744697,
      "balance_loss_mlp": 1.02075577,
      "epoch": 0.49191366560452116,
      "flos": 21102043875840.0,
      "grad_norm": 1.8921588816844086,
      "language_loss": 0.80599666,
      "learning_rate": 2.1494231969812114e-06,
      "loss": 0.82907212,
      "num_input_tokens_seen": 88073780,
      "step": 4091,
      "time_per_iteration": 3.616732358932495
    },
    {
      "auxiliary_loss_clip": 0.01342677,
      "auxiliary_loss_mlp": 0.01028141,
      "balance_loss_clip": 1.05230904,
      "balance_loss_mlp": 1.02077985,
      "epoch": 0.4920339084951602,
      "flos": 26067520091520.0,
      "grad_norm": 2.1815065462757124,
      "language_loss": 0.81269741,
      "learning_rate": 2.1486463911030372e-06,
      "loss": 0.83640558,
      "num_input_tokens_seen": 88094430,
      "step": 4092,
      "time_per_iteration": 2.7819273471832275
    },
    {
      "auxiliary_loss_clip": 0.01281257,
      "auxiliary_loss_mlp": 0.01026325,
      "balance_loss_clip": 1.04741979,
      "balance_loss_mlp": 1.01932454,
      "epoch": 0.4921541513857993,
      "flos": 25081269384960.0,
      "grad_norm": 1.8011909947546807,
      "language_loss": 0.74700707,
      "learning_rate": 2.147869562675324e-06,
      "loss": 0.77008289,
      "num_input_tokens_seen": 88113400,
      "step": 4093,
      "time_per_iteration": 2.737931489944458
    },
    {
      "auxiliary_loss_clip": 0.01235235,
      "auxiliary_loss_mlp": 0.01023218,
      "balance_loss_clip": 1.05195022,
      "balance_loss_mlp": 1.01524878,
      "epoch": 0.49227439427643843,
      "flos": 24389809597440.0,
      "grad_norm": 2.3884726910761764,
      "language_loss": 0.72565341,
      "learning_rate": 2.147092711815915e-06,
      "loss": 0.74823797,
      "num_input_tokens_seen": 88132750,
      "step": 4094,
      "time_per_iteration": 2.782163619995117
    },
    {
      "auxiliary_loss_clip": 0.01329956,
      "auxiliary_loss_mlp": 0.0102591,
      "balance_loss_clip": 1.05011284,
      "balance_loss_mlp": 1.01860857,
      "epoch": 0.4923946371670775,
      "flos": 11363753018880.0,
      "grad_norm": 2.2295396392128706,
      "language_loss": 0.86331928,
      "learning_rate": 2.1463158386426593e-06,
      "loss": 0.88687789,
      "num_input_tokens_seen": 88150560,
      "step": 4095,
      "time_per_iteration": 3.529221296310425
    },
    {
      "auxiliary_loss_clip": 0.01288965,
      "auxiliary_loss_mlp": 0.0102779,
      "balance_loss_clip": 1.05057907,
      "balance_loss_mlp": 1.01999629,
      "epoch": 0.4925148800577166,
      "flos": 30445964334720.0,
      "grad_norm": 1.902701111938967,
      "language_loss": 0.76874733,
      "learning_rate": 2.145538943273407e-06,
      "loss": 0.79191488,
      "num_input_tokens_seen": 88170835,
      "step": 4096,
      "time_per_iteration": 2.8493595123291016
    },
    {
      "auxiliary_loss_clip": 0.01188854,
      "auxiliary_loss_mlp": 0.01032103,
      "balance_loss_clip": 1.05439425,
      "balance_loss_mlp": 1.02450633,
      "epoch": 0.49263512294835565,
      "flos": 20850454039680.0,
      "grad_norm": 2.1293576258097278,
      "language_loss": 0.71816349,
      "learning_rate": 2.144762025826013e-06,
      "loss": 0.74037302,
      "num_input_tokens_seen": 88189925,
      "step": 4097,
      "time_per_iteration": 3.4842960834503174
    },
    {
      "auxiliary_loss_clip": 0.01238199,
      "auxiliary_loss_mlp": 0.01028508,
      "balance_loss_clip": 1.04968524,
      "balance_loss_mlp": 1.02094412,
      "epoch": 0.49275536583899476,
      "flos": 23767477534080.0,
      "grad_norm": 2.51941093502445,
      "language_loss": 0.87001157,
      "learning_rate": 2.143985086418334e-06,
      "loss": 0.89267862,
      "num_input_tokens_seen": 88205105,
      "step": 4098,
      "time_per_iteration": 2.8092446327209473
    },
    {
      "auxiliary_loss_clip": 0.01285027,
      "auxiliary_loss_mlp": 0.0102496,
      "balance_loss_clip": 1.04794168,
      "balance_loss_mlp": 1.01783085,
      "epoch": 0.4928756087296339,
      "flos": 22273522041600.0,
      "grad_norm": 1.7898924274984098,
      "language_loss": 0.76638609,
      "learning_rate": 2.1432081251682324e-06,
      "loss": 0.78948599,
      "num_input_tokens_seen": 88225475,
      "step": 4099,
      "time_per_iteration": 2.718900203704834
    },
    {
      "auxiliary_loss_clip": 0.01245585,
      "auxiliary_loss_mlp": 0.01026588,
      "balance_loss_clip": 1.05926132,
      "balance_loss_mlp": 1.0193584,
      "epoch": 0.49299585162027293,
      "flos": 19645471463040.0,
      "grad_norm": 1.832696759631203,
      "language_loss": 0.8698895,
      "learning_rate": 2.142431142193572e-06,
      "loss": 0.89261121,
      "num_input_tokens_seen": 88243255,
      "step": 4100,
      "time_per_iteration": 2.673250675201416
    },
    {
      "auxiliary_loss_clip": 0.01184203,
      "auxiliary_loss_mlp": 0.0102769,
      "balance_loss_clip": 1.05251741,
      "balance_loss_mlp": 1.02034426,
      "epoch": 0.49311609451091204,
      "flos": 38837138497920.0,
      "grad_norm": 2.8837080944613485,
      "language_loss": 0.71365535,
      "learning_rate": 2.1416541376122207e-06,
      "loss": 0.73577428,
      "num_input_tokens_seen": 88263435,
      "step": 4101,
      "time_per_iteration": 2.7697925567626953
    },
    {
      "auxiliary_loss_clip": 0.01184945,
      "auxiliary_loss_mlp": 0.01026237,
      "balance_loss_clip": 1.04997706,
      "balance_loss_mlp": 1.01870263,
      "epoch": 0.49323633740155115,
      "flos": 28329102161280.0,
      "grad_norm": 2.0264516454909995,
      "language_loss": 0.73169315,
      "learning_rate": 2.1408771115420496e-06,
      "loss": 0.75380498,
      "num_input_tokens_seen": 88283295,
      "step": 4102,
      "time_per_iteration": 2.6863934993743896
    },
    {
      "auxiliary_loss_clip": 0.01397027,
      "auxiliary_loss_mlp": 0.01031422,
      "balance_loss_clip": 1.05727899,
      "balance_loss_mlp": 1.02417743,
      "epoch": 0.4933565802921902,
      "flos": 21135584200320.0,
      "grad_norm": 2.008328418644424,
      "language_loss": 0.64935887,
      "learning_rate": 2.140100064100932e-06,
      "loss": 0.67364329,
      "num_input_tokens_seen": 88299270,
      "step": 4103,
      "time_per_iteration": 2.858157157897949
    },
    {
      "auxiliary_loss_clip": 0.01231575,
      "auxiliary_loss_mlp": 0.01025943,
      "balance_loss_clip": 1.05229402,
      "balance_loss_mlp": 1.0189693,
      "epoch": 0.4934768231828293,
      "flos": 18039007595520.0,
      "grad_norm": 3.1755702641989827,
      "language_loss": 0.75709891,
      "learning_rate": 2.139322995406746e-06,
      "loss": 0.77967405,
      "num_input_tokens_seen": 88316905,
      "step": 4104,
      "time_per_iteration": 2.6984400749206543
    },
    {
      "auxiliary_loss_clip": 0.01186743,
      "auxiliary_loss_mlp": 0.01031155,
      "balance_loss_clip": 1.05378222,
      "balance_loss_mlp": 1.02349579,
      "epoch": 0.4935970660734684,
      "flos": 23469957181440.0,
      "grad_norm": 2.1668601685003255,
      "language_loss": 0.79699898,
      "learning_rate": 2.1385459055773727e-06,
      "loss": 0.81917799,
      "num_input_tokens_seen": 88335095,
      "step": 4105,
      "time_per_iteration": 2.685393810272217
    },
    {
      "auxiliary_loss_clip": 0.01368972,
      "auxiliary_loss_mlp": 0.02561272,
      "balance_loss_clip": 1.04313493,
      "balance_loss_mlp": 1.00019491,
      "epoch": 0.4937173089641075,
      "flos": 64479258840960.0,
      "grad_norm": 2.55958037116647,
      "language_loss": 0.7370832,
      "learning_rate": 2.137768794730696e-06,
      "loss": 0.77638566,
      "num_input_tokens_seen": 88358545,
      "step": 4106,
      "time_per_iteration": 3.2117631435394287
    },
    {
      "auxiliary_loss_clip": 0.01291072,
      "auxiliary_loss_mlp": 0.01028483,
      "balance_loss_clip": 1.05547488,
      "balance_loss_mlp": 1.02000082,
      "epoch": 0.4938375518547466,
      "flos": 22346025644160.0,
      "grad_norm": 1.8044477092487992,
      "language_loss": 0.80359882,
      "learning_rate": 2.1369916629846026e-06,
      "loss": 0.82679439,
      "num_input_tokens_seen": 88378295,
      "step": 4107,
      "time_per_iteration": 2.7997939586639404
    },
    {
      "auxiliary_loss_clip": 0.0128547,
      "auxiliary_loss_mlp": 0.01024446,
      "balance_loss_clip": 1.0492177,
      "balance_loss_mlp": 1.017061,
      "epoch": 0.4939577947453857,
      "flos": 17858700299520.0,
      "grad_norm": 1.8581552381994435,
      "language_loss": 0.75212002,
      "learning_rate": 2.136214510456983e-06,
      "loss": 0.7752192,
      "num_input_tokens_seen": 88396750,
      "step": 4108,
      "time_per_iteration": 2.961371421813965
    },
    {
      "auxiliary_loss_clip": 0.01294481,
      "auxiliary_loss_mlp": 0.02505666,
      "balance_loss_clip": 1.01515555,
      "balance_loss_mlp": 1.00014925,
      "epoch": 0.49407803763602476,
      "flos": 70066746875520.0,
      "grad_norm": 0.9234287445643253,
      "language_loss": 0.63131922,
      "learning_rate": 2.1354373372657296e-06,
      "loss": 0.6693207,
      "num_input_tokens_seen": 88455190,
      "step": 4109,
      "time_per_iteration": 3.343641519546509
    },
    {
      "auxiliary_loss_clip": 0.01186518,
      "auxiliary_loss_mlp": 0.01031284,
      "balance_loss_clip": 1.05332255,
      "balance_loss_mlp": 1.02403665,
      "epoch": 0.49419828052666387,
      "flos": 24317485562880.0,
      "grad_norm": 4.680266806208277,
      "language_loss": 0.7100426,
      "learning_rate": 2.1346601435287404e-06,
      "loss": 0.73222059,
      "num_input_tokens_seen": 88477460,
      "step": 4110,
      "time_per_iteration": 2.668780565261841
    },
    {
      "auxiliary_loss_clip": 0.01283876,
      "auxiliary_loss_mlp": 0.01026941,
      "balance_loss_clip": 1.04887652,
      "balance_loss_mlp": 1.01941621,
      "epoch": 0.494318523417303,
      "flos": 29386060790400.0,
      "grad_norm": 1.8642869406623956,
      "language_loss": 0.80206895,
      "learning_rate": 2.1338829293639144e-06,
      "loss": 0.82517713,
      "num_input_tokens_seen": 88497820,
      "step": 4111,
      "time_per_iteration": 2.7817623615264893
    },
    {
      "auxiliary_loss_clip": 0.01390324,
      "auxiliary_loss_mlp": 0.01032528,
      "balance_loss_clip": 1.0515151,
      "balance_loss_mlp": 1.02495217,
      "epoch": 0.49443876630794203,
      "flos": 15268284195840.0,
      "grad_norm": 3.018294106867742,
      "language_loss": 0.83110428,
      "learning_rate": 2.1331056948891547e-06,
      "loss": 0.85533279,
      "num_input_tokens_seen": 88514920,
      "step": 4112,
      "time_per_iteration": 2.835930824279785
    },
    {
      "auxiliary_loss_clip": 0.01280888,
      "auxiliary_loss_mlp": 0.01028117,
      "balance_loss_clip": 1.04872656,
      "balance_loss_mlp": 1.02061236,
      "epoch": 0.49455900919858115,
      "flos": 12347453859840.0,
      "grad_norm": 2.2283260197248094,
      "language_loss": 0.75715554,
      "learning_rate": 2.1323284402223666e-06,
      "loss": 0.78024566,
      "num_input_tokens_seen": 88530910,
      "step": 4113,
      "time_per_iteration": 2.7047362327575684
    },
    {
      "auxiliary_loss_clip": 0.01182911,
      "auxiliary_loss_mlp": 0.02558821,
      "balance_loss_clip": 1.05447924,
      "balance_loss_mlp": 1.00030851,
      "epoch": 0.4946792520892202,
      "flos": 22779610715520.0,
      "grad_norm": 1.777977710740508,
      "language_loss": 0.87982798,
      "learning_rate": 2.1315511654814597e-06,
      "loss": 0.91724533,
      "num_input_tokens_seen": 88549320,
      "step": 4114,
      "time_per_iteration": 2.6819467544555664
    },
    {
      "auxiliary_loss_clip": 0.0127598,
      "auxiliary_loss_mlp": 0.0102562,
      "balance_loss_clip": 1.05074704,
      "balance_loss_mlp": 1.01847041,
      "epoch": 0.4947994949798593,
      "flos": 23148126299520.0,
      "grad_norm": 2.000460230129795,
      "language_loss": 0.78479874,
      "learning_rate": 2.1307738707843456e-06,
      "loss": 0.80781472,
      "num_input_tokens_seen": 88568985,
      "step": 4115,
      "time_per_iteration": 3.734989643096924
    },
    {
      "auxiliary_loss_clip": 0.01242246,
      "auxiliary_loss_mlp": 0.01022292,
      "balance_loss_clip": 1.05353498,
      "balance_loss_mlp": 1.01424301,
      "epoch": 0.4949197378704984,
      "flos": 23659997063040.0,
      "grad_norm": 2.2713729443568265,
      "language_loss": 0.69336343,
      "learning_rate": 2.1299965562489385e-06,
      "loss": 0.71600878,
      "num_input_tokens_seen": 88588790,
      "step": 4116,
      "time_per_iteration": 2.6821272373199463
    },
    {
      "auxiliary_loss_clip": 0.01234056,
      "auxiliary_loss_mlp": 0.01024241,
      "balance_loss_clip": 1.04947352,
      "balance_loss_mlp": 1.01654053,
      "epoch": 0.4950399807611375,
      "flos": 26911493026560.0,
      "grad_norm": 1.6561417931797737,
      "language_loss": 0.79005951,
      "learning_rate": 2.129219221993158e-06,
      "loss": 0.81264246,
      "num_input_tokens_seen": 88613575,
      "step": 4117,
      "time_per_iteration": 3.688939094543457
    },
    {
      "auxiliary_loss_clip": 0.0123443,
      "auxiliary_loss_mlp": 0.00999281,
      "balance_loss_clip": 1.01385188,
      "balance_loss_mlp": 0.9983328,
      "epoch": 0.4951602236517766,
      "flos": 67315270187520.0,
      "grad_norm": 0.7883896176894193,
      "language_loss": 0.59912694,
      "learning_rate": 2.128441868134924e-06,
      "loss": 0.62146401,
      "num_input_tokens_seen": 88675510,
      "step": 4118,
      "time_per_iteration": 3.3732128143310547
    },
    {
      "auxiliary_loss_clip": 0.01334338,
      "auxiliary_loss_mlp": 0.01030946,
      "balance_loss_clip": 1.04743481,
      "balance_loss_mlp": 1.02312636,
      "epoch": 0.4952804665424157,
      "flos": 19901442758400.0,
      "grad_norm": 2.576263561979592,
      "language_loss": 0.82660443,
      "learning_rate": 2.1276644947921606e-06,
      "loss": 0.85025728,
      "num_input_tokens_seen": 88694425,
      "step": 4119,
      "time_per_iteration": 2.7892258167266846
    },
    {
      "auxiliary_loss_clip": 0.01234679,
      "auxiliary_loss_mlp": 0.01029331,
      "balance_loss_clip": 1.05057621,
      "balance_loss_mlp": 1.0217731,
      "epoch": 0.49540070943305475,
      "flos": 18806813740800.0,
      "grad_norm": 1.896184670079025,
      "language_loss": 0.82611477,
      "learning_rate": 2.126887102082795e-06,
      "loss": 0.84875488,
      "num_input_tokens_seen": 88714450,
      "step": 4120,
      "time_per_iteration": 2.743011236190796
    },
    {
      "auxiliary_loss_clip": 0.01329868,
      "auxiliary_loss_mlp": 0.01022568,
      "balance_loss_clip": 1.04468274,
      "balance_loss_mlp": 1.01512921,
      "epoch": 0.49552095232369386,
      "flos": 24934179191040.0,
      "grad_norm": 2.0209503693419366,
      "language_loss": 0.70768607,
      "learning_rate": 2.126109690124757e-06,
      "loss": 0.73121047,
      "num_input_tokens_seen": 88735265,
      "step": 4121,
      "time_per_iteration": 3.665943145751953
    },
    {
      "auxiliary_loss_clip": 0.01381985,
      "auxiliary_loss_mlp": 0.01033814,
      "balance_loss_clip": 1.04541183,
      "balance_loss_mlp": 1.02663171,
      "epoch": 0.495641195214333,
      "flos": 22857249962880.0,
      "grad_norm": 1.662271360464674,
      "language_loss": 0.70968306,
      "learning_rate": 2.1253322590359786e-06,
      "loss": 0.73384106,
      "num_input_tokens_seen": 88754600,
      "step": 4122,
      "time_per_iteration": 2.8502326011657715
    },
    {
      "auxiliary_loss_clip": 0.01236865,
      "auxiliary_loss_mlp": 0.01031389,
      "balance_loss_clip": 1.0512054,
      "balance_loss_mlp": 1.02375388,
      "epoch": 0.49576143810497203,
      "flos": 25769748343680.0,
      "grad_norm": 1.9843581420581209,
      "language_loss": 0.74111766,
      "learning_rate": 2.124554808934397e-06,
      "loss": 0.7638002,
      "num_input_tokens_seen": 88775180,
      "step": 4123,
      "time_per_iteration": 3.661850929260254
    },
    {
      "auxiliary_loss_clip": 0.01383197,
      "auxiliary_loss_mlp": 0.0102725,
      "balance_loss_clip": 1.04421806,
      "balance_loss_mlp": 1.01979947,
      "epoch": 0.49588168099561114,
      "flos": 22128838058880.0,
      "grad_norm": 1.959225912449428,
      "language_loss": 0.73079509,
      "learning_rate": 2.1237773399379496e-06,
      "loss": 0.75489956,
      "num_input_tokens_seen": 88796145,
      "step": 4124,
      "time_per_iteration": 2.771360158920288
    },
    {
      "auxiliary_loss_clip": 0.01188454,
      "auxiliary_loss_mlp": 0.01028127,
      "balance_loss_clip": 1.04475677,
      "balance_loss_mlp": 1.02087331,
      "epoch": 0.49600192388625025,
      "flos": 24387331559040.0,
      "grad_norm": 2.1712499920897956,
      "language_loss": 0.87085676,
      "learning_rate": 2.122999852164578e-06,
      "loss": 0.8930226,
      "num_input_tokens_seen": 88816765,
      "step": 4125,
      "time_per_iteration": 2.806997776031494
    },
    {
      "auxiliary_loss_clip": 0.01286484,
      "auxiliary_loss_mlp": 0.01029957,
      "balance_loss_clip": 1.05077624,
      "balance_loss_mlp": 1.0223279,
      "epoch": 0.4961221667768893,
      "flos": 22857429530880.0,
      "grad_norm": 2.72483758024556,
      "language_loss": 0.58656317,
      "learning_rate": 2.122222345732227e-06,
      "loss": 0.60972756,
      "num_input_tokens_seen": 88836680,
      "step": 4126,
      "time_per_iteration": 2.769883155822754
    },
    {
      "auxiliary_loss_clip": 0.01336034,
      "auxiliary_loss_mlp": 0.01033973,
      "balance_loss_clip": 1.04873097,
      "balance_loss_mlp": 1.02655792,
      "epoch": 0.4962424096675284,
      "flos": 17858089768320.0,
      "grad_norm": 1.9269496863668125,
      "language_loss": 0.83362901,
      "learning_rate": 2.121444820758843e-06,
      "loss": 0.85732913,
      "num_input_tokens_seen": 88855320,
      "step": 4127,
      "time_per_iteration": 2.675095796585083
    },
    {
      "auxiliary_loss_clip": 0.01383968,
      "auxiliary_loss_mlp": 0.01026862,
      "balance_loss_clip": 1.04857564,
      "balance_loss_mlp": 1.01877332,
      "epoch": 0.49636265255816747,
      "flos": 21793611404160.0,
      "grad_norm": 2.1996909702834,
      "language_loss": 0.7898978,
      "learning_rate": 2.120667277362376e-06,
      "loss": 0.81400609,
      "num_input_tokens_seen": 88874035,
      "step": 4128,
      "time_per_iteration": 2.8890936374664307
    },
    {
      "auxiliary_loss_clip": 0.01193313,
      "auxiliary_loss_mlp": 0.01039329,
      "balance_loss_clip": 1.05761123,
      "balance_loss_mlp": 1.03124046,
      "epoch": 0.4964828954488066,
      "flos": 16358603581440.0,
      "grad_norm": 2.0442151403814095,
      "language_loss": 0.8543098,
      "learning_rate": 2.1198897156607796e-06,
      "loss": 0.87663627,
      "num_input_tokens_seen": 88891390,
      "step": 4129,
      "time_per_iteration": 2.6068060398101807
    },
    {
      "auxiliary_loss_clip": 0.01143447,
      "auxiliary_loss_mlp": 0.01025197,
      "balance_loss_clip": 1.05213952,
      "balance_loss_mlp": 1.01737738,
      "epoch": 0.4966031383394457,
      "flos": 24711101775360.0,
      "grad_norm": 2.2830114380656044,
      "language_loss": 0.74228281,
      "learning_rate": 2.1191121357720085e-06,
      "loss": 0.7639693,
      "num_input_tokens_seen": 88909450,
      "step": 4130,
      "time_per_iteration": 2.687800645828247
    },
    {
      "auxiliary_loss_clip": 0.01385584,
      "auxiliary_loss_mlp": 0.01028703,
      "balance_loss_clip": 1.04923987,
      "balance_loss_mlp": 1.02073956,
      "epoch": 0.49672338123008475,
      "flos": 22930615491840.0,
      "grad_norm": 1.7530694012475256,
      "language_loss": 0.7465927,
      "learning_rate": 2.1183345378140206e-06,
      "loss": 0.77073562,
      "num_input_tokens_seen": 88929195,
      "step": 4131,
      "time_per_iteration": 2.7826993465423584
    },
    {
      "auxiliary_loss_clip": 0.01129958,
      "auxiliary_loss_mlp": 0.00998167,
      "balance_loss_clip": 1.01160979,
      "balance_loss_mlp": 0.99710578,
      "epoch": 0.49684362412072386,
      "flos": 65976736844160.0,
      "grad_norm": 0.8536412097201204,
      "language_loss": 0.61993551,
      "learning_rate": 2.1175569219047783e-06,
      "loss": 0.64121675,
      "num_input_tokens_seen": 88990635,
      "step": 4132,
      "time_per_iteration": 3.391906499862671
    },
    {
      "auxiliary_loss_clip": 0.01185225,
      "auxiliary_loss_mlp": 0.0103217,
      "balance_loss_clip": 1.05251336,
      "balance_loss_mlp": 1.02390838,
      "epoch": 0.49696386701136297,
      "flos": 19971288754560.0,
      "grad_norm": 3.2406818973788813,
      "language_loss": 0.73328173,
      "learning_rate": 2.1167792881622437e-06,
      "loss": 0.75545567,
      "num_input_tokens_seen": 89009655,
      "step": 4133,
      "time_per_iteration": 2.6174063682556152
    },
    {
      "auxiliary_loss_clip": 0.01280159,
      "auxiliary_loss_mlp": 0.01028959,
      "balance_loss_clip": 1.05011284,
      "balance_loss_mlp": 1.02169371,
      "epoch": 0.497084109902002,
      "flos": 24750819239040.0,
      "grad_norm": 2.3839916320621595,
      "language_loss": 0.809376,
      "learning_rate": 2.116001636704384e-06,
      "loss": 0.8324672,
      "num_input_tokens_seen": 89030040,
      "step": 4134,
      "time_per_iteration": 2.7341067790985107
    },
    {
      "auxiliary_loss_clip": 0.01296179,
      "auxiliary_loss_mlp": 0.01033046,
      "balance_loss_clip": 1.04823005,
      "balance_loss_mlp": 1.02592611,
      "epoch": 0.49720435279264114,
      "flos": 21871825269120.0,
      "grad_norm": 2.000744197278107,
      "language_loss": 0.80563879,
      "learning_rate": 2.1152239676491685e-06,
      "loss": 0.82893097,
      "num_input_tokens_seen": 89048145,
      "step": 4135,
      "time_per_iteration": 2.771641969680786
    },
    {
      "auxiliary_loss_clip": 0.01289147,
      "auxiliary_loss_mlp": 0.01024978,
      "balance_loss_clip": 1.0470649,
      "balance_loss_mlp": 1.01802778,
      "epoch": 0.49732459568328025,
      "flos": 23805794367360.0,
      "grad_norm": 1.7735905969710455,
      "language_loss": 0.73600388,
      "learning_rate": 2.114446281114569e-06,
      "loss": 0.75914514,
      "num_input_tokens_seen": 89067165,
      "step": 4136,
      "time_per_iteration": 2.7590909004211426
    },
    {
      "auxiliary_loss_clip": 0.01278161,
      "auxiliary_loss_mlp": 0.01024806,
      "balance_loss_clip": 1.04982471,
      "balance_loss_mlp": 1.01783586,
      "epoch": 0.4974448385739193,
      "flos": 20047742853120.0,
      "grad_norm": 2.0366895695850125,
      "language_loss": 0.76135612,
      "learning_rate": 2.1136685772185587e-06,
      "loss": 0.7843858,
      "num_input_tokens_seen": 89086190,
      "step": 4137,
      "time_per_iteration": 2.7410471439361572
    },
    {
      "auxiliary_loss_clip": 0.01286062,
      "auxiliary_loss_mlp": 0.02568736,
      "balance_loss_clip": 1.045946,
      "balance_loss_mlp": 1.00043595,
      "epoch": 0.4975650814645584,
      "flos": 24821347593600.0,
      "grad_norm": 1.990783538618622,
      "language_loss": 0.77984929,
      "learning_rate": 2.1128908560791163e-06,
      "loss": 0.81839734,
      "num_input_tokens_seen": 89106020,
      "step": 4138,
      "time_per_iteration": 2.75510573387146
    },
    {
      "auxiliary_loss_clip": 0.0118583,
      "auxiliary_loss_mlp": 0.01021808,
      "balance_loss_clip": 1.0529933,
      "balance_loss_mlp": 1.01460195,
      "epoch": 0.4976853243551975,
      "flos": 19829477859840.0,
      "grad_norm": 2.1673512861386497,
      "language_loss": 0.78505754,
      "learning_rate": 2.1121131178142203e-06,
      "loss": 0.80713391,
      "num_input_tokens_seen": 89125385,
      "step": 4139,
      "time_per_iteration": 2.6770565509796143
    },
    {
      "auxiliary_loss_clip": 0.01281165,
      "auxiliary_loss_mlp": 0.01029763,
      "balance_loss_clip": 1.04693365,
      "balance_loss_mlp": 1.02181745,
      "epoch": 0.4978055672458366,
      "flos": 23142990654720.0,
      "grad_norm": 1.5166479663822954,
      "language_loss": 0.82638192,
      "learning_rate": 2.1113353625418544e-06,
      "loss": 0.84949124,
      "num_input_tokens_seen": 89143935,
      "step": 4140,
      "time_per_iteration": 4.192366600036621
    },
    {
      "auxiliary_loss_clip": 0.01228932,
      "auxiliary_loss_mlp": 0.01025331,
      "balance_loss_clip": 1.05331278,
      "balance_loss_mlp": 1.01876569,
      "epoch": 0.4979258101364757,
      "flos": 15559914718080.0,
      "grad_norm": 1.767591121728797,
      "language_loss": 0.78912973,
      "learning_rate": 2.1105575903800017e-06,
      "loss": 0.81167233,
      "num_input_tokens_seen": 89162655,
      "step": 4141,
      "time_per_iteration": 2.7218494415283203
    },
    {
      "auxiliary_loss_clip": 0.01142432,
      "auxiliary_loss_mlp": 0.01030382,
      "balance_loss_clip": 1.05168831,
      "balance_loss_mlp": 1.02313113,
      "epoch": 0.4980460530271148,
      "flos": 26356169784960.0,
      "grad_norm": 2.1399130191919413,
      "language_loss": 0.85309893,
      "learning_rate": 2.1097798014466502e-06,
      "loss": 0.87482703,
      "num_input_tokens_seen": 89182255,
      "step": 4142,
      "time_per_iteration": 2.7035794258117676
    },
    {
      "auxiliary_loss_clip": 0.01239797,
      "auxiliary_loss_mlp": 0.01026651,
      "balance_loss_clip": 1.05290616,
      "balance_loss_mlp": 1.01895571,
      "epoch": 0.49816629591775385,
      "flos": 17274541415040.0,
      "grad_norm": 8.457942669996953,
      "language_loss": 0.59238994,
      "learning_rate": 2.109001995859791e-06,
      "loss": 0.61505437,
      "num_input_tokens_seen": 89201155,
      "step": 4143,
      "time_per_iteration": 3.457157611846924
    },
    {
      "auxiliary_loss_clip": 0.01176141,
      "auxiliary_loss_mlp": 0.01012074,
      "balance_loss_clip": 1.0135684,
      "balance_loss_mlp": 1.01117361,
      "epoch": 0.49828653880839296,
      "flos": 64930947344640.0,
      "grad_norm": 0.8071210640303216,
      "language_loss": 0.60024107,
      "learning_rate": 2.108224173737415e-06,
      "loss": 0.62212324,
      "num_input_tokens_seen": 89264455,
      "step": 4144,
      "time_per_iteration": 3.2433512210845947
    },
    {
      "auxiliary_loss_clip": 0.01280665,
      "auxiliary_loss_mlp": 0.01030796,
      "balance_loss_clip": 1.04719234,
      "balance_loss_mlp": 1.02231431,
      "epoch": 0.498406781699032,
      "flos": 27484806003840.0,
      "grad_norm": 15.776375112544851,
      "language_loss": 0.75651228,
      "learning_rate": 2.1074463351975183e-06,
      "loss": 0.77962697,
      "num_input_tokens_seen": 89283340,
      "step": 4145,
      "time_per_iteration": 2.6877658367156982
    },
    {
      "auxiliary_loss_clip": 0.01242824,
      "auxiliary_loss_mlp": 0.01025006,
      "balance_loss_clip": 1.05129957,
      "balance_loss_mlp": 1.01768064,
      "epoch": 0.49852702458967113,
      "flos": 31499870307840.0,
      "grad_norm": 1.7591895907636477,
      "language_loss": 0.71433794,
      "learning_rate": 2.106668480358098e-06,
      "loss": 0.7370162,
      "num_input_tokens_seen": 89303565,
      "step": 4146,
      "time_per_iteration": 2.8086867332458496
    },
    {
      "auxiliary_loss_clip": 0.01247041,
      "auxiliary_loss_mlp": 0.01030543,
      "balance_loss_clip": 1.04713202,
      "balance_loss_mlp": 1.02207315,
      "epoch": 0.49864726748031024,
      "flos": 22852868503680.0,
      "grad_norm": 2.1804721297455645,
      "language_loss": 0.71078098,
      "learning_rate": 2.105890609337154e-06,
      "loss": 0.73355681,
      "num_input_tokens_seen": 89322080,
      "step": 4147,
      "time_per_iteration": 3.666923999786377
    },
    {
      "auxiliary_loss_clip": 0.01072866,
      "auxiliary_loss_mlp": 0.01003056,
      "balance_loss_clip": 1.01148808,
      "balance_loss_mlp": 1.00210786,
      "epoch": 0.4987675103709493,
      "flos": 70405708544640.0,
      "grad_norm": 0.6902665917836267,
      "language_loss": 0.63811994,
      "learning_rate": 2.1051127222526883e-06,
      "loss": 0.65887916,
      "num_input_tokens_seen": 89394195,
      "step": 4148,
      "time_per_iteration": 3.2904574871063232
    },
    {
      "auxiliary_loss_clip": 0.01232325,
      "auxiliary_loss_mlp": 0.01029012,
      "balance_loss_clip": 1.05274963,
      "balance_loss_mlp": 1.021842,
      "epoch": 0.4988877532615884,
      "flos": 28767571482240.0,
      "grad_norm": 1.7297559188118679,
      "language_loss": 0.80757177,
      "learning_rate": 2.1043348192227067e-06,
      "loss": 0.83018517,
      "num_input_tokens_seen": 89414565,
      "step": 4149,
      "time_per_iteration": 3.670820951461792
    },
    {
      "auxiliary_loss_clip": 0.01327635,
      "auxiliary_loss_mlp": 0.01021702,
      "balance_loss_clip": 1.04896104,
      "balance_loss_mlp": 1.01471615,
      "epoch": 0.4990079961522275,
      "flos": 16872700988160.0,
      "grad_norm": 1.840613908049841,
      "language_loss": 0.6196779,
      "learning_rate": 2.1035569003652156e-06,
      "loss": 0.64317131,
      "num_input_tokens_seen": 89433195,
      "step": 4150,
      "time_per_iteration": 2.7445223331451416
    },
    {
      "auxiliary_loss_clip": 0.0138814,
      "auxiliary_loss_mlp": 0.01029009,
      "balance_loss_clip": 1.04949856,
      "balance_loss_mlp": 1.02074814,
      "epoch": 0.4991282390428666,
      "flos": 13291042187520.0,
      "grad_norm": 2.0128663384647028,
      "language_loss": 0.81533909,
      "learning_rate": 2.1027789657982255e-06,
      "loss": 0.83951062,
      "num_input_tokens_seen": 89447410,
      "step": 4151,
      "time_per_iteration": 2.704967737197876
    },
    {
      "auxiliary_loss_clip": 0.01283563,
      "auxiliary_loss_mlp": 0.0102936,
      "balance_loss_clip": 1.04754543,
      "balance_loss_mlp": 1.02208495,
      "epoch": 0.4992484819335057,
      "flos": 21537496454400.0,
      "grad_norm": 2.010187240126557,
      "language_loss": 0.77461052,
      "learning_rate": 2.1020010156397482e-06,
      "loss": 0.79773968,
      "num_input_tokens_seen": 89464630,
      "step": 4152,
      "time_per_iteration": 2.7819645404815674
    },
    {
      "auxiliary_loss_clip": 0.01237203,
      "auxiliary_loss_mlp": 0.01033576,
      "balance_loss_clip": 1.05195916,
      "balance_loss_mlp": 1.02580357,
      "epoch": 0.4993687248241448,
      "flos": 24860095390080.0,
      "grad_norm": 2.0869407868930256,
      "language_loss": 0.77526015,
      "learning_rate": 2.101223050007797e-06,
      "loss": 0.79796791,
      "num_input_tokens_seen": 89483180,
      "step": 4153,
      "time_per_iteration": 2.6645030975341797
    },
    {
      "auxiliary_loss_clip": 0.01072512,
      "auxiliary_loss_mlp": 0.01001767,
      "balance_loss_clip": 1.01132619,
      "balance_loss_mlp": 1.00081944,
      "epoch": 0.49948896771478385,
      "flos": 62941602453120.0,
      "grad_norm": 0.8101946151117984,
      "language_loss": 0.53807145,
      "learning_rate": 2.1004450690203904e-06,
      "loss": 0.55881423,
      "num_input_tokens_seen": 89539260,
      "step": 4154,
      "time_per_iteration": 3.2060556411743164
    },
    {
      "auxiliary_loss_clip": 0.01072374,
      "auxiliary_loss_mlp": 0.01000375,
      "balance_loss_clip": 1.01115263,
      "balance_loss_mlp": 0.99943954,
      "epoch": 0.49960921060542296,
      "flos": 68284213516800.0,
      "grad_norm": 0.8522600901112465,
      "language_loss": 0.63299453,
      "learning_rate": 2.099667072795546e-06,
      "loss": 0.65372199,
      "num_input_tokens_seen": 89601380,
      "step": 4155,
      "time_per_iteration": 3.2112271785736084
    },
    {
      "auxiliary_loss_clip": 0.01232141,
      "auxiliary_loss_mlp": 0.01023518,
      "balance_loss_clip": 1.05015349,
      "balance_loss_mlp": 1.01637793,
      "epoch": 0.49972945349606207,
      "flos": 23659350618240.0,
      "grad_norm": 1.9862424875835254,
      "language_loss": 0.79621822,
      "learning_rate": 2.0988890614512864e-06,
      "loss": 0.81877482,
      "num_input_tokens_seen": 89621270,
      "step": 4156,
      "time_per_iteration": 2.664026975631714
    },
    {
      "auxiliary_loss_clip": 0.01285848,
      "auxiliary_loss_mlp": 0.01029702,
      "balance_loss_clip": 1.05180836,
      "balance_loss_mlp": 1.0221622,
      "epoch": 0.4998496963867011,
      "flos": 19755825022080.0,
      "grad_norm": 1.7470603831291087,
      "language_loss": 0.84282792,
      "learning_rate": 2.098111035105635e-06,
      "loss": 0.86598337,
      "num_input_tokens_seen": 89639695,
      "step": 4157,
      "time_per_iteration": 2.6992902755737305
    },
    {
      "auxiliary_loss_clip": 0.01281077,
      "auxiliary_loss_mlp": 0.01027089,
      "balance_loss_clip": 1.04899049,
      "balance_loss_mlp": 1.01921558,
      "epoch": 0.49996993927734024,
      "flos": 22265728790400.0,
      "grad_norm": 2.2433393757423588,
      "language_loss": 0.73282903,
      "learning_rate": 2.0973329938766176e-06,
      "loss": 0.75591075,
      "num_input_tokens_seen": 89657125,
      "step": 4158,
      "time_per_iteration": 2.738851547241211
    },
    {
      "auxiliary_loss_clip": 0.01242852,
      "auxiliary_loss_mlp": 0.0103466,
      "balance_loss_clip": 1.05370235,
      "balance_loss_mlp": 1.02705789,
      "epoch": 0.5000901821679793,
      "flos": 23327212533120.0,
      "grad_norm": 2.0303538648133053,
      "language_loss": 0.78890651,
      "learning_rate": 2.0965549378822618e-06,
      "loss": 0.81168163,
      "num_input_tokens_seen": 89678415,
      "step": 4159,
      "time_per_iteration": 2.7415287494659424
    },
    {
      "auxiliary_loss_clip": 0.0143216,
      "auxiliary_loss_mlp": 0.01031685,
      "balance_loss_clip": 1.04184544,
      "balance_loss_mlp": 1.02366853,
      "epoch": 0.5002104250586185,
      "flos": 20339014239360.0,
      "grad_norm": 2.0146628136041844,
      "language_loss": 0.84187973,
      "learning_rate": 2.095776867240599e-06,
      "loss": 0.8665182,
      "num_input_tokens_seen": 89695405,
      "step": 4160,
      "time_per_iteration": 2.911146402359009
    },
    {
      "auxiliary_loss_clip": 0.01331172,
      "auxiliary_loss_mlp": 0.01029009,
      "balance_loss_clip": 1.04958868,
      "balance_loss_mlp": 1.02139807,
      "epoch": 0.5003306679492575,
      "flos": 13991372634240.0,
      "grad_norm": 2.8157786245615264,
      "language_loss": 0.82708722,
      "learning_rate": 2.094998782069661e-06,
      "loss": 0.85068905,
      "num_input_tokens_seen": 89713110,
      "step": 4161,
      "time_per_iteration": 2.91485333442688
    },
    {
      "auxiliary_loss_clip": 0.01187479,
      "auxiliary_loss_mlp": 0.01026727,
      "balance_loss_clip": 1.05373144,
      "balance_loss_mlp": 1.01942897,
      "epoch": 0.5004509108398966,
      "flos": 27672762896640.0,
      "grad_norm": 1.6805356846806552,
      "language_loss": 0.75627899,
      "learning_rate": 2.0942206824874845e-06,
      "loss": 0.77842104,
      "num_input_tokens_seen": 89735885,
      "step": 4162,
      "time_per_iteration": 2.640352487564087
    },
    {
      "auxiliary_loss_clip": 0.01240632,
      "auxiliary_loss_mlp": 0.01027818,
      "balance_loss_clip": 1.05512667,
      "balance_loss_mlp": 1.02004242,
      "epoch": 0.5005711537305357,
      "flos": 14976186796800.0,
      "grad_norm": 2.1457899289953666,
      "language_loss": 0.79001355,
      "learning_rate": 2.093442568612105e-06,
      "loss": 0.81269801,
      "num_input_tokens_seen": 89753690,
      "step": 4163,
      "time_per_iteration": 2.6331303119659424
    },
    {
      "auxiliary_loss_clip": 0.01187753,
      "auxiliary_loss_mlp": 0.01028757,
      "balance_loss_clip": 1.05287504,
      "balance_loss_mlp": 1.02149153,
      "epoch": 0.5006913966211748,
      "flos": 26503259978880.0,
      "grad_norm": 1.587308737945686,
      "language_loss": 0.85126543,
      "learning_rate": 2.0926644405615613e-06,
      "loss": 0.87343055,
      "num_input_tokens_seen": 89774590,
      "step": 4164,
      "time_per_iteration": 2.707958936691284
    },
    {
      "auxiliary_loss_clip": 0.01329532,
      "auxiliary_loss_mlp": 0.01024689,
      "balance_loss_clip": 1.04966211,
      "balance_loss_mlp": 1.01722646,
      "epoch": 0.5008116395118138,
      "flos": 20449295971200.0,
      "grad_norm": 9.649132619440723,
      "language_loss": 0.80976325,
      "learning_rate": 2.091886298453897e-06,
      "loss": 0.83330548,
      "num_input_tokens_seen": 89792775,
      "step": 4165,
      "time_per_iteration": 2.7181267738342285
    },
    {
      "auxiliary_loss_clip": 0.01233665,
      "auxiliary_loss_mlp": 0.01031561,
      "balance_loss_clip": 1.05107296,
      "balance_loss_mlp": 1.02436352,
      "epoch": 0.500931882402453,
      "flos": 21579871524480.0,
      "grad_norm": 1.912825706434224,
      "language_loss": 0.72579354,
      "learning_rate": 2.091108142407153e-06,
      "loss": 0.74844581,
      "num_input_tokens_seen": 89811515,
      "step": 4166,
      "time_per_iteration": 3.884100914001465
    },
    {
      "auxiliary_loss_clip": 0.01197489,
      "auxiliary_loss_mlp": 0.01008342,
      "balance_loss_clip": 1.02698922,
      "balance_loss_mlp": 1.00731647,
      "epoch": 0.5010521252930921,
      "flos": 57785011925760.0,
      "grad_norm": 0.8401286919300145,
      "language_loss": 0.62291181,
      "learning_rate": 2.090329972539377e-06,
      "loss": 0.64497012,
      "num_input_tokens_seen": 89870080,
      "step": 4167,
      "time_per_iteration": 3.32360577583313
    },
    {
      "auxiliary_loss_clip": 0.01521977,
      "auxiliary_loss_mlp": 0.01031405,
      "balance_loss_clip": 1.04220855,
      "balance_loss_mlp": 1.02397788,
      "epoch": 0.5011723681837311,
      "flos": 18625500864000.0,
      "grad_norm": 3.5072877160300338,
      "language_loss": 0.68609804,
      "learning_rate": 2.089551788968616e-06,
      "loss": 0.71163189,
      "num_input_tokens_seen": 89888045,
      "step": 4168,
      "time_per_iteration": 3.78326153755188
    },
    {
      "auxiliary_loss_clip": 0.01070641,
      "auxiliary_loss_mlp": 0.01001765,
      "balance_loss_clip": 1.00962353,
      "balance_loss_mlp": 1.00081182,
      "epoch": 0.5012926110743702,
      "flos": 55883146608000.0,
      "grad_norm": 0.8410451818228101,
      "language_loss": 0.60762852,
      "learning_rate": 2.08877359181292e-06,
      "loss": 0.62835258,
      "num_input_tokens_seen": 89944610,
      "step": 4169,
      "time_per_iteration": 3.8049914836883545
    },
    {
      "auxiliary_loss_clip": 0.01342119,
      "auxiliary_loss_mlp": 0.01028836,
      "balance_loss_clip": 1.04808426,
      "balance_loss_mlp": 1.02088451,
      "epoch": 0.5014128539650093,
      "flos": 24238266117120.0,
      "grad_norm": 2.426558944242154,
      "language_loss": 0.85644233,
      "learning_rate": 2.0879953811903396e-06,
      "loss": 0.88015187,
      "num_input_tokens_seen": 89959495,
      "step": 4170,
      "time_per_iteration": 2.837725877761841
    },
    {
      "auxiliary_loss_clip": 0.01235814,
      "auxiliary_loss_mlp": 0.01028845,
      "balance_loss_clip": 1.05456996,
      "balance_loss_mlp": 1.01998758,
      "epoch": 0.5015330968556484,
      "flos": 27527468382720.0,
      "grad_norm": 1.9730809085668162,
      "language_loss": 0.78524286,
      "learning_rate": 2.08721715721893e-06,
      "loss": 0.80788946,
      "num_input_tokens_seen": 89978820,
      "step": 4171,
      "time_per_iteration": 2.6633224487304688
    },
    {
      "auxiliary_loss_clip": 0.01234404,
      "auxiliary_loss_mlp": 0.01024426,
      "balance_loss_clip": 1.05093741,
      "balance_loss_mlp": 1.01646876,
      "epoch": 0.5016533397462875,
      "flos": 23800802376960.0,
      "grad_norm": 1.905723966956785,
      "language_loss": 0.76909906,
      "learning_rate": 2.0864389200167477e-06,
      "loss": 0.79168737,
      "num_input_tokens_seen": 89997075,
      "step": 4172,
      "time_per_iteration": 2.7141213417053223
    },
    {
      "auxiliary_loss_clip": 0.01239386,
      "auxiliary_loss_mlp": 0.02565915,
      "balance_loss_clip": 1.05244291,
      "balance_loss_mlp": 1.00048578,
      "epoch": 0.5017735826369266,
      "flos": 25295009264640.0,
      "grad_norm": 2.116187847421211,
      "language_loss": 0.79133689,
      "learning_rate": 2.0856606697018504e-06,
      "loss": 0.82938993,
      "num_input_tokens_seen": 90015085,
      "step": 4173,
      "time_per_iteration": 3.571125030517578
    },
    {
      "auxiliary_loss_clip": 0.01284435,
      "auxiliary_loss_mlp": 0.01027567,
      "balance_loss_clip": 1.04952598,
      "balance_loss_mlp": 1.01871574,
      "epoch": 0.5018938255275657,
      "flos": 16873203778560.0,
      "grad_norm": 3.5182449984103172,
      "language_loss": 0.73382968,
      "learning_rate": 2.084882406392297e-06,
      "loss": 0.75694966,
      "num_input_tokens_seen": 90033045,
      "step": 4174,
      "time_per_iteration": 2.6927130222320557
    },
    {
      "auxiliary_loss_clip": 0.01240715,
      "auxiliary_loss_mlp": 0.01028782,
      "balance_loss_clip": 1.05541301,
      "balance_loss_mlp": 1.02146208,
      "epoch": 0.5020140684182047,
      "flos": 25515429073920.0,
      "grad_norm": 2.8084566104658393,
      "language_loss": 0.7156949,
      "learning_rate": 2.0841041302061496e-06,
      "loss": 0.73838985,
      "num_input_tokens_seen": 90052505,
      "step": 4175,
      "time_per_iteration": 3.7973413467407227
    },
    {
      "auxiliary_loss_clip": 0.01278505,
      "auxiliary_loss_mlp": 0.01028959,
      "balance_loss_clip": 1.04601645,
      "balance_loss_mlp": 1.02121079,
      "epoch": 0.5021343113088439,
      "flos": 23659278791040.0,
      "grad_norm": 1.9015787389038572,
      "language_loss": 0.75530612,
      "learning_rate": 2.083325841261473e-06,
      "loss": 0.77838081,
      "num_input_tokens_seen": 90071565,
      "step": 4176,
      "time_per_iteration": 2.6683974266052246
    },
    {
      "auxiliary_loss_clip": 0.01283065,
      "auxiliary_loss_mlp": 0.01027186,
      "balance_loss_clip": 1.04815793,
      "balance_loss_mlp": 1.01935947,
      "epoch": 0.502254554199483,
      "flos": 24534673148160.0,
      "grad_norm": 2.261233225763131,
      "language_loss": 0.65920979,
      "learning_rate": 2.0825475396763322e-06,
      "loss": 0.68231237,
      "num_input_tokens_seen": 90092215,
      "step": 4177,
      "time_per_iteration": 2.726410150527954
    },
    {
      "auxiliary_loss_clip": 0.01534559,
      "auxiliary_loss_mlp": 0.01027558,
      "balance_loss_clip": 1.04472065,
      "balance_loss_mlp": 1.0194819,
      "epoch": 0.502374797090122,
      "flos": 34240285607040.0,
      "grad_norm": 1.7478749717544888,
      "language_loss": 0.65740216,
      "learning_rate": 2.081769225568796e-06,
      "loss": 0.68302333,
      "num_input_tokens_seen": 90114665,
      "step": 4178,
      "time_per_iteration": 3.1406784057617188
    },
    {
      "auxiliary_loss_clip": 0.01237608,
      "auxiliary_loss_mlp": 0.01028301,
      "balance_loss_clip": 1.04973626,
      "balance_loss_mlp": 1.02050161,
      "epoch": 0.5024950399807612,
      "flos": 26031106679040.0,
      "grad_norm": 1.566373579950536,
      "language_loss": 0.7585972,
      "learning_rate": 2.0809908990569327e-06,
      "loss": 0.78125632,
      "num_input_tokens_seen": 90136445,
      "step": 4179,
      "time_per_iteration": 3.1817538738250732
    },
    {
      "auxiliary_loss_clip": 0.01284186,
      "auxiliary_loss_mlp": 0.01030164,
      "balance_loss_clip": 1.05062604,
      "balance_loss_mlp": 1.02204251,
      "epoch": 0.5026152828714002,
      "flos": 21252438120960.0,
      "grad_norm": 1.750413583068922,
      "language_loss": 0.78999078,
      "learning_rate": 2.0802125602588146e-06,
      "loss": 0.81313431,
      "num_input_tokens_seen": 90155710,
      "step": 4180,
      "time_per_iteration": 2.7144672870635986
    },
    {
      "auxiliary_loss_clip": 0.01190055,
      "auxiliary_loss_mlp": 0.01028905,
      "balance_loss_clip": 1.0556668,
      "balance_loss_mlp": 1.02059019,
      "epoch": 0.5027355257620393,
      "flos": 30956111245440.0,
      "grad_norm": 2.214441960041844,
      "language_loss": 0.66843641,
      "learning_rate": 2.0794342092925146e-06,
      "loss": 0.69062603,
      "num_input_tokens_seen": 90176845,
      "step": 4181,
      "time_per_iteration": 2.677269458770752
    },
    {
      "auxiliary_loss_clip": 0.01242279,
      "auxiliary_loss_mlp": 0.01028336,
      "balance_loss_clip": 1.05615783,
      "balance_loss_mlp": 1.02090883,
      "epoch": 0.5028557686526784,
      "flos": 24791147233920.0,
      "grad_norm": 2.287320450264016,
      "language_loss": 0.678177,
      "learning_rate": 2.078655846276108e-06,
      "loss": 0.70088309,
      "num_input_tokens_seen": 90197175,
      "step": 4182,
      "time_per_iteration": 2.7539780139923096
    },
    {
      "auxiliary_loss_clip": 0.01284764,
      "auxiliary_loss_mlp": 0.01029413,
      "balance_loss_clip": 1.05112958,
      "balance_loss_mlp": 1.02141988,
      "epoch": 0.5029760115433175,
      "flos": 22966992990720.0,
      "grad_norm": 1.8946451812654899,
      "language_loss": 0.68885142,
      "learning_rate": 2.0778774713276727e-06,
      "loss": 0.71199322,
      "num_input_tokens_seen": 90216650,
      "step": 4183,
      "time_per_iteration": 2.6224398612976074
    },
    {
      "auxiliary_loss_clip": 0.01235754,
      "auxiliary_loss_mlp": 0.01025452,
      "balance_loss_clip": 1.04925478,
      "balance_loss_mlp": 1.01732826,
      "epoch": 0.5030962544339566,
      "flos": 15305164485120.0,
      "grad_norm": 2.5403464611550466,
      "language_loss": 0.68414193,
      "learning_rate": 2.077099084565287e-06,
      "loss": 0.70675397,
      "num_input_tokens_seen": 90234055,
      "step": 4184,
      "time_per_iteration": 2.719621419906616
    },
    {
      "auxiliary_loss_clip": 0.01284044,
      "auxiliary_loss_mlp": 0.01033391,
      "balance_loss_clip": 1.04665053,
      "balance_loss_mlp": 1.02530301,
      "epoch": 0.5032164973245957,
      "flos": 24494847943680.0,
      "grad_norm": 2.3741521794356055,
      "language_loss": 0.65171528,
      "learning_rate": 2.0763206861070313e-06,
      "loss": 0.67488962,
      "num_input_tokens_seen": 90253115,
      "step": 4185,
      "time_per_iteration": 2.634406566619873
    },
    {
      "auxiliary_loss_clip": 0.01189975,
      "auxiliary_loss_mlp": 0.01031747,
      "balance_loss_clip": 1.05536473,
      "balance_loss_mlp": 1.02401042,
      "epoch": 0.5033367402152348,
      "flos": 16213452721920.0,
      "grad_norm": 2.1040253436877943,
      "language_loss": 0.75981146,
      "learning_rate": 2.0755422760709876e-06,
      "loss": 0.78202868,
      "num_input_tokens_seen": 90270515,
      "step": 4186,
      "time_per_iteration": 2.6740145683288574
    },
    {
      "auxiliary_loss_clip": 0.01284398,
      "auxiliary_loss_mlp": 0.01026977,
      "balance_loss_clip": 1.04457641,
      "balance_loss_mlp": 1.01888263,
      "epoch": 0.5034569831058738,
      "flos": 21391375927680.0,
      "grad_norm": 1.9083566981690696,
      "language_loss": 0.77168965,
      "learning_rate": 2.0747638545752417e-06,
      "loss": 0.79480338,
      "num_input_tokens_seen": 90289075,
      "step": 4187,
      "time_per_iteration": 2.7839784622192383
    },
    {
      "auxiliary_loss_clip": 0.01283621,
      "auxiliary_loss_mlp": 0.01025892,
      "balance_loss_clip": 1.05370426,
      "balance_loss_mlp": 1.01790464,
      "epoch": 0.503577225996513,
      "flos": 20558751690240.0,
      "grad_norm": 1.8970379046166974,
      "language_loss": 0.83245009,
      "learning_rate": 2.073985421737878e-06,
      "loss": 0.85554522,
      "num_input_tokens_seen": 90306385,
      "step": 4188,
      "time_per_iteration": 2.690967082977295
    },
    {
      "auxiliary_loss_clip": 0.01241776,
      "auxiliary_loss_mlp": 0.01025468,
      "balance_loss_clip": 1.05461788,
      "balance_loss_mlp": 1.01727879,
      "epoch": 0.5036974688871521,
      "flos": 27229157930880.0,
      "grad_norm": 2.305684477621911,
      "language_loss": 0.74217683,
      "learning_rate": 2.0732069776769844e-06,
      "loss": 0.76484931,
      "num_input_tokens_seen": 90323795,
      "step": 4189,
      "time_per_iteration": 2.673722743988037
    },
    {
      "auxiliary_loss_clip": 0.01191145,
      "auxiliary_loss_mlp": 0.01029879,
      "balance_loss_clip": 1.05723321,
      "balance_loss_mlp": 1.0222733,
      "epoch": 0.5038177117777911,
      "flos": 20412164286720.0,
      "grad_norm": 10.46057131021812,
      "language_loss": 0.73487639,
      "learning_rate": 2.072428522510651e-06,
      "loss": 0.75708663,
      "num_input_tokens_seen": 90340360,
      "step": 4190,
      "time_per_iteration": 2.612955331802368
    },
    {
      "auxiliary_loss_clip": 0.01331866,
      "auxiliary_loss_mlp": 0.01032406,
      "balance_loss_clip": 1.04935002,
      "balance_loss_mlp": 1.02526498,
      "epoch": 0.5039379546684303,
      "flos": 21907987286400.0,
      "grad_norm": 2.2039779530997445,
      "language_loss": 0.76280499,
      "learning_rate": 2.071650056356968e-06,
      "loss": 0.78644776,
      "num_input_tokens_seen": 90357900,
      "step": 4191,
      "time_per_iteration": 2.685763359069824
    },
    {
      "auxiliary_loss_clip": 0.01191054,
      "auxiliary_loss_mlp": 0.01027691,
      "balance_loss_clip": 1.05636883,
      "balance_loss_mlp": 1.02035999,
      "epoch": 0.5040581975590693,
      "flos": 20010718909440.0,
      "grad_norm": 1.9832538753152447,
      "language_loss": 0.80039144,
      "learning_rate": 2.070871579334028e-06,
      "loss": 0.82257885,
      "num_input_tokens_seen": 90377010,
      "step": 4192,
      "time_per_iteration": 2.5858988761901855
    },
    {
      "auxiliary_loss_clip": 0.0118812,
      "auxiliary_loss_mlp": 0.01031708,
      "balance_loss_clip": 1.0535295,
      "balance_loss_mlp": 1.02358413,
      "epoch": 0.5041784404497084,
      "flos": 20959837931520.0,
      "grad_norm": 1.6006667600027416,
      "language_loss": 0.71854258,
      "learning_rate": 2.0700930915599264e-06,
      "loss": 0.7407409,
      "num_input_tokens_seen": 90396740,
      "step": 4193,
      "time_per_iteration": 3.750080108642578
    },
    {
      "auxiliary_loss_clip": 0.01186393,
      "auxiliary_loss_mlp": 0.01021453,
      "balance_loss_clip": 1.05391502,
      "balance_loss_mlp": 1.01403236,
      "epoch": 0.5042986833403476,
      "flos": 12495082757760.0,
      "grad_norm": 3.2738584540713105,
      "language_loss": 0.78579724,
      "learning_rate": 2.0693145931527583e-06,
      "loss": 0.80787569,
      "num_input_tokens_seen": 90413220,
      "step": 4194,
      "time_per_iteration": 3.4346187114715576
    },
    {
      "auxiliary_loss_clip": 0.01279121,
      "auxiliary_loss_mlp": 0.01027659,
      "balance_loss_clip": 1.04917383,
      "balance_loss_mlp": 1.01985669,
      "epoch": 0.5044189262309866,
      "flos": 29202305788800.0,
      "grad_norm": 1.9869202257030683,
      "language_loss": 0.78253293,
      "learning_rate": 2.068536084230622e-06,
      "loss": 0.80560076,
      "num_input_tokens_seen": 90435085,
      "step": 4195,
      "time_per_iteration": 2.7597763538360596
    },
    {
      "auxiliary_loss_clip": 0.01239354,
      "auxiliary_loss_mlp": 0.01028235,
      "balance_loss_clip": 1.05551744,
      "balance_loss_mlp": 1.02031386,
      "epoch": 0.5045391691216257,
      "flos": 23873198238720.0,
      "grad_norm": 2.537407436432198,
      "language_loss": 0.889521,
      "learning_rate": 2.067757564911616e-06,
      "loss": 0.91219687,
      "num_input_tokens_seen": 90453660,
      "step": 4196,
      "time_per_iteration": 2.6904523372650146
    },
    {
      "auxiliary_loss_clip": 0.0129691,
      "auxiliary_loss_mlp": 0.02569666,
      "balance_loss_clip": 1.05273366,
      "balance_loss_mlp": 1.0005393,
      "epoch": 0.5046594120122648,
      "flos": 24644990793600.0,
      "grad_norm": 2.2254169153339043,
      "language_loss": 0.92705804,
      "learning_rate": 2.0669790353138407e-06,
      "loss": 0.96572375,
      "num_input_tokens_seen": 90472625,
      "step": 4197,
      "time_per_iteration": 2.6739180088043213
    },
    {
      "auxiliary_loss_clip": 0.01328878,
      "auxiliary_loss_mlp": 0.02570618,
      "balance_loss_clip": 1.05130911,
      "balance_loss_mlp": 1.00055337,
      "epoch": 0.5047796549029039,
      "flos": 23362835846400.0,
      "grad_norm": 2.1347324845978695,
      "language_loss": 0.72884488,
      "learning_rate": 2.0662004955553995e-06,
      "loss": 0.76783985,
      "num_input_tokens_seen": 90492325,
      "step": 4198,
      "time_per_iteration": 2.73905611038208
    },
    {
      "auxiliary_loss_clip": 0.01284692,
      "auxiliary_loss_mlp": 0.01028187,
      "balance_loss_clip": 1.0485723,
      "balance_loss_mlp": 1.02081943,
      "epoch": 0.5048998977935429,
      "flos": 17304095329920.0,
      "grad_norm": 1.9477206343537974,
      "language_loss": 0.769292,
      "learning_rate": 2.065421945754395e-06,
      "loss": 0.79242074,
      "num_input_tokens_seen": 90510055,
      "step": 4199,
      "time_per_iteration": 3.543562412261963
    },
    {
      "auxiliary_loss_clip": 0.01389835,
      "auxiliary_loss_mlp": 0.01037343,
      "balance_loss_clip": 1.05311096,
      "balance_loss_mlp": 1.03017211,
      "epoch": 0.505020140684182,
      "flos": 34856979235200.0,
      "grad_norm": 1.6791075064865157,
      "language_loss": 0.7816667,
      "learning_rate": 2.0646433860289344e-06,
      "loss": 0.80593848,
      "num_input_tokens_seen": 90528980,
      "step": 4200,
      "time_per_iteration": 2.8779854774475098
    },
    {
      "auxiliary_loss_clip": 0.0124322,
      "auxiliary_loss_mlp": 0.02573353,
      "balance_loss_clip": 1.05360389,
      "balance_loss_mlp": 1.00055194,
      "epoch": 0.5051403835748212,
      "flos": 24863974058880.0,
      "grad_norm": 2.8214318576296606,
      "language_loss": 0.82905763,
      "learning_rate": 2.0638648164971233e-06,
      "loss": 0.86722338,
      "num_input_tokens_seen": 90547445,
      "step": 4201,
      "time_per_iteration": 3.58199405670166
    },
    {
      "auxiliary_loss_clip": 0.0128707,
      "auxiliary_loss_mlp": 0.01023906,
      "balance_loss_clip": 1.05251765,
      "balance_loss_mlp": 1.0162437,
      "epoch": 0.5052606264654602,
      "flos": 20959694277120.0,
      "grad_norm": 2.3428560908845606,
      "language_loss": 0.88343406,
      "learning_rate": 2.06308623727707e-06,
      "loss": 0.90654373,
      "num_input_tokens_seen": 90567545,
      "step": 4202,
      "time_per_iteration": 2.663426399230957
    },
    {
      "auxiliary_loss_clip": 0.01235687,
      "auxiliary_loss_mlp": 0.01022011,
      "balance_loss_clip": 1.053424,
      "balance_loss_mlp": 1.0139879,
      "epoch": 0.5053808693560993,
      "flos": 19642382893440.0,
      "grad_norm": 2.5719106333400243,
      "language_loss": 0.76371419,
      "learning_rate": 2.0623076484868846e-06,
      "loss": 0.78629118,
      "num_input_tokens_seen": 90585000,
      "step": 4203,
      "time_per_iteration": 2.6759989261627197
    },
    {
      "auxiliary_loss_clip": 0.0117674,
      "auxiliary_loss_mlp": 0.01006927,
      "balance_loss_clip": 1.01415157,
      "balance_loss_mlp": 1.00602663,
      "epoch": 0.5055011122467384,
      "flos": 67504915019520.0,
      "grad_norm": 0.8289232627244497,
      "language_loss": 0.60675198,
      "learning_rate": 2.061529050244679e-06,
      "loss": 0.62858862,
      "num_input_tokens_seen": 90644745,
      "step": 4204,
      "time_per_iteration": 3.156445026397705
    },
    {
      "auxiliary_loss_clip": 0.0124618,
      "auxiliary_loss_mlp": 0.01025355,
      "balance_loss_clip": 1.05105078,
      "balance_loss_mlp": 1.01817858,
      "epoch": 0.5056213551373775,
      "flos": 16872952383360.0,
      "grad_norm": 5.989393459890356,
      "language_loss": 0.74401772,
      "learning_rate": 2.060750442668565e-06,
      "loss": 0.76673305,
      "num_input_tokens_seen": 90662500,
      "step": 4205,
      "time_per_iteration": 2.682685136795044
    },
    {
      "auxiliary_loss_clip": 0.01239139,
      "auxiliary_loss_mlp": 0.01025416,
      "balance_loss_clip": 1.05547762,
      "balance_loss_mlp": 1.01828766,
      "epoch": 0.5057415980280165,
      "flos": 15334179696000.0,
      "grad_norm": 2.9219108892922705,
      "language_loss": 0.63892376,
      "learning_rate": 2.059971825876657e-06,
      "loss": 0.66156936,
      "num_input_tokens_seen": 90677010,
      "step": 4206,
      "time_per_iteration": 2.573573350906372
    },
    {
      "auxiliary_loss_clip": 0.01239888,
      "auxiliary_loss_mlp": 0.01030922,
      "balance_loss_clip": 1.05309677,
      "balance_loss_mlp": 1.02347708,
      "epoch": 0.5058618409186557,
      "flos": 19025976574080.0,
      "grad_norm": 2.7387074534438556,
      "language_loss": 0.76459318,
      "learning_rate": 2.0591931999870713e-06,
      "loss": 0.7873013,
      "num_input_tokens_seen": 90695935,
      "step": 4207,
      "time_per_iteration": 2.6271414756774902
    },
    {
      "auxiliary_loss_clip": 0.01122366,
      "auxiliary_loss_mlp": 0.01001601,
      "balance_loss_clip": 1.01337314,
      "balance_loss_mlp": 1.00068903,
      "epoch": 0.5059820838092948,
      "flos": 63453114080640.0,
      "grad_norm": 0.8284379637430926,
      "language_loss": 0.57508707,
      "learning_rate": 2.0584145651179234e-06,
      "loss": 0.59632683,
      "num_input_tokens_seen": 90751645,
      "step": 4208,
      "time_per_iteration": 3.200458288192749
    },
    {
      "auxiliary_loss_clip": 0.01282613,
      "auxiliary_loss_mlp": 0.02569526,
      "balance_loss_clip": 1.05268574,
      "balance_loss_mlp": 1.00052142,
      "epoch": 0.5061023266999338,
      "flos": 15441803821440.0,
      "grad_norm": 3.063184148546686,
      "language_loss": 0.80095911,
      "learning_rate": 2.0576359213873327e-06,
      "loss": 0.83948046,
      "num_input_tokens_seen": 90766795,
      "step": 4209,
      "time_per_iteration": 2.6157121658325195
    },
    {
      "auxiliary_loss_clip": 0.01295187,
      "auxiliary_loss_mlp": 0.01032884,
      "balance_loss_clip": 1.05045652,
      "balance_loss_mlp": 1.02488542,
      "epoch": 0.506222569590573,
      "flos": 22451063990400.0,
      "grad_norm": 5.444401714201307,
      "language_loss": 0.70604479,
      "learning_rate": 2.056857268913419e-06,
      "loss": 0.72932553,
      "num_input_tokens_seen": 90786845,
      "step": 4210,
      "time_per_iteration": 2.6921019554138184
    },
    {
      "auxiliary_loss_clip": 0.012376,
      "auxiliary_loss_mlp": 0.01034598,
      "balance_loss_clip": 1.05429411,
      "balance_loss_mlp": 1.02671862,
      "epoch": 0.506342812481212,
      "flos": 17558665994880.0,
      "grad_norm": 2.1400263757346654,
      "language_loss": 0.8387205,
      "learning_rate": 2.056078607814303e-06,
      "loss": 0.86144245,
      "num_input_tokens_seen": 90802630,
      "step": 4211,
      "time_per_iteration": 2.5700321197509766
    },
    {
      "auxiliary_loss_clip": 0.01233984,
      "auxiliary_loss_mlp": 0.01024565,
      "balance_loss_clip": 1.05309415,
      "balance_loss_mlp": 1.01684058,
      "epoch": 0.5064630553718511,
      "flos": 23402050519680.0,
      "grad_norm": 1.9596587596091506,
      "language_loss": 0.78643012,
      "learning_rate": 2.055299938208106e-06,
      "loss": 0.80901563,
      "num_input_tokens_seen": 90823620,
      "step": 4212,
      "time_per_iteration": 2.6806366443634033
    },
    {
      "auxiliary_loss_clip": 0.0124339,
      "auxiliary_loss_mlp": 0.01031669,
      "balance_loss_clip": 1.05478168,
      "balance_loss_mlp": 1.02394402,
      "epoch": 0.5065832982624903,
      "flos": 23987035416960.0,
      "grad_norm": 1.728046828147528,
      "language_loss": 0.8603934,
      "learning_rate": 2.0545212602129526e-06,
      "loss": 0.88314402,
      "num_input_tokens_seen": 90843475,
      "step": 4213,
      "time_per_iteration": 2.642712354660034
    },
    {
      "auxiliary_loss_clip": 0.01282021,
      "auxiliary_loss_mlp": 0.01026563,
      "balance_loss_clip": 1.04925275,
      "balance_loss_mlp": 1.01860607,
      "epoch": 0.5067035411531293,
      "flos": 21503058289920.0,
      "grad_norm": 2.1247600311230426,
      "language_loss": 0.66554207,
      "learning_rate": 2.0537425739469673e-06,
      "loss": 0.6886279,
      "num_input_tokens_seen": 90862410,
      "step": 4214,
      "time_per_iteration": 2.7188549041748047
    },
    {
      "auxiliary_loss_clip": 0.01127196,
      "auxiliary_loss_mlp": 0.01005,
      "balance_loss_clip": 1.01230872,
      "balance_loss_mlp": 1.00401103,
      "epoch": 0.5068237840437684,
      "flos": 65934397687680.0,
      "grad_norm": 0.8431149404073137,
      "language_loss": 0.59403682,
      "learning_rate": 2.052963879528276e-06,
      "loss": 0.61535877,
      "num_input_tokens_seen": 90922280,
      "step": 4215,
      "time_per_iteration": 3.220952033996582
    },
    {
      "auxiliary_loss_clip": 0.0123939,
      "auxiliary_loss_mlp": 0.01031699,
      "balance_loss_clip": 1.05442142,
      "balance_loss_mlp": 1.02352738,
      "epoch": 0.5069440269344075,
      "flos": 27264206626560.0,
      "grad_norm": 2.0269640873184636,
      "language_loss": 0.76393896,
      "learning_rate": 2.052185177075007e-06,
      "loss": 0.78664988,
      "num_input_tokens_seen": 90941850,
      "step": 4216,
      "time_per_iteration": 2.6739485263824463
    },
    {
      "auxiliary_loss_clip": 0.01240241,
      "auxiliary_loss_mlp": 0.01030818,
      "balance_loss_clip": 1.05228543,
      "balance_loss_mlp": 1.02285504,
      "epoch": 0.5070642698250466,
      "flos": 23366319465600.0,
      "grad_norm": 1.6238099056653537,
      "language_loss": 0.82751811,
      "learning_rate": 2.051406466705288e-06,
      "loss": 0.85022867,
      "num_input_tokens_seen": 90961390,
      "step": 4217,
      "time_per_iteration": 2.597687244415283
    },
    {
      "auxiliary_loss_clip": 0.01187301,
      "auxiliary_loss_mlp": 0.01029807,
      "balance_loss_clip": 1.05295885,
      "balance_loss_mlp": 1.02220726,
      "epoch": 0.5071845127156857,
      "flos": 20340127560960.0,
      "grad_norm": 1.8460121900165614,
      "language_loss": 0.81105375,
      "learning_rate": 2.0506277485372486e-06,
      "loss": 0.83322483,
      "num_input_tokens_seen": 90980215,
      "step": 4218,
      "time_per_iteration": 2.6549878120422363
    },
    {
      "auxiliary_loss_clip": 0.01236668,
      "auxiliary_loss_mlp": 0.01030534,
      "balance_loss_clip": 1.05251944,
      "balance_loss_mlp": 1.02281499,
      "epoch": 0.5073047556063248,
      "flos": 12092955022080.0,
      "grad_norm": 5.878785431905208,
      "language_loss": 0.67412138,
      "learning_rate": 2.04984902268902e-06,
      "loss": 0.69679344,
      "num_input_tokens_seen": 90997415,
      "step": 4219,
      "time_per_iteration": 3.5617048740386963
    },
    {
      "auxiliary_loss_clip": 0.01245154,
      "auxiliary_loss_mlp": 0.01027959,
      "balance_loss_clip": 1.05242956,
      "balance_loss_mlp": 1.01924467,
      "epoch": 0.5074249984969639,
      "flos": 19682854542720.0,
      "grad_norm": 5.971070889515067,
      "language_loss": 0.75483704,
      "learning_rate": 2.0490702892787345e-06,
      "loss": 0.77756816,
      "num_input_tokens_seen": 91016475,
      "step": 4220,
      "time_per_iteration": 3.5786798000335693
    },
    {
      "auxiliary_loss_clip": 0.01231295,
      "auxiliary_loss_mlp": 0.0102868,
      "balance_loss_clip": 1.05019808,
      "balance_loss_mlp": 1.02128863,
      "epoch": 0.5075452413876029,
      "flos": 28765703975040.0,
      "grad_norm": 3.7536291219538898,
      "language_loss": 0.62510169,
      "learning_rate": 2.0482915484245246e-06,
      "loss": 0.6477015,
      "num_input_tokens_seen": 91038095,
      "step": 4221,
      "time_per_iteration": 2.7080938816070557
    },
    {
      "auxiliary_loss_clip": 0.01382383,
      "auxiliary_loss_mlp": 0.01027722,
      "balance_loss_clip": 1.04826188,
      "balance_loss_mlp": 1.01942551,
      "epoch": 0.5076654842782421,
      "flos": 20339445202560.0,
      "grad_norm": 3.3378579871423626,
      "language_loss": 0.84123242,
      "learning_rate": 2.047512800244526e-06,
      "loss": 0.86533344,
      "num_input_tokens_seen": 91053360,
      "step": 4222,
      "time_per_iteration": 2.771091938018799
    },
    {
      "auxiliary_loss_clip": 0.01237817,
      "auxiliary_loss_mlp": 0.01033165,
      "balance_loss_clip": 1.05505002,
      "balance_loss_mlp": 1.02481461,
      "epoch": 0.5077857271688812,
      "flos": 26359653404160.0,
      "grad_norm": 1.9705214048520783,
      "language_loss": 0.79063851,
      "learning_rate": 2.046734044856873e-06,
      "loss": 0.81334829,
      "num_input_tokens_seen": 91072770,
      "step": 4223,
      "time_per_iteration": 2.662229537963867
    },
    {
      "auxiliary_loss_clip": 0.01236442,
      "auxiliary_loss_mlp": 0.01028252,
      "balance_loss_clip": 1.05267775,
      "balance_loss_mlp": 1.02013421,
      "epoch": 0.5079059700595202,
      "flos": 21798962530560.0,
      "grad_norm": 1.9023128432605896,
      "language_loss": 0.8152343,
      "learning_rate": 2.045955282379702e-06,
      "loss": 0.83788121,
      "num_input_tokens_seen": 91091430,
      "step": 4224,
      "time_per_iteration": 2.6408963203430176
    },
    {
      "auxiliary_loss_clip": 0.01236168,
      "auxiliary_loss_mlp": 0.01029535,
      "balance_loss_clip": 1.05070519,
      "balance_loss_mlp": 1.02183759,
      "epoch": 0.5080262129501594,
      "flos": 13187943175680.0,
      "grad_norm": 2.9954433392108704,
      "language_loss": 0.76118684,
      "learning_rate": 2.045176512931152e-06,
      "loss": 0.78384382,
      "num_input_tokens_seen": 91106060,
      "step": 4225,
      "time_per_iteration": 3.531169891357422
    },
    {
      "auxiliary_loss_clip": 0.01239522,
      "auxiliary_loss_mlp": 0.01032781,
      "balance_loss_clip": 1.04839158,
      "balance_loss_mlp": 1.02516985,
      "epoch": 0.5081464558407984,
      "flos": 25301473712640.0,
      "grad_norm": 1.9848145498258551,
      "language_loss": 0.76051116,
      "learning_rate": 2.0443977366293604e-06,
      "loss": 0.78323424,
      "num_input_tokens_seen": 91124100,
      "step": 4226,
      "time_per_iteration": 2.728571653366089
    },
    {
      "auxiliary_loss_clip": 0.01440728,
      "auxiliary_loss_mlp": 0.01032557,
      "balance_loss_clip": 1.04666281,
      "balance_loss_mlp": 1.02396834,
      "epoch": 0.5082666987314375,
      "flos": 30951226995840.0,
      "grad_norm": 1.6389621680992141,
      "language_loss": 0.76954323,
      "learning_rate": 2.043618953592468e-06,
      "loss": 0.79427612,
      "num_input_tokens_seen": 91146555,
      "step": 4227,
      "time_per_iteration": 3.812769651412964
    },
    {
      "auxiliary_loss_clip": 0.01287987,
      "auxiliary_loss_mlp": 0.01027318,
      "balance_loss_clip": 1.05336356,
      "balance_loss_mlp": 1.01934886,
      "epoch": 0.5083869416220766,
      "flos": 19682495406720.0,
      "grad_norm": 2.17422309593626,
      "language_loss": 0.81266522,
      "learning_rate": 2.0428401639386144e-06,
      "loss": 0.83581829,
      "num_input_tokens_seen": 91167120,
      "step": 4228,
      "time_per_iteration": 2.668104887008667
    },
    {
      "auxiliary_loss_clip": 0.01170093,
      "auxiliary_loss_mlp": 0.0100215,
      "balance_loss_clip": 1.00923491,
      "balance_loss_mlp": 1.00131571,
      "epoch": 0.5085071845127157,
      "flos": 71817535589760.0,
      "grad_norm": 0.8215744846731133,
      "language_loss": 0.58070374,
      "learning_rate": 2.042061367785943e-06,
      "loss": 0.60242617,
      "num_input_tokens_seen": 91220260,
      "step": 4229,
      "time_per_iteration": 3.2362186908721924
    },
    {
      "auxiliary_loss_clip": 0.01335754,
      "auxiliary_loss_mlp": 0.01028367,
      "balance_loss_clip": 1.04765582,
      "balance_loss_mlp": 1.0200398,
      "epoch": 0.5086274274033548,
      "flos": 35951608252800.0,
      "grad_norm": 3.0495571992711183,
      "language_loss": 0.7539506,
      "learning_rate": 2.041282565252594e-06,
      "loss": 0.77759176,
      "num_input_tokens_seen": 91240425,
      "step": 4230,
      "time_per_iteration": 2.823314666748047
    },
    {
      "auxiliary_loss_clip": 0.01340157,
      "auxiliary_loss_mlp": 0.01027805,
      "balance_loss_clip": 1.04914904,
      "balance_loss_mlp": 1.01994884,
      "epoch": 0.5087476702939938,
      "flos": 23513732881920.0,
      "grad_norm": 1.7369516380144414,
      "language_loss": 0.77350593,
      "learning_rate": 2.040503756456714e-06,
      "loss": 0.79718554,
      "num_input_tokens_seen": 91259635,
      "step": 4231,
      "time_per_iteration": 2.7437095642089844
    },
    {
      "auxiliary_loss_clip": 0.01234241,
      "auxiliary_loss_mlp": 0.01028323,
      "balance_loss_clip": 1.05049086,
      "balance_loss_mlp": 1.02046096,
      "epoch": 0.508867913184633,
      "flos": 15122091841920.0,
      "grad_norm": 2.0555860998315936,
      "language_loss": 0.78925979,
      "learning_rate": 2.0397249415164456e-06,
      "loss": 0.81188536,
      "num_input_tokens_seen": 91276990,
      "step": 4232,
      "time_per_iteration": 2.616964340209961
    },
    {
      "auxiliary_loss_clip": 0.01283965,
      "auxiliary_loss_mlp": 0.01034695,
      "balance_loss_clip": 1.04798937,
      "balance_loss_mlp": 1.02617717,
      "epoch": 0.508988156075272,
      "flos": 25885309374720.0,
      "grad_norm": 2.508623787729711,
      "language_loss": 0.80017138,
      "learning_rate": 2.0389461205499354e-06,
      "loss": 0.82335794,
      "num_input_tokens_seen": 91296125,
      "step": 4233,
      "time_per_iteration": 2.7122769355773926
    },
    {
      "auxiliary_loss_clip": 0.01334185,
      "auxiliary_loss_mlp": 0.01025387,
      "balance_loss_clip": 1.04744315,
      "balance_loss_mlp": 1.01763213,
      "epoch": 0.5091083989659111,
      "flos": 13844857057920.0,
      "grad_norm": 1.806077978939886,
      "language_loss": 0.73307949,
      "learning_rate": 2.03816729367533e-06,
      "loss": 0.75667524,
      "num_input_tokens_seen": 91314280,
      "step": 4234,
      "time_per_iteration": 2.6615328788757324
    },
    {
      "auxiliary_loss_clip": 0.01299333,
      "auxiliary_loss_mlp": 0.01033273,
      "balance_loss_clip": 1.05762625,
      "balance_loss_mlp": 1.02523279,
      "epoch": 0.5092286418565503,
      "flos": 21104881050240.0,
      "grad_norm": 2.116606876631003,
      "language_loss": 0.71598321,
      "learning_rate": 2.0373884610107765e-06,
      "loss": 0.73930925,
      "num_input_tokens_seen": 91334595,
      "step": 4235,
      "time_per_iteration": 2.7179880142211914
    },
    {
      "auxiliary_loss_clip": 0.01240771,
      "auxiliary_loss_mlp": 0.01026968,
      "balance_loss_clip": 1.05146158,
      "balance_loss_mlp": 1.01969075,
      "epoch": 0.5093488847471893,
      "flos": 18621298972800.0,
      "grad_norm": 2.3324043164801593,
      "language_loss": 0.69568568,
      "learning_rate": 2.0366096226744225e-06,
      "loss": 0.71836311,
      "num_input_tokens_seen": 91349790,
      "step": 4236,
      "time_per_iteration": 2.640141725540161
    },
    {
      "auxiliary_loss_clip": 0.0123305,
      "auxiliary_loss_mlp": 0.01032805,
      "balance_loss_clip": 1.05184615,
      "balance_loss_mlp": 1.02496409,
      "epoch": 0.5094691276378284,
      "flos": 23803783205760.0,
      "grad_norm": 1.6931631147167616,
      "language_loss": 0.77264929,
      "learning_rate": 2.035830778784418e-06,
      "loss": 0.79530787,
      "num_input_tokens_seen": 91370465,
      "step": 4237,
      "time_per_iteration": 2.663135290145874
    },
    {
      "auxiliary_loss_clip": 0.01194743,
      "auxiliary_loss_mlp": 0.01034952,
      "balance_loss_clip": 1.05619037,
      "balance_loss_mlp": 1.02655387,
      "epoch": 0.5095893705284675,
      "flos": 17420410546560.0,
      "grad_norm": 1.9718020760316641,
      "language_loss": 0.80211222,
      "learning_rate": 2.0350519294589134e-06,
      "loss": 0.82440913,
      "num_input_tokens_seen": 91388505,
      "step": 4238,
      "time_per_iteration": 2.6724162101745605
    },
    {
      "auxiliary_loss_clip": 0.0138154,
      "auxiliary_loss_mlp": 0.01026812,
      "balance_loss_clip": 1.04576206,
      "balance_loss_mlp": 1.01849115,
      "epoch": 0.5097096134191066,
      "flos": 25849362839040.0,
      "grad_norm": 1.7910671899559647,
      "language_loss": 0.83052605,
      "learning_rate": 2.0342730748160588e-06,
      "loss": 0.85460961,
      "num_input_tokens_seen": 91408970,
      "step": 4239,
      "time_per_iteration": 2.742356061935425
    },
    {
      "auxiliary_loss_clip": 0.01284879,
      "auxiliary_loss_mlp": 0.01023936,
      "balance_loss_clip": 1.04792595,
      "balance_loss_mlp": 1.01578164,
      "epoch": 0.5098298563097456,
      "flos": 27745122844800.0,
      "grad_norm": 2.0385225423691273,
      "language_loss": 0.70159161,
      "learning_rate": 2.033494214974006e-06,
      "loss": 0.72467983,
      "num_input_tokens_seen": 91430115,
      "step": 4240,
      "time_per_iteration": 2.803779363632202
    },
    {
      "auxiliary_loss_clip": 0.01277811,
      "auxiliary_loss_mlp": 0.0102678,
      "balance_loss_clip": 1.05006266,
      "balance_loss_mlp": 1.01934683,
      "epoch": 0.5099500992003848,
      "flos": 21358913011200.0,
      "grad_norm": 1.8080347863697357,
      "language_loss": 0.83853912,
      "learning_rate": 2.0327153500509067e-06,
      "loss": 0.86158502,
      "num_input_tokens_seen": 91449140,
      "step": 4241,
      "time_per_iteration": 2.672940969467163
    },
    {
      "auxiliary_loss_clip": 0.01289795,
      "auxiliary_loss_mlp": 0.01027703,
      "balance_loss_clip": 1.05437326,
      "balance_loss_mlp": 1.01948929,
      "epoch": 0.5100703420910239,
      "flos": 19865999013120.0,
      "grad_norm": 3.006609471102507,
      "language_loss": 0.85113436,
      "learning_rate": 2.031936480164916e-06,
      "loss": 0.8743093,
      "num_input_tokens_seen": 91466880,
      "step": 4242,
      "time_per_iteration": 2.657680034637451
    },
    {
      "auxiliary_loss_clip": 0.01282713,
      "auxiliary_loss_mlp": 0.01030451,
      "balance_loss_clip": 1.05206311,
      "balance_loss_mlp": 1.02287471,
      "epoch": 0.5101905849816629,
      "flos": 24648797635200.0,
      "grad_norm": 2.8059184832033415,
      "language_loss": 0.79845887,
      "learning_rate": 2.0311576054341857e-06,
      "loss": 0.82159048,
      "num_input_tokens_seen": 91487495,
      "step": 4243,
      "time_per_iteration": 2.688277006149292
    },
    {
      "auxiliary_loss_clip": 0.0119046,
      "auxiliary_loss_mlp": 0.01027799,
      "balance_loss_clip": 1.0563972,
      "balance_loss_mlp": 1.01999664,
      "epoch": 0.5103108278723021,
      "flos": 22930076787840.0,
      "grad_norm": 1.6153995750640837,
      "language_loss": 0.62437236,
      "learning_rate": 2.0303787259768715e-06,
      "loss": 0.64655495,
      "num_input_tokens_seen": 91508395,
      "step": 4244,
      "time_per_iteration": 2.6395647525787354
    },
    {
      "auxiliary_loss_clip": 0.01289369,
      "auxiliary_loss_mlp": 0.01030685,
      "balance_loss_clip": 1.05465412,
      "balance_loss_mlp": 1.02262032,
      "epoch": 0.5104310707629411,
      "flos": 21506613736320.0,
      "grad_norm": 2.516874305093746,
      "language_loss": 0.68901467,
      "learning_rate": 2.0295998419111294e-06,
      "loss": 0.71221519,
      "num_input_tokens_seen": 91525685,
      "step": 4245,
      "time_per_iteration": 3.658639907836914
    },
    {
      "auxiliary_loss_clip": 0.01439348,
      "auxiliary_loss_mlp": 0.01027257,
      "balance_loss_clip": 1.0440495,
      "balance_loss_mlp": 1.01965714,
      "epoch": 0.5105513136535802,
      "flos": 14903180403840.0,
      "grad_norm": 4.458959566805954,
      "language_loss": 0.74135137,
      "learning_rate": 2.028820953355115e-06,
      "loss": 0.76601738,
      "num_input_tokens_seen": 91543785,
      "step": 4246,
      "time_per_iteration": 2.7507364749908447
    },
    {
      "auxiliary_loss_clip": 0.01289544,
      "auxiliary_loss_mlp": 0.0102835,
      "balance_loss_clip": 1.04961109,
      "balance_loss_mlp": 1.01996386,
      "epoch": 0.5106715565442194,
      "flos": 22602212421120.0,
      "grad_norm": 2.0004514258679467,
      "language_loss": 0.78386688,
      "learning_rate": 2.0280420604269834e-06,
      "loss": 0.80704588,
      "num_input_tokens_seen": 91563325,
      "step": 4247,
      "time_per_iteration": 3.659754514694214
    },
    {
      "auxiliary_loss_clip": 0.0111878,
      "auxiliary_loss_mlp": 0.00999592,
      "balance_loss_clip": 1.00897229,
      "balance_loss_mlp": 0.99862051,
      "epoch": 0.5107917994348584,
      "flos": 71027645558400.0,
      "grad_norm": 0.7114740193597943,
      "language_loss": 0.58912587,
      "learning_rate": 2.027263163244895e-06,
      "loss": 0.6103096,
      "num_input_tokens_seen": 91632450,
      "step": 4248,
      "time_per_iteration": 3.3793118000030518
    },
    {
      "auxiliary_loss_clip": 0.01237731,
      "auxiliary_loss_mlp": 0.01026249,
      "balance_loss_clip": 1.05460024,
      "balance_loss_mlp": 1.01870358,
      "epoch": 0.5109120423254975,
      "flos": 24827416992000.0,
      "grad_norm": 1.5920930467843653,
      "language_loss": 0.74736702,
      "learning_rate": 2.026484261927005e-06,
      "loss": 0.77000684,
      "num_input_tokens_seen": 91651945,
      "step": 4249,
      "time_per_iteration": 2.647348165512085
    },
    {
      "auxiliary_loss_clip": 0.01245208,
      "auxiliary_loss_mlp": 0.01031251,
      "balance_loss_clip": 1.05622792,
      "balance_loss_mlp": 1.02309704,
      "epoch": 0.5110322852161366,
      "flos": 21247661612160.0,
      "grad_norm": 2.781167373177434,
      "language_loss": 0.73800516,
      "learning_rate": 2.025705356591475e-06,
      "loss": 0.76076984,
      "num_input_tokens_seen": 91669635,
      "step": 4250,
      "time_per_iteration": 2.6712985038757324
    },
    {
      "auxiliary_loss_clip": 0.01224596,
      "auxiliary_loss_mlp": 0.02507124,
      "balance_loss_clip": 1.00825834,
      "balance_loss_mlp": 1.00042903,
      "epoch": 0.5111525281067757,
      "flos": 66457114358400.0,
      "grad_norm": 0.7725360168070031,
      "language_loss": 0.57937801,
      "learning_rate": 2.024926447356462e-06,
      "loss": 0.61669523,
      "num_input_tokens_seen": 91731920,
      "step": 4251,
      "time_per_iteration": 4.049286365509033
    },
    {
      "auxiliary_loss_clip": 0.01238061,
      "auxiliary_loss_mlp": 0.01032132,
      "balance_loss_clip": 1.05401576,
      "balance_loss_mlp": 1.02344179,
      "epoch": 0.5112727709974147,
      "flos": 14866731077760.0,
      "grad_norm": 14.610791524427741,
      "language_loss": 0.78675854,
      "learning_rate": 2.024147534340127e-06,
      "loss": 0.80946052,
      "num_input_tokens_seen": 91749780,
      "step": 4252,
      "time_per_iteration": 2.662172555923462
    },
    {
      "auxiliary_loss_clip": 0.01186746,
      "auxiliary_loss_mlp": 0.01032284,
      "balance_loss_clip": 1.0478071,
      "balance_loss_mlp": 1.02471161,
      "epoch": 0.5113930138880539,
      "flos": 21177600134400.0,
      "grad_norm": 1.6818946675170077,
      "language_loss": 0.80127656,
      "learning_rate": 2.02336861766063e-06,
      "loss": 0.8234669,
      "num_input_tokens_seen": 91768840,
      "step": 4253,
      "time_per_iteration": 3.579575538635254
    },
    {
      "auxiliary_loss_clip": 0.01246008,
      "auxiliary_loss_mlp": 0.01033563,
      "balance_loss_clip": 1.05392623,
      "balance_loss_mlp": 1.02576065,
      "epoch": 0.511513256778693,
      "flos": 20409111630720.0,
      "grad_norm": 2.6929512944112926,
      "language_loss": 0.79048747,
      "learning_rate": 2.0225896974361327e-06,
      "loss": 0.81328321,
      "num_input_tokens_seen": 91788945,
      "step": 4254,
      "time_per_iteration": 2.6093294620513916
    },
    {
      "auxiliary_loss_clip": 0.0122833,
      "auxiliary_loss_mlp": 0.00999602,
      "balance_loss_clip": 1.01029587,
      "balance_loss_mlp": 0.99855345,
      "epoch": 0.511633499669332,
      "flos": 69879975131520.0,
      "grad_norm": 0.8531134335463075,
      "language_loss": 0.59903169,
      "learning_rate": 2.0218107737847962e-06,
      "loss": 0.62131095,
      "num_input_tokens_seen": 91850990,
      "step": 4255,
      "time_per_iteration": 3.343179702758789
    },
    {
      "auxiliary_loss_clip": 0.01187575,
      "auxiliary_loss_mlp": 0.01029709,
      "balance_loss_clip": 1.05419636,
      "balance_loss_mlp": 1.02182281,
      "epoch": 0.5117537425599712,
      "flos": 24097855852800.0,
      "grad_norm": 1.8594719890262934,
      "language_loss": 0.74970937,
      "learning_rate": 2.0210318468247826e-06,
      "loss": 0.77188218,
      "num_input_tokens_seen": 91869960,
      "step": 4256,
      "time_per_iteration": 2.6054012775421143
    },
    {
      "auxiliary_loss_clip": 0.01282454,
      "auxiliary_loss_mlp": 0.01029309,
      "balance_loss_clip": 1.04832077,
      "balance_loss_mlp": 1.02204609,
      "epoch": 0.5118739854506102,
      "flos": 20959550622720.0,
      "grad_norm": 1.9452928956673092,
      "language_loss": 0.81803334,
      "learning_rate": 2.020252916674255e-06,
      "loss": 0.841151,
      "num_input_tokens_seen": 91889075,
      "step": 4257,
      "time_per_iteration": 2.6939377784729004
    },
    {
      "auxiliary_loss_clip": 0.01238826,
      "auxiliary_loss_mlp": 0.01026808,
      "balance_loss_clip": 1.05132389,
      "balance_loss_mlp": 1.01858854,
      "epoch": 0.5119942283412493,
      "flos": 17457326749440.0,
      "grad_norm": 1.7521781587305745,
      "language_loss": 0.80921811,
      "learning_rate": 2.019473983451375e-06,
      "loss": 0.83187449,
      "num_input_tokens_seen": 91907495,
      "step": 4258,
      "time_per_iteration": 2.6142094135284424
    },
    {
      "auxiliary_loss_clip": 0.01242092,
      "auxiliary_loss_mlp": 0.01033789,
      "balance_loss_clip": 1.04723024,
      "balance_loss_mlp": 1.02551627,
      "epoch": 0.5121144712318885,
      "flos": 21066743784960.0,
      "grad_norm": 1.823092720779081,
      "language_loss": 0.71387392,
      "learning_rate": 2.0186950472743076e-06,
      "loss": 0.73663276,
      "num_input_tokens_seen": 91927400,
      "step": 4259,
      "time_per_iteration": 2.71830415725708
    },
    {
      "auxiliary_loss_clip": 0.01190253,
      "auxiliary_loss_mlp": 0.01026817,
      "balance_loss_clip": 1.05512059,
      "balance_loss_mlp": 1.01951218,
      "epoch": 0.5122347141225275,
      "flos": 19860791541120.0,
      "grad_norm": 1.5989056658616305,
      "language_loss": 0.73877674,
      "learning_rate": 2.0179161082612162e-06,
      "loss": 0.76094741,
      "num_input_tokens_seen": 91946790,
      "step": 4260,
      "time_per_iteration": 2.6248505115509033
    },
    {
      "auxiliary_loss_clip": 0.01281789,
      "auxiliary_loss_mlp": 0.01025409,
      "balance_loss_clip": 1.04923868,
      "balance_loss_mlp": 1.01797032,
      "epoch": 0.5123549570131666,
      "flos": 22528487756160.0,
      "grad_norm": 2.064618032287243,
      "language_loss": 0.72970164,
      "learning_rate": 2.017137166530266e-06,
      "loss": 0.75277364,
      "num_input_tokens_seen": 91966325,
      "step": 4261,
      "time_per_iteration": 2.6486153602600098
    },
    {
      "auxiliary_loss_clip": 0.01293864,
      "auxiliary_loss_mlp": 0.01029053,
      "balance_loss_clip": 1.05302405,
      "balance_loss_mlp": 1.02142906,
      "epoch": 0.5124751999038056,
      "flos": 20333375804160.0,
      "grad_norm": 1.8940156636822334,
      "language_loss": 0.80092394,
      "learning_rate": 2.0163582221996213e-06,
      "loss": 0.82415318,
      "num_input_tokens_seen": 91984700,
      "step": 4262,
      "time_per_iteration": 2.7023940086364746
    },
    {
      "auxiliary_loss_clip": 0.01288048,
      "auxiliary_loss_mlp": 0.01036277,
      "balance_loss_clip": 1.05241191,
      "balance_loss_mlp": 1.02819419,
      "epoch": 0.5125954427944448,
      "flos": 39785970211200.0,
      "grad_norm": 2.341472384373308,
      "language_loss": 0.68159473,
      "learning_rate": 2.015579275387446e-06,
      "loss": 0.70483798,
      "num_input_tokens_seen": 92010020,
      "step": 4263,
      "time_per_iteration": 2.7968685626983643
    },
    {
      "auxiliary_loss_clip": 0.01283087,
      "auxiliary_loss_mlp": 0.01032956,
      "balance_loss_clip": 1.05216002,
      "balance_loss_mlp": 1.02490306,
      "epoch": 0.5127156856850839,
      "flos": 29205394358400.0,
      "grad_norm": 2.4082693937591304,
      "language_loss": 0.68681848,
      "learning_rate": 2.0148003262119085e-06,
      "loss": 0.70997894,
      "num_input_tokens_seen": 92030990,
      "step": 4264,
      "time_per_iteration": 2.723036766052246
    },
    {
      "auxiliary_loss_clip": 0.01333685,
      "auxiliary_loss_mlp": 0.01030335,
      "balance_loss_clip": 1.0490737,
      "balance_loss_mlp": 1.022259,
      "epoch": 0.5128359285757229,
      "flos": 13553693412480.0,
      "grad_norm": 1.8038179350853971,
      "language_loss": 0.7661432,
      "learning_rate": 2.0140213747911728e-06,
      "loss": 0.78978336,
      "num_input_tokens_seen": 92049525,
      "step": 4265,
      "time_per_iteration": 2.647580862045288
    },
    {
      "auxiliary_loss_clip": 0.0133555,
      "auxiliary_loss_mlp": 0.0102716,
      "balance_loss_clip": 1.05336642,
      "balance_loss_mlp": 1.01861238,
      "epoch": 0.5129561714663621,
      "flos": 25192089820800.0,
      "grad_norm": 2.8848474023546635,
      "language_loss": 0.80472827,
      "learning_rate": 2.013242421243406e-06,
      "loss": 0.82835537,
      "num_input_tokens_seen": 92068430,
      "step": 4266,
      "time_per_iteration": 2.765040397644043
    },
    {
      "auxiliary_loss_clip": 0.01386744,
      "auxiliary_loss_mlp": 0.01028761,
      "balance_loss_clip": 1.05184996,
      "balance_loss_mlp": 1.02158141,
      "epoch": 0.5130764143570011,
      "flos": 18150223080960.0,
      "grad_norm": 1.7232823809930522,
      "language_loss": 0.79030257,
      "learning_rate": 2.012463465686774e-06,
      "loss": 0.8144576,
      "num_input_tokens_seen": 92088180,
      "step": 4267,
      "time_per_iteration": 2.6798245906829834
    },
    {
      "auxiliary_loss_clip": 0.01209287,
      "auxiliary_loss_mlp": 0.01001669,
      "balance_loss_clip": 1.03021502,
      "balance_loss_mlp": 1.00049484,
      "epoch": 0.5131966572476402,
      "flos": 59794896418560.0,
      "grad_norm": 1.168157537934399,
      "language_loss": 0.54784179,
      "learning_rate": 2.0116845082394446e-06,
      "loss": 0.56995142,
      "num_input_tokens_seen": 92153015,
      "step": 4268,
      "time_per_iteration": 3.3915586471557617
    },
    {
      "auxiliary_loss_clip": 0.01241829,
      "auxiliary_loss_mlp": 0.01027058,
      "balance_loss_clip": 1.05357826,
      "balance_loss_mlp": 1.01972628,
      "epoch": 0.5133169001382794,
      "flos": 18515219132160.0,
      "grad_norm": 2.381449979427126,
      "language_loss": 0.78605795,
      "learning_rate": 2.0109055490195836e-06,
      "loss": 0.80874681,
      "num_input_tokens_seen": 92171470,
      "step": 4269,
      "time_per_iteration": 2.7886970043182373
    },
    {
      "auxiliary_loss_clip": 0.01438463,
      "auxiliary_loss_mlp": 0.01027745,
      "balance_loss_clip": 1.03992963,
      "balance_loss_mlp": 1.01964426,
      "epoch": 0.5134371430289184,
      "flos": 15523537219200.0,
      "grad_norm": 2.112732016406456,
      "language_loss": 0.64699078,
      "learning_rate": 2.0101265881453605e-06,
      "loss": 0.67165285,
      "num_input_tokens_seen": 92189945,
      "step": 4270,
      "time_per_iteration": 3.7338459491729736
    },
    {
      "auxiliary_loss_clip": 0.01284309,
      "auxiliary_loss_mlp": 0.01037885,
      "balance_loss_clip": 1.05405831,
      "balance_loss_mlp": 1.03025556,
      "epoch": 0.5135573859195575,
      "flos": 21433786911360.0,
      "grad_norm": 2.2956386984582706,
      "language_loss": 0.78291684,
      "learning_rate": 2.009347625734941e-06,
      "loss": 0.80613875,
      "num_input_tokens_seen": 92209855,
      "step": 4271,
      "time_per_iteration": 2.724912166595459
    },
    {
      "auxiliary_loss_clip": 0.01193157,
      "auxiliary_loss_mlp": 0.01031826,
      "balance_loss_clip": 1.05759168,
      "balance_loss_mlp": 1.02396464,
      "epoch": 0.5136776288101966,
      "flos": 17712651600000.0,
      "grad_norm": 2.4464263475880634,
      "language_loss": 0.74854326,
      "learning_rate": 2.0085686619064954e-06,
      "loss": 0.77079308,
      "num_input_tokens_seen": 92226295,
      "step": 4272,
      "time_per_iteration": 3.581024408340454
    },
    {
      "auxiliary_loss_clip": 0.01242672,
      "auxiliary_loss_mlp": 0.01038599,
      "balance_loss_clip": 1.05399919,
      "balance_loss_mlp": 1.03059399,
      "epoch": 0.5137978717008357,
      "flos": 16581680997120.0,
      "grad_norm": 2.1526519812786646,
      "language_loss": 0.82929325,
      "learning_rate": 2.00778969677819e-06,
      "loss": 0.85210592,
      "num_input_tokens_seen": 92243330,
      "step": 4273,
      "time_per_iteration": 2.6501219272613525
    },
    {
      "auxiliary_loss_clip": 0.01284642,
      "auxiliary_loss_mlp": 0.01027495,
      "balance_loss_clip": 1.04964423,
      "balance_loss_mlp": 1.0197432,
      "epoch": 0.5139181145914747,
      "flos": 20668243322880.0,
      "grad_norm": 1.754762230018292,
      "language_loss": 0.6419667,
      "learning_rate": 2.0070107304681934e-06,
      "loss": 0.66508806,
      "num_input_tokens_seen": 92262285,
      "step": 4274,
      "time_per_iteration": 2.719332218170166
    },
    {
      "auxiliary_loss_clip": 0.01332546,
      "auxiliary_loss_mlp": 0.01025178,
      "balance_loss_clip": 1.05067003,
      "balance_loss_mlp": 1.01744092,
      "epoch": 0.5140383574821139,
      "flos": 32926996546560.0,
      "grad_norm": 1.8494326660740519,
      "language_loss": 0.78349459,
      "learning_rate": 2.006231763094675e-06,
      "loss": 0.80707186,
      "num_input_tokens_seen": 92283305,
      "step": 4275,
      "time_per_iteration": 2.8273448944091797
    },
    {
      "auxiliary_loss_clip": 0.01280575,
      "auxiliary_loss_mlp": 0.0102841,
      "balance_loss_clip": 1.05287683,
      "balance_loss_mlp": 1.02082217,
      "epoch": 0.514158600372753,
      "flos": 19537093152000.0,
      "grad_norm": 1.9713472316979852,
      "language_loss": 0.8778044,
      "learning_rate": 2.0054527947758027e-06,
      "loss": 0.90089428,
      "num_input_tokens_seen": 92302105,
      "step": 4276,
      "time_per_iteration": 2.7359025478363037
    },
    {
      "auxiliary_loss_clip": 0.01118641,
      "auxiliary_loss_mlp": 0.00998716,
      "balance_loss_clip": 1.00860357,
      "balance_loss_mlp": 0.99779224,
      "epoch": 0.514278843263392,
      "flos": 62523855279360.0,
      "grad_norm": 0.7216920381716123,
      "language_loss": 0.55852109,
      "learning_rate": 2.004673825629746e-06,
      "loss": 0.57969469,
      "num_input_tokens_seen": 92362885,
      "step": 4277,
      "time_per_iteration": 4.058689117431641
    },
    {
      "auxiliary_loss_clip": 0.01285013,
      "auxiliary_loss_mlp": 0.0102624,
      "balance_loss_clip": 1.04882884,
      "balance_loss_mlp": 1.0184319,
      "epoch": 0.5143990861540312,
      "flos": 25882328545920.0,
      "grad_norm": 1.6528576129152224,
      "language_loss": 0.72561109,
      "learning_rate": 2.0038948557746744e-06,
      "loss": 0.74872357,
      "num_input_tokens_seen": 92384740,
      "step": 4278,
      "time_per_iteration": 2.7297794818878174
    },
    {
      "auxiliary_loss_clip": 0.0123543,
      "auxiliary_loss_mlp": 0.01028896,
      "balance_loss_clip": 1.0552907,
      "balance_loss_mlp": 1.02152634,
      "epoch": 0.5145193290446702,
      "flos": 23330660238720.0,
      "grad_norm": 1.835274850110765,
      "language_loss": 0.75228453,
      "learning_rate": 2.0031158853287558e-06,
      "loss": 0.77492779,
      "num_input_tokens_seen": 92405175,
      "step": 4279,
      "time_per_iteration": 3.5825581550598145
    },
    {
      "auxiliary_loss_clip": 0.01283678,
      "auxiliary_loss_mlp": 0.01030266,
      "balance_loss_clip": 1.05291915,
      "balance_loss_mlp": 1.0222013,
      "epoch": 0.5146395719353093,
      "flos": 22856603518080.0,
      "grad_norm": 2.3332309026224904,
      "language_loss": 0.70217454,
      "learning_rate": 2.0023369144101593e-06,
      "loss": 0.72531396,
      "num_input_tokens_seen": 92423345,
      "step": 4280,
      "time_per_iteration": 2.710209608078003
    },
    {
      "auxiliary_loss_clip": 0.01278442,
      "auxiliary_loss_mlp": 0.01027735,
      "balance_loss_clip": 1.04855108,
      "balance_loss_mlp": 1.02038598,
      "epoch": 0.5147598148259485,
      "flos": 26391577616640.0,
      "grad_norm": 2.5871030695860484,
      "language_loss": 0.76766217,
      "learning_rate": 2.0015579431370555e-06,
      "loss": 0.79072392,
      "num_input_tokens_seen": 92445025,
      "step": 4281,
      "time_per_iteration": 2.698371171951294
    },
    {
      "auxiliary_loss_clip": 0.01236429,
      "auxiliary_loss_mlp": 0.01031451,
      "balance_loss_clip": 1.05368066,
      "balance_loss_mlp": 1.02429867,
      "epoch": 0.5148800577165875,
      "flos": 29965694561280.0,
      "grad_norm": 1.9258201872589449,
      "language_loss": 0.69634855,
      "learning_rate": 2.000778971627612e-06,
      "loss": 0.7190274,
      "num_input_tokens_seen": 92464490,
      "step": 4282,
      "time_per_iteration": 2.7236742973327637
    },
    {
      "auxiliary_loss_clip": 0.01282756,
      "auxiliary_loss_mlp": 0.01031456,
      "balance_loss_clip": 1.04888916,
      "balance_loss_mlp": 1.02401114,
      "epoch": 0.5150003006072266,
      "flos": 17931383470080.0,
      "grad_norm": 1.9187140600716615,
      "language_loss": 0.90327203,
      "learning_rate": 2e-06,
      "loss": 0.92641413,
      "num_input_tokens_seen": 92482085,
      "step": 4283,
      "time_per_iteration": 2.6352896690368652
    },
    {
      "auxiliary_loss_clip": 0.01189091,
      "auxiliary_loss_mlp": 0.01031602,
      "balance_loss_clip": 1.05735302,
      "balance_loss_mlp": 1.02417207,
      "epoch": 0.5151205434978657,
      "flos": 18478733892480.0,
      "grad_norm": 6.434070578374005,
      "language_loss": 0.85754848,
      "learning_rate": 1.9992210283723878e-06,
      "loss": 0.87975538,
      "num_input_tokens_seen": 92499325,
      "step": 4284,
      "time_per_iteration": 2.608607530593872
    },
    {
      "auxiliary_loss_clip": 0.01187267,
      "auxiliary_loss_mlp": 0.01031304,
      "balance_loss_clip": 1.05706251,
      "balance_loss_mlp": 1.02424085,
      "epoch": 0.5152407863885048,
      "flos": 25341263003520.0,
      "grad_norm": 1.6222002262950437,
      "language_loss": 0.79616487,
      "learning_rate": 1.9984420568629448e-06,
      "loss": 0.81835055,
      "num_input_tokens_seen": 92522090,
      "step": 4285,
      "time_per_iteration": 2.6762595176696777
    },
    {
      "auxiliary_loss_clip": 0.01238807,
      "auxiliary_loss_mlp": 0.0102745,
      "balance_loss_clip": 1.05366778,
      "balance_loss_mlp": 1.02024651,
      "epoch": 0.5153610292791438,
      "flos": 18329740277760.0,
      "grad_norm": 2.193212647535323,
      "language_loss": 0.78316987,
      "learning_rate": 1.9976630855898405e-06,
      "loss": 0.80583245,
      "num_input_tokens_seen": 92539845,
      "step": 4286,
      "time_per_iteration": 2.6561484336853027
    },
    {
      "auxiliary_loss_clip": 0.01282507,
      "auxiliary_loss_mlp": 0.01027409,
      "balance_loss_clip": 1.04712129,
      "balance_loss_mlp": 1.02014315,
      "epoch": 0.515481272169783,
      "flos": 30409945971840.0,
      "grad_norm": 2.2467997312766483,
      "language_loss": 0.75075918,
      "learning_rate": 1.9968841146712445e-06,
      "loss": 0.77385831,
      "num_input_tokens_seen": 92559460,
      "step": 4287,
      "time_per_iteration": 2.7067112922668457
    },
    {
      "auxiliary_loss_clip": 0.01431291,
      "auxiliary_loss_mlp": 0.02569499,
      "balance_loss_clip": 1.04663324,
      "balance_loss_mlp": 1.00067377,
      "epoch": 0.5156015150604221,
      "flos": 23037305863680.0,
      "grad_norm": 1.8705311528836928,
      "language_loss": 0.7142396,
      "learning_rate": 1.996105144225326e-06,
      "loss": 0.75424749,
      "num_input_tokens_seen": 92579695,
      "step": 4288,
      "time_per_iteration": 2.880500555038452
    },
    {
      "auxiliary_loss_clip": 0.01237584,
      "auxiliary_loss_mlp": 0.01031558,
      "balance_loss_clip": 1.05594945,
      "balance_loss_mlp": 1.02389336,
      "epoch": 0.5157217579510611,
      "flos": 17858556645120.0,
      "grad_norm": 2.1892422949518906,
      "language_loss": 0.79114175,
      "learning_rate": 1.995326174370254e-06,
      "loss": 0.81383324,
      "num_input_tokens_seen": 92598795,
      "step": 4289,
      "time_per_iteration": 2.61956524848938
    },
    {
      "auxiliary_loss_clip": 0.0123511,
      "auxiliary_loss_mlp": 0.0256255,
      "balance_loss_clip": 1.05257833,
      "balance_loss_mlp": 1.00063884,
      "epoch": 0.5158420008417003,
      "flos": 19171486569600.0,
      "grad_norm": 1.6906978312900784,
      "language_loss": 0.73179364,
      "learning_rate": 1.994547205224197e-06,
      "loss": 0.76977026,
      "num_input_tokens_seen": 92617700,
      "step": 4290,
      "time_per_iteration": 2.6912481784820557
    },
    {
      "auxiliary_loss_clip": 0.01280326,
      "auxiliary_loss_mlp": 0.01036663,
      "balance_loss_clip": 1.05169928,
      "balance_loss_mlp": 1.02942741,
      "epoch": 0.5159622437323393,
      "flos": 22419534827520.0,
      "grad_norm": 2.203955283101594,
      "language_loss": 0.67531121,
      "learning_rate": 1.993768236905325e-06,
      "loss": 0.69848108,
      "num_input_tokens_seen": 92638370,
      "step": 4291,
      "time_per_iteration": 2.6730446815490723
    },
    {
      "auxiliary_loss_clip": 0.01282904,
      "auxiliary_loss_mlp": 0.01028538,
      "balance_loss_clip": 1.04981017,
      "balance_loss_mlp": 1.02090859,
      "epoch": 0.5160824866229784,
      "flos": 24603010773120.0,
      "grad_norm": 2.5137559908477467,
      "language_loss": 0.65991074,
      "learning_rate": 1.992989269531807e-06,
      "loss": 0.68302518,
      "num_input_tokens_seen": 92657180,
      "step": 4292,
      "time_per_iteration": 2.733698844909668
    },
    {
      "auxiliary_loss_clip": 0.01284829,
      "auxiliary_loss_mlp": 0.01030852,
      "balance_loss_clip": 1.05051112,
      "balance_loss_mlp": 1.0226748,
      "epoch": 0.5162027295136175,
      "flos": 18002737837440.0,
      "grad_norm": 2.4138290103333766,
      "language_loss": 0.6777814,
      "learning_rate": 1.99221030322181e-06,
      "loss": 0.70093822,
      "num_input_tokens_seen": 92673985,
      "step": 4293,
      "time_per_iteration": 2.6164045333862305
    },
    {
      "auxiliary_loss_clip": 0.01289191,
      "auxiliary_loss_mlp": 0.0102258,
      "balance_loss_clip": 1.05067301,
      "balance_loss_mlp": 1.01583529,
      "epoch": 0.5163229724042566,
      "flos": 27344611221120.0,
      "grad_norm": 3.611248364540869,
      "language_loss": 0.81007981,
      "learning_rate": 1.991431338093505e-06,
      "loss": 0.83319759,
      "num_input_tokens_seen": 92696340,
      "step": 4294,
      "time_per_iteration": 2.7282557487487793
    },
    {
      "auxiliary_loss_clip": 0.01286524,
      "auxiliary_loss_mlp": 0.0103272,
      "balance_loss_clip": 1.05554175,
      "balance_loss_mlp": 1.02556181,
      "epoch": 0.5164432152948957,
      "flos": 21762764599680.0,
      "grad_norm": 1.7891481541501169,
      "language_loss": 0.79374963,
      "learning_rate": 1.9906523742650587e-06,
      "loss": 0.8169421,
      "num_input_tokens_seen": 92715200,
      "step": 4295,
      "time_per_iteration": 2.6433379650115967
    },
    {
      "auxiliary_loss_clip": 0.01186875,
      "auxiliary_loss_mlp": 0.01028921,
      "balance_loss_clip": 1.05154443,
      "balance_loss_mlp": 1.02018881,
      "epoch": 0.5165634581855347,
      "flos": 25550334115200.0,
      "grad_norm": 2.4376782443202942,
      "language_loss": 0.77664089,
      "learning_rate": 1.9898734118546397e-06,
      "loss": 0.7987988,
      "num_input_tokens_seen": 92735150,
      "step": 4296,
      "time_per_iteration": 2.7678468227386475
    },
    {
      "auxiliary_loss_clip": 0.01484065,
      "auxiliary_loss_mlp": 0.01027802,
      "balance_loss_clip": 1.04541314,
      "balance_loss_mlp": 1.02026224,
      "epoch": 0.5166837010761739,
      "flos": 19901191363200.0,
      "grad_norm": 1.5684395904048862,
      "language_loss": 0.80615342,
      "learning_rate": 1.989094450980416e-06,
      "loss": 0.83127213,
      "num_input_tokens_seen": 92755250,
      "step": 4297,
      "time_per_iteration": 4.031324148178101
    },
    {
      "auxiliary_loss_clip": 0.01231964,
      "auxiliary_loss_mlp": 0.0103011,
      "balance_loss_clip": 1.0522604,
      "balance_loss_mlp": 1.02254665,
      "epoch": 0.516803943966813,
      "flos": 26646076454400.0,
      "grad_norm": 2.1416244666748825,
      "language_loss": 0.76982868,
      "learning_rate": 1.9883154917605556e-06,
      "loss": 0.79244941,
      "num_input_tokens_seen": 92774460,
      "step": 4298,
      "time_per_iteration": 4.15547776222229
    },
    {
      "auxiliary_loss_clip": 0.01187958,
      "auxiliary_loss_mlp": 0.01029003,
      "balance_loss_clip": 1.05472851,
      "balance_loss_mlp": 1.02161813,
      "epoch": 0.516924186857452,
      "flos": 19682854542720.0,
      "grad_norm": 1.690022109053751,
      "language_loss": 0.83576149,
      "learning_rate": 1.9875365343132262e-06,
      "loss": 0.85793108,
      "num_input_tokens_seen": 92791580,
      "step": 4299,
      "time_per_iteration": 2.6159043312072754
    },
    {
      "auxiliary_loss_clip": 0.01239496,
      "auxiliary_loss_mlp": 0.02566568,
      "balance_loss_clip": 1.05692339,
      "balance_loss_mlp": 1.00054502,
      "epoch": 0.5170444297480912,
      "flos": 15956583586560.0,
      "grad_norm": 3.9756936073571087,
      "language_loss": 0.84977263,
      "learning_rate": 1.9867575787565946e-06,
      "loss": 0.88783324,
      "num_input_tokens_seen": 92806240,
      "step": 4300,
      "time_per_iteration": 2.7158455848693848
    },
    {
      "auxiliary_loss_clip": 0.01238748,
      "auxiliary_loss_mlp": 0.01033712,
      "balance_loss_clip": 1.05584979,
      "balance_loss_mlp": 1.02601659,
      "epoch": 0.5171646726387302,
      "flos": 14174157968640.0,
      "grad_norm": 1.8735579176970072,
      "language_loss": 0.85688365,
      "learning_rate": 1.9859786252088275e-06,
      "loss": 0.87960827,
      "num_input_tokens_seen": 92823420,
      "step": 4301,
      "time_per_iteration": 2.599133014678955
    },
    {
      "auxiliary_loss_clip": 0.01337262,
      "auxiliary_loss_mlp": 0.0102568,
      "balance_loss_clip": 1.0509305,
      "balance_loss_mlp": 1.01752555,
      "epoch": 0.5172849155293693,
      "flos": 23578550974080.0,
      "grad_norm": 2.4928819911120708,
      "language_loss": 0.66851121,
      "learning_rate": 1.9851996737880914e-06,
      "loss": 0.69214058,
      "num_input_tokens_seen": 92838605,
      "step": 4302,
      "time_per_iteration": 3.6310720443725586
    },
    {
      "auxiliary_loss_clip": 0.01239182,
      "auxiliary_loss_mlp": 0.01027593,
      "balance_loss_clip": 1.05212033,
      "balance_loss_mlp": 1.0196476,
      "epoch": 0.5174051584200084,
      "flos": 14283541860480.0,
      "grad_norm": 2.2591712061277085,
      "language_loss": 0.74668694,
      "learning_rate": 1.9844207246125537e-06,
      "loss": 0.7693547,
      "num_input_tokens_seen": 92855185,
      "step": 4303,
      "time_per_iteration": 2.5822744369506836
    },
    {
      "auxiliary_loss_clip": 0.01284541,
      "auxiliary_loss_mlp": 0.0102683,
      "balance_loss_clip": 1.05104566,
      "balance_loss_mlp": 1.0194627,
      "epoch": 0.5175254013106475,
      "flos": 37889384192640.0,
      "grad_norm": 1.7908900442780344,
      "language_loss": 0.68280149,
      "learning_rate": 1.983641777800379e-06,
      "loss": 0.70591521,
      "num_input_tokens_seen": 92877830,
      "step": 4304,
      "time_per_iteration": 2.820469617843628
    },
    {
      "auxiliary_loss_clip": 0.01185049,
      "auxiliary_loss_mlp": 0.01006471,
      "balance_loss_clip": 1.00879979,
      "balance_loss_mlp": 1.00553536,
      "epoch": 0.5176456442012866,
      "flos": 68549737829760.0,
      "grad_norm": 0.7442690253247953,
      "language_loss": 0.5876438,
      "learning_rate": 1.9828628334697343e-06,
      "loss": 0.609559,
      "num_input_tokens_seen": 92945040,
      "step": 4305,
      "time_per_iteration": 4.31735897064209
    },
    {
      "auxiliary_loss_clip": 0.01179371,
      "auxiliary_loss_mlp": 0.01005507,
      "balance_loss_clip": 1.00886405,
      "balance_loss_mlp": 1.00456512,
      "epoch": 0.5177658870919257,
      "flos": 64084137235200.0,
      "grad_norm": 0.7595450418083342,
      "language_loss": 0.54622197,
      "learning_rate": 1.982083891738784e-06,
      "loss": 0.56807077,
      "num_input_tokens_seen": 93005910,
      "step": 4306,
      "time_per_iteration": 3.2857279777526855
    },
    {
      "auxiliary_loss_clip": 0.01279529,
      "auxiliary_loss_mlp": 0.01029128,
      "balance_loss_clip": 1.05340171,
      "balance_loss_mlp": 1.02215099,
      "epoch": 0.5178861299825648,
      "flos": 26651248012800.0,
      "grad_norm": 1.497768734514173,
      "language_loss": 0.82938784,
      "learning_rate": 1.9813049527256923e-06,
      "loss": 0.85247445,
      "num_input_tokens_seen": 93026305,
      "step": 4307,
      "time_per_iteration": 2.675826072692871
    },
    {
      "auxiliary_loss_clip": 0.01337312,
      "auxiliary_loss_mlp": 0.01027603,
      "balance_loss_clip": 1.04719174,
      "balance_loss_mlp": 1.01968157,
      "epoch": 0.5180063728732038,
      "flos": 17931886260480.0,
      "grad_norm": 2.3384362165082995,
      "language_loss": 0.82050824,
      "learning_rate": 1.9805260165486252e-06,
      "loss": 0.8441574,
      "num_input_tokens_seen": 93045675,
      "step": 4308,
      "time_per_iteration": 2.7308170795440674
    },
    {
      "auxiliary_loss_clip": 0.012368,
      "auxiliary_loss_mlp": 0.01028328,
      "balance_loss_clip": 1.05358243,
      "balance_loss_mlp": 1.02047205,
      "epoch": 0.518126615763843,
      "flos": 19500895221120.0,
      "grad_norm": 3.4179440234518865,
      "language_loss": 0.86617184,
      "learning_rate": 1.9797470833257457e-06,
      "loss": 0.88882315,
      "num_input_tokens_seen": 93065375,
      "step": 4309,
      "time_per_iteration": 2.6335628032684326
    },
    {
      "auxiliary_loss_clip": 0.01241751,
      "auxiliary_loss_mlp": 0.01031321,
      "balance_loss_clip": 1.05748165,
      "balance_loss_mlp": 1.02380538,
      "epoch": 0.5182468586544821,
      "flos": 20704082117760.0,
      "grad_norm": 1.916818794600053,
      "language_loss": 0.77740729,
      "learning_rate": 1.9789681531752177e-06,
      "loss": 0.800138,
      "num_input_tokens_seen": 93085595,
      "step": 4310,
      "time_per_iteration": 2.6624913215637207
    },
    {
      "auxiliary_loss_clip": 0.01381287,
      "auxiliary_loss_mlp": 0.01028435,
      "balance_loss_clip": 1.05009615,
      "balance_loss_mlp": 1.02124715,
      "epoch": 0.5183671015451211,
      "flos": 23112107936640.0,
      "grad_norm": 1.8533470607369493,
      "language_loss": 0.72616667,
      "learning_rate": 1.978189226215204e-06,
      "loss": 0.75026387,
      "num_input_tokens_seen": 93106140,
      "step": 4311,
      "time_per_iteration": 2.752135992050171
    },
    {
      "auxiliary_loss_clip": 0.01188117,
      "auxiliary_loss_mlp": 0.01027192,
      "balance_loss_clip": 1.05510724,
      "balance_loss_mlp": 1.01894248,
      "epoch": 0.5184873444357603,
      "flos": 17597090568960.0,
      "grad_norm": 1.9373131513217623,
      "language_loss": 0.76788878,
      "learning_rate": 1.9774103025638675e-06,
      "loss": 0.7900418,
      "num_input_tokens_seen": 93124265,
      "step": 4312,
      "time_per_iteration": 2.583927631378174
    },
    {
      "auxiliary_loss_clip": 0.01390578,
      "auxiliary_loss_mlp": 0.01032314,
      "balance_loss_clip": 1.05479085,
      "balance_loss_mlp": 1.02463078,
      "epoch": 0.5186075873263993,
      "flos": 24936800883840.0,
      "grad_norm": 1.6329006416483713,
      "language_loss": 0.76581007,
      "learning_rate": 1.9766313823393696e-06,
      "loss": 0.79003894,
      "num_input_tokens_seen": 93145130,
      "step": 4313,
      "time_per_iteration": 2.769239664077759
    },
    {
      "auxiliary_loss_clip": 0.01374245,
      "auxiliary_loss_mlp": 0.01026938,
      "balance_loss_clip": 1.04327226,
      "balance_loss_mlp": 1.01964593,
      "epoch": 0.5187278302170384,
      "flos": 15190106244480.0,
      "grad_norm": 2.5115238212886224,
      "language_loss": 0.69066679,
      "learning_rate": 1.975852465659873e-06,
      "loss": 0.71467865,
      "num_input_tokens_seen": 93161110,
      "step": 4314,
      "time_per_iteration": 2.726191759109497
    },
    {
      "auxiliary_loss_clip": 0.01238677,
      "auxiliary_loss_mlp": 0.01032128,
      "balance_loss_clip": 1.05460048,
      "balance_loss_mlp": 1.02403963,
      "epoch": 0.5188480731076776,
      "flos": 25009412227200.0,
      "grad_norm": 2.92627360893441,
      "language_loss": 0.70018351,
      "learning_rate": 1.9750735526435377e-06,
      "loss": 0.72289157,
      "num_input_tokens_seen": 93178055,
      "step": 4315,
      "time_per_iteration": 2.6177966594696045
    },
    {
      "auxiliary_loss_clip": 0.01289634,
      "auxiliary_loss_mlp": 0.0102779,
      "balance_loss_clip": 1.05500746,
      "balance_loss_mlp": 1.01951051,
      "epoch": 0.5189683159983166,
      "flos": 24790141653120.0,
      "grad_norm": 2.58351472849173,
      "language_loss": 0.79185927,
      "learning_rate": 1.974294643408525e-06,
      "loss": 0.8150335,
      "num_input_tokens_seen": 93195850,
      "step": 4316,
      "time_per_iteration": 2.7322659492492676
    },
    {
      "auxiliary_loss_clip": 0.01241952,
      "auxiliary_loss_mlp": 0.01026869,
      "balance_loss_clip": 1.05245519,
      "balance_loss_mlp": 1.01909614,
      "epoch": 0.5190885588889557,
      "flos": 24754266944640.0,
      "grad_norm": 1.9478848300427416,
      "language_loss": 0.66817045,
      "learning_rate": 1.9735157380729947e-06,
      "loss": 0.69085866,
      "num_input_tokens_seen": 93216260,
      "step": 4317,
      "time_per_iteration": 2.6463534832000732
    },
    {
      "auxiliary_loss_clip": 0.01287366,
      "auxiliary_loss_mlp": 0.01025707,
      "balance_loss_clip": 1.05083704,
      "balance_loss_mlp": 1.01862931,
      "epoch": 0.5192088017795948,
      "flos": 24712646060160.0,
      "grad_norm": 2.7198960169598494,
      "language_loss": 0.84327376,
      "learning_rate": 1.9727368367551053e-06,
      "loss": 0.86640447,
      "num_input_tokens_seen": 93234810,
      "step": 4318,
      "time_per_iteration": 2.742995500564575
    },
    {
      "auxiliary_loss_clip": 0.01279769,
      "auxiliary_loss_mlp": 0.01028256,
      "balance_loss_clip": 1.05071568,
      "balance_loss_mlp": 1.02099574,
      "epoch": 0.5193290446702339,
      "flos": 27229588894080.0,
      "grad_norm": 1.937388422785079,
      "language_loss": 0.68397534,
      "learning_rate": 1.9719579395730164e-06,
      "loss": 0.70705557,
      "num_input_tokens_seen": 93254185,
      "step": 4319,
      "time_per_iteration": 2.7188711166381836
    },
    {
      "auxiliary_loss_clip": 0.01190514,
      "auxiliary_loss_mlp": 0.01032241,
      "balance_loss_clip": 1.05775952,
      "balance_loss_mlp": 1.0239892,
      "epoch": 0.5194492875608729,
      "flos": 11473352392320.0,
      "grad_norm": 2.9196186893576774,
      "language_loss": 0.93654859,
      "learning_rate": 1.9711790466448854e-06,
      "loss": 0.95877618,
      "num_input_tokens_seen": 93268205,
      "step": 4320,
      "time_per_iteration": 2.5437870025634766
    },
    {
      "auxiliary_loss_clip": 0.01385766,
      "auxiliary_loss_mlp": 0.01031599,
      "balance_loss_clip": 1.04968274,
      "balance_loss_mlp": 1.02247965,
      "epoch": 0.5195695304515121,
      "flos": 20338906498560.0,
      "grad_norm": 2.260103335477977,
      "language_loss": 0.71671426,
      "learning_rate": 1.9704001580888704e-06,
      "loss": 0.74088788,
      "num_input_tokens_seen": 93286945,
      "step": 4321,
      "time_per_iteration": 2.7915499210357666
    },
    {
      "auxiliary_loss_clip": 0.01280698,
      "auxiliary_loss_mlp": 0.02568368,
      "balance_loss_clip": 1.04977894,
      "balance_loss_mlp": 1.00063062,
      "epoch": 0.5196897733421512,
      "flos": 20048317470720.0,
      "grad_norm": 2.0477472260874032,
      "language_loss": 0.86790967,
      "learning_rate": 1.9696212740231283e-06,
      "loss": 0.90640032,
      "num_input_tokens_seen": 93305595,
      "step": 4322,
      "time_per_iteration": 3.5986363887786865
    },
    {
      "auxiliary_loss_clip": 0.0124189,
      "auxiliary_loss_mlp": 0.01030264,
      "balance_loss_clip": 1.05070102,
      "balance_loss_mlp": 1.0220449,
      "epoch": 0.5198100162327902,
      "flos": 23805507058560.0,
      "grad_norm": 2.1409868787869013,
      "language_loss": 0.82088947,
      "learning_rate": 1.9688423945658146e-06,
      "loss": 0.843611,
      "num_input_tokens_seen": 93326460,
      "step": 4323,
      "time_per_iteration": 2.6712229251861572
    },
    {
      "auxiliary_loss_clip": 0.01377446,
      "auxiliary_loss_mlp": 0.01029909,
      "balance_loss_clip": 1.04176986,
      "balance_loss_mlp": 1.02141571,
      "epoch": 0.5199302591234293,
      "flos": 24023951619840.0,
      "grad_norm": 2.4201044278583255,
      "language_loss": 0.72016865,
      "learning_rate": 1.9680635198350845e-06,
      "loss": 0.74424219,
      "num_input_tokens_seen": 93346170,
      "step": 4324,
      "time_per_iteration": 3.683738946914673
    },
    {
      "auxiliary_loss_clip": 0.01236639,
      "auxiliary_loss_mlp": 0.01034318,
      "balance_loss_clip": 1.05049133,
      "balance_loss_mlp": 1.02624726,
      "epoch": 0.5200505020140684,
      "flos": 26359366095360.0,
      "grad_norm": 5.9312439333795774,
      "language_loss": 0.72845531,
      "learning_rate": 1.967284649949093e-06,
      "loss": 0.75116491,
      "num_input_tokens_seen": 93365380,
      "step": 4325,
      "time_per_iteration": 2.621351718902588
    },
    {
      "auxiliary_loss_clip": 0.01338244,
      "auxiliary_loss_mlp": 0.01025585,
      "balance_loss_clip": 1.04957962,
      "balance_loss_mlp": 1.01788425,
      "epoch": 0.5201707449047075,
      "flos": 39604262284800.0,
      "grad_norm": 2.2014526690159575,
      "language_loss": 0.7213614,
      "learning_rate": 1.966505785025994e-06,
      "loss": 0.74499965,
      "num_input_tokens_seen": 93387285,
      "step": 4326,
      "time_per_iteration": 2.91732120513916
    },
    {
      "auxiliary_loss_clip": 0.01237802,
      "auxiliary_loss_mlp": 0.01027064,
      "balance_loss_clip": 1.05159366,
      "balance_loss_mlp": 1.01944625,
      "epoch": 0.5202909877953465,
      "flos": 53682788292480.0,
      "grad_norm": 2.5647337287136676,
      "language_loss": 0.76324397,
      "learning_rate": 1.965726925183941e-06,
      "loss": 0.78589261,
      "num_input_tokens_seen": 93410390,
      "step": 4327,
      "time_per_iteration": 3.0606443881988525
    },
    {
      "auxiliary_loss_clip": 0.01187729,
      "auxiliary_loss_mlp": 0.01025954,
      "balance_loss_clip": 1.05490661,
      "balance_loss_mlp": 1.01828909,
      "epoch": 0.5204112306859857,
      "flos": 19537021324800.0,
      "grad_norm": 1.95108666452016,
      "language_loss": 0.85008502,
      "learning_rate": 1.964948070541087e-06,
      "loss": 0.87222183,
      "num_input_tokens_seen": 93429050,
      "step": 4328,
      "time_per_iteration": 3.4612600803375244
    },
    {
      "auxiliary_loss_clip": 0.01230145,
      "auxiliary_loss_mlp": 0.01030903,
      "balance_loss_clip": 1.0490334,
      "balance_loss_mlp": 1.02330303,
      "epoch": 0.5205314735766248,
      "flos": 15304697608320.0,
      "grad_norm": 2.2101095567880615,
      "language_loss": 0.70133555,
      "learning_rate": 1.9641692212155816e-06,
      "loss": 0.72394603,
      "num_input_tokens_seen": 93446815,
      "step": 4329,
      "time_per_iteration": 2.6175167560577393
    },
    {
      "auxiliary_loss_clip": 0.01384678,
      "auxiliary_loss_mlp": 0.01032771,
      "balance_loss_clip": 1.05166292,
      "balance_loss_mlp": 1.02530205,
      "epoch": 0.5206517164672638,
      "flos": 59263701160320.0,
      "grad_norm": 2.188769119920752,
      "language_loss": 0.72715378,
      "learning_rate": 1.9633903773255777e-06,
      "loss": 0.75132829,
      "num_input_tokens_seen": 93469130,
      "step": 4330,
      "time_per_iteration": 3.9817237854003906
    },
    {
      "auxiliary_loss_clip": 0.01185065,
      "auxiliary_loss_mlp": 0.01029118,
      "balance_loss_clip": 1.05131555,
      "balance_loss_mlp": 1.02160788,
      "epoch": 0.520771959357903,
      "flos": 26871129118080.0,
      "grad_norm": 1.5605933611123073,
      "language_loss": 0.74599254,
      "learning_rate": 1.9626115389892237e-06,
      "loss": 0.76813436,
      "num_input_tokens_seen": 93489920,
      "step": 4331,
      "time_per_iteration": 2.639089345932007
    },
    {
      "auxiliary_loss_clip": 0.01243736,
      "auxiliary_loss_mlp": 0.01033271,
      "balance_loss_clip": 1.05101168,
      "balance_loss_mlp": 1.0251472,
      "epoch": 0.520892202248542,
      "flos": 26907075653760.0,
      "grad_norm": 6.277888899689654,
      "language_loss": 0.85904956,
      "learning_rate": 1.96183270632467e-06,
      "loss": 0.88181973,
      "num_input_tokens_seen": 93509770,
      "step": 4332,
      "time_per_iteration": 2.70280122756958
    },
    {
      "auxiliary_loss_clip": 0.01234204,
      "auxiliary_loss_mlp": 0.0257199,
      "balance_loss_clip": 1.04754663,
      "balance_loss_mlp": 1.00046062,
      "epoch": 0.5210124451391811,
      "flos": 25849434666240.0,
      "grad_norm": 1.8837176155762203,
      "language_loss": 0.79119194,
      "learning_rate": 1.9610538794500644e-06,
      "loss": 0.82925385,
      "num_input_tokens_seen": 93529320,
      "step": 4333,
      "time_per_iteration": 2.7595667839050293
    },
    {
      "auxiliary_loss_clip": 0.01234965,
      "auxiliary_loss_mlp": 0.01000573,
      "balance_loss_clip": 1.0105623,
      "balance_loss_mlp": 0.99962503,
      "epoch": 0.5211326880298203,
      "flos": 70553804319360.0,
      "grad_norm": 0.7672028944906849,
      "language_loss": 0.59413958,
      "learning_rate": 1.9602750584835542e-06,
      "loss": 0.61649501,
      "num_input_tokens_seen": 93595255,
      "step": 4334,
      "time_per_iteration": 3.379917621612549
    },
    {
      "auxiliary_loss_clip": 0.01285563,
      "auxiliary_loss_mlp": 0.01027955,
      "balance_loss_clip": 1.04904974,
      "balance_loss_mlp": 1.0208261,
      "epoch": 0.5212529309204593,
      "flos": 15628898787840.0,
      "grad_norm": 1.8762834780335629,
      "language_loss": 0.82510853,
      "learning_rate": 1.959496243543286e-06,
      "loss": 0.84824371,
      "num_input_tokens_seen": 93613135,
      "step": 4335,
      "time_per_iteration": 2.629377603530884
    },
    {
      "auxiliary_loss_clip": 0.01243694,
      "auxiliary_loss_mlp": 0.010284,
      "balance_loss_clip": 1.05906928,
      "balance_loss_mlp": 1.02053809,
      "epoch": 0.5213731738110984,
      "flos": 26242655829120.0,
      "grad_norm": 2.1934465193050703,
      "language_loss": 0.79182589,
      "learning_rate": 1.9587174347474057e-06,
      "loss": 0.81454682,
      "num_input_tokens_seen": 93629645,
      "step": 4336,
      "time_per_iteration": 2.6937756538391113
    },
    {
      "auxiliary_loss_clip": 0.01374732,
      "auxiliary_loss_mlp": 0.01031651,
      "balance_loss_clip": 1.04581821,
      "balance_loss_mlp": 1.02346683,
      "epoch": 0.5214934167017375,
      "flos": 19418407637760.0,
      "grad_norm": 2.1327781413731706,
      "language_loss": 0.82022023,
      "learning_rate": 1.9579386322140574e-06,
      "loss": 0.84428406,
      "num_input_tokens_seen": 93645325,
      "step": 4337,
      "time_per_iteration": 2.6899569034576416
    },
    {
      "auxiliary_loss_clip": 0.01192087,
      "auxiliary_loss_mlp": 0.02571493,
      "balance_loss_clip": 1.05725825,
      "balance_loss_mlp": 1.00039113,
      "epoch": 0.5216136595923766,
      "flos": 30955788023040.0,
      "grad_norm": 1.6876930415929043,
      "language_loss": 0.80760574,
      "learning_rate": 1.9571598360613854e-06,
      "loss": 0.84524155,
      "num_input_tokens_seen": 93668200,
      "step": 4338,
      "time_per_iteration": 2.7201995849609375
    },
    {
      "auxiliary_loss_clip": 0.01277848,
      "auxiliary_loss_mlp": 0.01026432,
      "balance_loss_clip": 1.0465858,
      "balance_loss_mlp": 1.01872778,
      "epoch": 0.5217339024830157,
      "flos": 21945047143680.0,
      "grad_norm": 4.031970000213029,
      "language_loss": 0.69579482,
      "learning_rate": 1.956381046407532e-06,
      "loss": 0.71883762,
      "num_input_tokens_seen": 93688495,
      "step": 4339,
      "time_per_iteration": 2.630032777786255
    },
    {
      "auxiliary_loss_clip": 0.01337434,
      "auxiliary_loss_mlp": 0.0103186,
      "balance_loss_clip": 1.0501523,
      "balance_loss_mlp": 1.02436161,
      "epoch": 0.5218541453736548,
      "flos": 20923209037440.0,
      "grad_norm": 1.7576520104422564,
      "language_loss": 0.86462438,
      "learning_rate": 1.9556022633706394e-06,
      "loss": 0.88831723,
      "num_input_tokens_seen": 93707285,
      "step": 4340,
      "time_per_iteration": 2.7641727924346924
    },
    {
      "auxiliary_loss_clip": 0.01281698,
      "auxiliary_loss_mlp": 0.01027574,
      "balance_loss_clip": 1.05162716,
      "balance_loss_mlp": 1.01975393,
      "epoch": 0.5219743882642939,
      "flos": 23951663498880.0,
      "grad_norm": 1.6896153027725294,
      "language_loss": 0.79706281,
      "learning_rate": 1.954823487068848e-06,
      "loss": 0.82015562,
      "num_input_tokens_seen": 93727495,
      "step": 4341,
      "time_per_iteration": 2.6593313217163086
    },
    {
      "auxiliary_loss_clip": 0.01234508,
      "auxiliary_loss_mlp": 0.0103069,
      "balance_loss_clip": 1.05590737,
      "balance_loss_mlp": 1.02363002,
      "epoch": 0.5220946311549329,
      "flos": 28799280213120.0,
      "grad_norm": 2.3372791669616655,
      "language_loss": 0.80849111,
      "learning_rate": 1.9540447176202976e-06,
      "loss": 0.83114302,
      "num_input_tokens_seen": 93748740,
      "step": 4342,
      "time_per_iteration": 2.724114179611206
    },
    {
      "auxiliary_loss_clip": 0.01119268,
      "auxiliary_loss_mlp": 0.01001363,
      "balance_loss_clip": 1.01079834,
      "balance_loss_mlp": 1.00046277,
      "epoch": 0.5222148740455721,
      "flos": 67189369017600.0,
      "grad_norm": 0.8748230963469118,
      "language_loss": 0.60737705,
      "learning_rate": 1.9532659551431272e-06,
      "loss": 0.62858331,
      "num_input_tokens_seen": 93815770,
      "step": 4343,
      "time_per_iteration": 3.372098684310913
    },
    {
      "auxiliary_loss_clip": 0.0123698,
      "auxiliary_loss_mlp": 0.0102807,
      "balance_loss_clip": 1.05072343,
      "balance_loss_mlp": 1.02082777,
      "epoch": 0.5223351169362112,
      "flos": 61856164339200.0,
      "grad_norm": 2.2231224961209737,
      "language_loss": 0.672966,
      "learning_rate": 1.9524871997554744e-06,
      "loss": 0.69561648,
      "num_input_tokens_seen": 93843530,
      "step": 4344,
      "time_per_iteration": 2.997997522354126
    },
    {
      "auxiliary_loss_clip": 0.01235729,
      "auxiliary_loss_mlp": 0.0103425,
      "balance_loss_clip": 1.053105,
      "balance_loss_mlp": 1.02654314,
      "epoch": 0.5224553598268502,
      "flos": 14647388676480.0,
      "grad_norm": 2.5540092213146552,
      "language_loss": 0.80514133,
      "learning_rate": 1.951708451575475e-06,
      "loss": 0.82784116,
      "num_input_tokens_seen": 93860595,
      "step": 4345,
      "time_per_iteration": 2.6836438179016113
    },
    {
      "auxiliary_loss_clip": 0.01346205,
      "auxiliary_loss_mlp": 0.01026581,
      "balance_loss_clip": 1.04903185,
      "balance_loss_mlp": 1.01912141,
      "epoch": 0.5225756027174894,
      "flos": 14826043946880.0,
      "grad_norm": 2.0565095741394144,
      "language_loss": 0.82200706,
      "learning_rate": 1.9509297107212657e-06,
      "loss": 0.84573495,
      "num_input_tokens_seen": 93877365,
      "step": 4346,
      "time_per_iteration": 2.6244020462036133
    },
    {
      "auxiliary_loss_clip": 0.01187757,
      "auxiliary_loss_mlp": 0.0103305,
      "balance_loss_clip": 1.05580723,
      "balance_loss_mlp": 1.02557564,
      "epoch": 0.5226958456081284,
      "flos": 23512009029120.0,
      "grad_norm": 1.678582240832953,
      "language_loss": 0.79004824,
      "learning_rate": 1.95015097731098e-06,
      "loss": 0.81225628,
      "num_input_tokens_seen": 93896855,
      "step": 4347,
      "time_per_iteration": 2.6659224033355713
    },
    {
      "auxiliary_loss_clip": 0.01187519,
      "auxiliary_loss_mlp": 0.01029213,
      "balance_loss_clip": 1.05551696,
      "balance_loss_mlp": 1.02135444,
      "epoch": 0.5228160884987675,
      "flos": 19062928690560.0,
      "grad_norm": 2.722424368247757,
      "language_loss": 0.82144129,
      "learning_rate": 1.949372251462751e-06,
      "loss": 0.84360862,
      "num_input_tokens_seen": 93914270,
      "step": 4348,
      "time_per_iteration": 3.5072882175445557
    },
    {
      "auxiliary_loss_clip": 0.01336087,
      "auxiliary_loss_mlp": 0.02565476,
      "balance_loss_clip": 1.05181313,
      "balance_loss_mlp": 1.00046873,
      "epoch": 0.5229363313894067,
      "flos": 21063224252160.0,
      "grad_norm": 2.065775578812693,
      "language_loss": 0.82822466,
      "learning_rate": 1.9485935332947124e-06,
      "loss": 0.86724031,
      "num_input_tokens_seen": 93932180,
      "step": 4349,
      "time_per_iteration": 2.7020936012268066
    },
    {
      "auxiliary_loss_clip": 0.01279234,
      "auxiliary_loss_mlp": 0.01028943,
      "balance_loss_clip": 1.05061436,
      "balance_loss_mlp": 1.02184057,
      "epoch": 0.5230565742800457,
      "flos": 14830389492480.0,
      "grad_norm": 2.5782680764155597,
      "language_loss": 0.83511668,
      "learning_rate": 1.947814822924993e-06,
      "loss": 0.85819846,
      "num_input_tokens_seen": 93949690,
      "step": 4350,
      "time_per_iteration": 3.510423183441162
    },
    {
      "auxiliary_loss_clip": 0.01188548,
      "auxiliary_loss_mlp": 0.01022607,
      "balance_loss_clip": 1.05723906,
      "balance_loss_mlp": 1.01555848,
      "epoch": 0.5231768171706848,
      "flos": 25813021253760.0,
      "grad_norm": 1.9080543316385168,
      "language_loss": 0.82965744,
      "learning_rate": 1.9470361204717236e-06,
      "loss": 0.85176897,
      "num_input_tokens_seen": 93968830,
      "step": 4351,
      "time_per_iteration": 2.584832191467285
    },
    {
      "auxiliary_loss_clip": 0.01336179,
      "auxiliary_loss_mlp": 0.02569373,
      "balance_loss_clip": 1.04802775,
      "balance_loss_mlp": 1.00040722,
      "epoch": 0.5232970600613239,
      "flos": 22743807834240.0,
      "grad_norm": 1.6683820784087944,
      "language_loss": 0.8090679,
      "learning_rate": 1.9462574260530326e-06,
      "loss": 0.84812343,
      "num_input_tokens_seen": 93989110,
      "step": 4352,
      "time_per_iteration": 2.774475574493408
    },
    {
      "auxiliary_loss_clip": 0.01229245,
      "auxiliary_loss_mlp": 0.01029366,
      "balance_loss_clip": 1.04942822,
      "balance_loss_mlp": 1.0213728,
      "epoch": 0.523417302951963,
      "flos": 17310703432320.0,
      "grad_norm": 1.713449073014465,
      "language_loss": 0.81057239,
      "learning_rate": 1.9454787397870472e-06,
      "loss": 0.83315849,
      "num_input_tokens_seen": 94006430,
      "step": 4353,
      "time_per_iteration": 2.5859947204589844
    },
    {
      "auxiliary_loss_clip": 0.01333458,
      "auxiliary_loss_mlp": 0.01028411,
      "balance_loss_clip": 1.05029106,
      "balance_loss_mlp": 1.02129698,
      "epoch": 0.523537545842602,
      "flos": 18551740285440.0,
      "grad_norm": 2.2572265607389963,
      "language_loss": 0.71807611,
      "learning_rate": 1.944700061791894e-06,
      "loss": 0.74169487,
      "num_input_tokens_seen": 94024825,
      "step": 4354,
      "time_per_iteration": 3.773970365524292
    },
    {
      "auxiliary_loss_clip": 0.01236019,
      "auxiliary_loss_mlp": 0.01028984,
      "balance_loss_clip": 1.054268,
      "balance_loss_mlp": 1.0216819,
      "epoch": 0.5236577887332411,
      "flos": 19719267955200.0,
      "grad_norm": 2.388066339957428,
      "language_loss": 0.65084988,
      "learning_rate": 1.943921392185698e-06,
      "loss": 0.67349994,
      "num_input_tokens_seen": 94043450,
      "step": 4355,
      "time_per_iteration": 2.8274667263031006
    },
    {
      "auxiliary_loss_clip": 0.01190584,
      "auxiliary_loss_mlp": 0.01027486,
      "balance_loss_clip": 1.04949367,
      "balance_loss_mlp": 1.02045834,
      "epoch": 0.5237780316238803,
      "flos": 23550218121600.0,
      "grad_norm": 2.0070271081645994,
      "language_loss": 0.7719239,
      "learning_rate": 1.9431427310865814e-06,
      "loss": 0.79410452,
      "num_input_tokens_seen": 94063055,
      "step": 4356,
      "time_per_iteration": 4.049981355667114
    },
    {
      "auxiliary_loss_clip": 0.01329103,
      "auxiliary_loss_mlp": 0.01029566,
      "balance_loss_clip": 1.05029154,
      "balance_loss_mlp": 1.02253246,
      "epoch": 0.5238982745145193,
      "flos": 22491894775680.0,
      "grad_norm": 2.560035319514936,
      "language_loss": 0.78547043,
      "learning_rate": 1.942364078612667e-06,
      "loss": 0.80905712,
      "num_input_tokens_seen": 94081785,
      "step": 4357,
      "time_per_iteration": 2.781801462173462
    },
    {
      "auxiliary_loss_clip": 0.01241783,
      "auxiliary_loss_mlp": 0.01031577,
      "balance_loss_clip": 1.04830813,
      "balance_loss_mlp": 1.0239594,
      "epoch": 0.5240185174051584,
      "flos": 27088927234560.0,
      "grad_norm": 1.8398002272572562,
      "language_loss": 0.75898385,
      "learning_rate": 1.9415854348820765e-06,
      "loss": 0.78171748,
      "num_input_tokens_seen": 94101635,
      "step": 4358,
      "time_per_iteration": 2.7578771114349365
    },
    {
      "auxiliary_loss_clip": 0.01139789,
      "auxiliary_loss_mlp": 0.01024916,
      "balance_loss_clip": 1.05058813,
      "balance_loss_mlp": 1.0172689,
      "epoch": 0.5241387602957975,
      "flos": 22674680110080.0,
      "grad_norm": 2.71802765816635,
      "language_loss": 0.6839357,
      "learning_rate": 1.940806800012929e-06,
      "loss": 0.70558274,
      "num_input_tokens_seen": 94121705,
      "step": 4359,
      "time_per_iteration": 2.5751709938049316
    },
    {
      "auxiliary_loss_clip": 0.0137596,
      "auxiliary_loss_mlp": 0.02570481,
      "balance_loss_clip": 1.04819143,
      "balance_loss_mlp": 1.00035024,
      "epoch": 0.5242590031864366,
      "flos": 40553453134080.0,
      "grad_norm": 10.445961543803138,
      "language_loss": 0.63726026,
      "learning_rate": 1.9400281741233432e-06,
      "loss": 0.67672467,
      "num_input_tokens_seen": 94146595,
      "step": 4360,
      "time_per_iteration": 2.9346792697906494
    },
    {
      "auxiliary_loss_clip": 0.01222633,
      "auxiliary_loss_mlp": 0.01002627,
      "balance_loss_clip": 1.00940406,
      "balance_loss_mlp": 1.00153661,
      "epoch": 0.5243792460770756,
      "flos": 66676313105280.0,
      "grad_norm": 0.6581290227637631,
      "language_loss": 0.52531052,
      "learning_rate": 1.939249557331435e-06,
      "loss": 0.5475632,
      "num_input_tokens_seen": 94212410,
      "step": 4361,
      "time_per_iteration": 3.300522804260254
    },
    {
      "auxiliary_loss_clip": 0.01342039,
      "auxiliary_loss_mlp": 0.01025389,
      "balance_loss_clip": 1.05023849,
      "balance_loss_mlp": 1.01771796,
      "epoch": 0.5244994889677148,
      "flos": 28183663992960.0,
      "grad_norm": 1.9515354800268836,
      "language_loss": 0.73284018,
      "learning_rate": 1.938470949755321e-06,
      "loss": 0.75651443,
      "num_input_tokens_seen": 94232290,
      "step": 4362,
      "time_per_iteration": 2.8130288124084473
    },
    {
      "auxiliary_loss_clip": 0.01230465,
      "auxiliary_loss_mlp": 0.01002695,
      "balance_loss_clip": 1.00902855,
      "balance_loss_mlp": 1.00166976,
      "epoch": 0.5246197318583539,
      "flos": 65950379239680.0,
      "grad_norm": 0.8085968802254015,
      "language_loss": 0.55663246,
      "learning_rate": 1.937692351513115e-06,
      "loss": 0.57896405,
      "num_input_tokens_seen": 94291285,
      "step": 4363,
      "time_per_iteration": 3.2948267459869385
    },
    {
      "auxiliary_loss_clip": 0.01238788,
      "auxiliary_loss_mlp": 0.01032416,
      "balance_loss_clip": 1.05140913,
      "balance_loss_mlp": 1.02436328,
      "epoch": 0.5247399747489929,
      "flos": 21033490769280.0,
      "grad_norm": 1.7289018902683848,
      "language_loss": 0.80647659,
      "learning_rate": 1.9369137627229297e-06,
      "loss": 0.82918859,
      "num_input_tokens_seen": 94309685,
      "step": 4364,
      "time_per_iteration": 2.652724504470825
    },
    {
      "auxiliary_loss_clip": 0.01232174,
      "auxiliary_loss_mlp": 0.01027017,
      "balance_loss_clip": 1.05250573,
      "balance_loss_mlp": 1.01932776,
      "epoch": 0.5248602176396321,
      "flos": 19025940660480.0,
      "grad_norm": 2.368646272956753,
      "language_loss": 0.88653195,
      "learning_rate": 1.936135183502877e-06,
      "loss": 0.90912384,
      "num_input_tokens_seen": 94326985,
      "step": 4365,
      "time_per_iteration": 2.833144187927246
    },
    {
      "auxiliary_loss_clip": 0.0123712,
      "auxiliary_loss_mlp": 0.01027038,
      "balance_loss_clip": 1.0492959,
      "balance_loss_mlp": 1.01935506,
      "epoch": 0.5249804605302711,
      "flos": 22200084685440.0,
      "grad_norm": 2.816656488995863,
      "language_loss": 0.80988425,
      "learning_rate": 1.935356613971066e-06,
      "loss": 0.83252585,
      "num_input_tokens_seen": 94347645,
      "step": 4366,
      "time_per_iteration": 2.721973180770874
    },
    {
      "auxiliary_loss_clip": 0.01282273,
      "auxiliary_loss_mlp": 0.02567534,
      "balance_loss_clip": 1.05166769,
      "balance_loss_mlp": 1.00040197,
      "epoch": 0.5251007034209102,
      "flos": 23805686626560.0,
      "grad_norm": 1.823372909035621,
      "language_loss": 0.76930439,
      "learning_rate": 1.9345780542456047e-06,
      "loss": 0.8078025,
      "num_input_tokens_seen": 94367020,
      "step": 4367,
      "time_per_iteration": 2.7309157848358154
    },
    {
      "auxiliary_loss_clip": 0.01230664,
      "auxiliary_loss_mlp": 0.01029062,
      "balance_loss_clip": 1.05181372,
      "balance_loss_mlp": 1.02142644,
      "epoch": 0.5252209463115494,
      "flos": 23294605962240.0,
      "grad_norm": 1.8886989854972738,
      "language_loss": 0.7149393,
      "learning_rate": 1.9337995044446007e-06,
      "loss": 0.73753655,
      "num_input_tokens_seen": 94385860,
      "step": 4368,
      "time_per_iteration": 2.6662302017211914
    },
    {
      "auxiliary_loss_clip": 0.01240054,
      "auxiliary_loss_mlp": 0.01024249,
      "balance_loss_clip": 1.05300331,
      "balance_loss_mlp": 1.01658988,
      "epoch": 0.5253411892021884,
      "flos": 19828687760640.0,
      "grad_norm": 2.6404054414212483,
      "language_loss": 0.79840845,
      "learning_rate": 1.9330209646861596e-06,
      "loss": 0.82105148,
      "num_input_tokens_seen": 94405010,
      "step": 4369,
      "time_per_iteration": 2.6257431507110596
    },
    {
      "auxiliary_loss_clip": 0.01279554,
      "auxiliary_loss_mlp": 0.01026532,
      "balance_loss_clip": 1.04967117,
      "balance_loss_mlp": 1.01942134,
      "epoch": 0.5254614320928275,
      "flos": 24133730561280.0,
      "grad_norm": 2.1095451305151585,
      "language_loss": 0.77695131,
      "learning_rate": 1.9322424350883843e-06,
      "loss": 0.80001211,
      "num_input_tokens_seen": 94426845,
      "step": 4370,
      "time_per_iteration": 2.705080509185791
    },
    {
      "auxiliary_loss_clip": 0.01283788,
      "auxiliary_loss_mlp": 0.01029936,
      "balance_loss_clip": 1.04891789,
      "balance_loss_mlp": 1.02288771,
      "epoch": 0.5255816749834666,
      "flos": 24644954880000.0,
      "grad_norm": 1.644014474061873,
      "language_loss": 0.78909898,
      "learning_rate": 1.931463915769379e-06,
      "loss": 0.81223619,
      "num_input_tokens_seen": 94446960,
      "step": 4371,
      "time_per_iteration": 2.736386775970459
    },
    {
      "auxiliary_loss_clip": 0.01380426,
      "auxiliary_loss_mlp": 0.01025952,
      "balance_loss_clip": 1.04610276,
      "balance_loss_mlp": 1.0186801,
      "epoch": 0.5257019178741057,
      "flos": 14136595320960.0,
      "grad_norm": 2.8298840898551103,
      "language_loss": 0.74177456,
      "learning_rate": 1.930685406847242e-06,
      "loss": 0.76583838,
      "num_input_tokens_seen": 94461535,
      "step": 4372,
      "time_per_iteration": 2.7104313373565674
    },
    {
      "auxiliary_loss_clip": 0.01279657,
      "auxiliary_loss_mlp": 0.01028835,
      "balance_loss_clip": 1.05026329,
      "balance_loss_mlp": 1.02117026,
      "epoch": 0.5258221607647448,
      "flos": 23548961145600.0,
      "grad_norm": 1.5513260804614328,
      "language_loss": 0.81631428,
      "learning_rate": 1.9299069084400734e-06,
      "loss": 0.83939922,
      "num_input_tokens_seen": 94482395,
      "step": 4373,
      "time_per_iteration": 2.730428695678711
    },
    {
      "auxiliary_loss_clip": 0.01328703,
      "auxiliary_loss_mlp": 0.01024081,
      "balance_loss_clip": 1.05078983,
      "balance_loss_mlp": 1.01653481,
      "epoch": 0.5259424036553839,
      "flos": 24966103403520.0,
      "grad_norm": 1.979560062245474,
      "language_loss": 0.69693476,
      "learning_rate": 1.9291284206659717e-06,
      "loss": 0.72046262,
      "num_input_tokens_seen": 94500580,
      "step": 4374,
      "time_per_iteration": 3.5764782428741455
    },
    {
      "auxiliary_loss_clip": 0.01186814,
      "auxiliary_loss_mlp": 0.01024917,
      "balance_loss_clip": 1.05377269,
      "balance_loss_mlp": 1.017627,
      "epoch": 0.526062646546023,
      "flos": 28763908295040.0,
      "grad_norm": 2.222806581810258,
      "language_loss": 0.7176162,
      "learning_rate": 1.928349943643032e-06,
      "loss": 0.73973346,
      "num_input_tokens_seen": 94519680,
      "step": 4375,
      "time_per_iteration": 2.6171255111694336
    },
    {
      "auxiliary_loss_clip": 0.01234359,
      "auxiliary_loss_mlp": 0.0102855,
      "balance_loss_clip": 1.05569887,
      "balance_loss_mlp": 1.02134109,
      "epoch": 0.526182889436662,
      "flos": 22821375254400.0,
      "grad_norm": 1.837858489579107,
      "language_loss": 0.82025409,
      "learning_rate": 1.9275714774893493e-06,
      "loss": 0.84288317,
      "num_input_tokens_seen": 94539135,
      "step": 4376,
      "time_per_iteration": 3.5779716968536377
    },
    {
      "auxiliary_loss_clip": 0.01325056,
      "auxiliary_loss_mlp": 0.01024399,
      "balance_loss_clip": 1.04433155,
      "balance_loss_mlp": 1.01606345,
      "epoch": 0.5263031323273012,
      "flos": 22929466256640.0,
      "grad_norm": 2.817806838562993,
      "language_loss": 0.72796285,
      "learning_rate": 1.9267930223230154e-06,
      "loss": 0.75145745,
      "num_input_tokens_seen": 94557610,
      "step": 4377,
      "time_per_iteration": 2.688720464706421
    },
    {
      "auxiliary_loss_clip": 0.01287244,
      "auxiliary_loss_mlp": 0.01029649,
      "balance_loss_clip": 1.05192029,
      "balance_loss_mlp": 1.02227581,
      "epoch": 0.5264233752179402,
      "flos": 17748634049280.0,
      "grad_norm": 2.2678925788639877,
      "language_loss": 0.78067261,
      "learning_rate": 1.9260145782621224e-06,
      "loss": 0.80384159,
      "num_input_tokens_seen": 94575390,
      "step": 4378,
      "time_per_iteration": 2.7068588733673096
    },
    {
      "auxiliary_loss_clip": 0.01279275,
      "auxiliary_loss_mlp": 0.01029823,
      "balance_loss_clip": 1.05218554,
      "balance_loss_mlp": 1.02253306,
      "epoch": 0.5265436181085793,
      "flos": 24421626069120.0,
      "grad_norm": 2.1262928666574044,
      "language_loss": 0.88012147,
      "learning_rate": 1.925236145424758e-06,
      "loss": 0.90321249,
      "num_input_tokens_seen": 94594210,
      "step": 4379,
      "time_per_iteration": 2.705787181854248
    },
    {
      "auxiliary_loss_clip": 0.01125212,
      "auxiliary_loss_mlp": 0.01002515,
      "balance_loss_clip": 1.01034737,
      "balance_loss_mlp": 1.00152004,
      "epoch": 0.5266638609992185,
      "flos": 69207298156800.0,
      "grad_norm": 0.6955536957429622,
      "language_loss": 0.57536066,
      "learning_rate": 1.924457723929012e-06,
      "loss": 0.59663796,
      "num_input_tokens_seen": 94665020,
      "step": 4380,
      "time_per_iteration": 4.267252206802368
    },
    {
      "auxiliary_loss_clip": 0.01232179,
      "auxiliary_loss_mlp": 0.01027393,
      "balance_loss_clip": 1.05097461,
      "balance_loss_mlp": 1.01972759,
      "epoch": 0.5267841038898575,
      "flos": 20738699850240.0,
      "grad_norm": 1.5197347547548996,
      "language_loss": 0.82527137,
      "learning_rate": 1.9236793138929685e-06,
      "loss": 0.84786713,
      "num_input_tokens_seen": 94684290,
      "step": 4381,
      "time_per_iteration": 2.6301937103271484
    },
    {
      "auxiliary_loss_clip": 0.01237902,
      "auxiliary_loss_mlp": 0.01024702,
      "balance_loss_clip": 1.05120325,
      "balance_loss_mlp": 1.01723409,
      "epoch": 0.5269043467804966,
      "flos": 17234392988160.0,
      "grad_norm": 2.6381634810182035,
      "language_loss": 0.81363046,
      "learning_rate": 1.9229009154347133e-06,
      "loss": 0.83625644,
      "num_input_tokens_seen": 94701880,
      "step": 4382,
      "time_per_iteration": 2.619692087173462
    },
    {
      "auxiliary_loss_clip": 0.01371745,
      "auxiliary_loss_mlp": 0.02564912,
      "balance_loss_clip": 1.04472291,
      "balance_loss_mlp": 1.000314,
      "epoch": 0.5270245896711357,
      "flos": 18223157646720.0,
      "grad_norm": 2.0412931460655424,
      "language_loss": 0.81044722,
      "learning_rate": 1.922122528672327e-06,
      "loss": 0.8498137,
      "num_input_tokens_seen": 94720545,
      "step": 4383,
      "time_per_iteration": 3.6411266326904297
    },
    {
      "auxiliary_loss_clip": 0.01182648,
      "auxiliary_loss_mlp": 0.01025453,
      "balance_loss_clip": 1.05365562,
      "balance_loss_mlp": 1.01817799,
      "epoch": 0.5271448325617748,
      "flos": 21287558643840.0,
      "grad_norm": 3.8609273611739106,
      "language_loss": 0.78727651,
      "learning_rate": 1.9213441537238914e-06,
      "loss": 0.80935752,
      "num_input_tokens_seen": 94737420,
      "step": 4384,
      "time_per_iteration": 2.6276931762695312
    },
    {
      "auxiliary_loss_clip": 0.01279281,
      "auxiliary_loss_mlp": 0.01003541,
      "balance_loss_clip": 1.01622629,
      "balance_loss_mlp": 1.00260568,
      "epoch": 0.5272650754524139,
      "flos": 65495497403520.0,
      "grad_norm": 0.9692769502701037,
      "language_loss": 0.57322073,
      "learning_rate": 1.920565790707485e-06,
      "loss": 0.59604895,
      "num_input_tokens_seen": 94802810,
      "step": 4385,
      "time_per_iteration": 3.407212495803833
    },
    {
      "auxiliary_loss_clip": 0.01289363,
      "auxiliary_loss_mlp": 0.01027933,
      "balance_loss_clip": 1.04477811,
      "balance_loss_mlp": 1.01984477,
      "epoch": 0.527385318343053,
      "flos": 19676426008320.0,
      "grad_norm": 2.0414887742862007,
      "language_loss": 0.66100335,
      "learning_rate": 1.9197874397411853e-06,
      "loss": 0.68417633,
      "num_input_tokens_seen": 94819440,
      "step": 4386,
      "time_per_iteration": 2.7652432918548584
    },
    {
      "auxiliary_loss_clip": 0.01335735,
      "auxiliary_loss_mlp": 0.01023068,
      "balance_loss_clip": 1.04485822,
      "balance_loss_mlp": 1.01527119,
      "epoch": 0.5275055612336921,
      "flos": 12712018947840.0,
      "grad_norm": 3.206329760329269,
      "language_loss": 0.67306674,
      "learning_rate": 1.919009100943067e-06,
      "loss": 0.69665468,
      "num_input_tokens_seen": 94835130,
      "step": 4387,
      "time_per_iteration": 2.7600257396698
    },
    {
      "auxiliary_loss_clip": 0.01390785,
      "auxiliary_loss_mlp": 0.01030066,
      "balance_loss_clip": 1.0498184,
      "balance_loss_mlp": 1.02200747,
      "epoch": 0.5276258041243311,
      "flos": 17749029098880.0,
      "grad_norm": 1.9415243942071785,
      "language_loss": 0.65773731,
      "learning_rate": 1.9182307744312043e-06,
      "loss": 0.68194586,
      "num_input_tokens_seen": 94852235,
      "step": 4388,
      "time_per_iteration": 2.829319477081299
    },
    {
      "auxiliary_loss_clip": 0.01288593,
      "auxiliary_loss_mlp": 0.01030596,
      "balance_loss_clip": 1.05049539,
      "balance_loss_mlp": 1.02293718,
      "epoch": 0.5277460470149702,
      "flos": 22710447077760.0,
      "grad_norm": 1.9537225621420309,
      "language_loss": 0.76615983,
      "learning_rate": 1.9174524603236676e-06,
      "loss": 0.7893517,
      "num_input_tokens_seen": 94871185,
      "step": 4389,
      "time_per_iteration": 2.698467969894409
    },
    {
      "auxiliary_loss_clip": 0.01282301,
      "auxiliary_loss_mlp": 0.01027874,
      "balance_loss_clip": 1.05097365,
      "balance_loss_mlp": 1.02035785,
      "epoch": 0.5278662899056094,
      "flos": 19902699734400.0,
      "grad_norm": 2.173295876279596,
      "language_loss": 0.76662105,
      "learning_rate": 1.916674158738527e-06,
      "loss": 0.7897228,
      "num_input_tokens_seen": 94890090,
      "step": 4390,
      "time_per_iteration": 2.6360151767730713
    },
    {
      "auxiliary_loss_clip": 0.01325762,
      "auxiliary_loss_mlp": 0.02572572,
      "balance_loss_clip": 1.04967356,
      "balance_loss_mlp": 1.00032401,
      "epoch": 0.5279865327962484,
      "flos": 18005215875840.0,
      "grad_norm": 2.2596280714456154,
      "language_loss": 0.60418236,
      "learning_rate": 1.9158958697938506e-06,
      "loss": 0.64316571,
      "num_input_tokens_seen": 94908470,
      "step": 4391,
      "time_per_iteration": 2.697342872619629
    },
    {
      "auxiliary_loss_clip": 0.01281044,
      "auxiliary_loss_mlp": 0.01030816,
      "balance_loss_clip": 1.04882574,
      "balance_loss_mlp": 1.02329421,
      "epoch": 0.5281067756868875,
      "flos": 15924443892480.0,
      "grad_norm": 2.417463263801839,
      "language_loss": 0.85552049,
      "learning_rate": 1.9151175936077032e-06,
      "loss": 0.8786391,
      "num_input_tokens_seen": 94923440,
      "step": 4392,
      "time_per_iteration": 2.6112420558929443
    },
    {
      "auxiliary_loss_clip": 0.01230658,
      "auxiliary_loss_mlp": 0.01024686,
      "balance_loss_clip": 1.05122983,
      "balance_loss_mlp": 1.01714325,
      "epoch": 0.5282270185775266,
      "flos": 19426488197760.0,
      "grad_norm": 1.669200631043679,
      "language_loss": 0.79697526,
      "learning_rate": 1.9143393302981507e-06,
      "loss": 0.8195287,
      "num_input_tokens_seen": 94941125,
      "step": 4393,
      "time_per_iteration": 2.6463088989257812
    },
    {
      "auxiliary_loss_clip": 0.01285358,
      "auxiliary_loss_mlp": 0.01030193,
      "balance_loss_clip": 1.05105245,
      "balance_loss_mlp": 1.02177095,
      "epoch": 0.5283472614681657,
      "flos": 16399613934720.0,
      "grad_norm": 1.9047231284573756,
      "language_loss": 0.83469629,
      "learning_rate": 1.913561079983252e-06,
      "loss": 0.85785186,
      "num_input_tokens_seen": 94959950,
      "step": 4394,
      "time_per_iteration": 2.647071599960327
    },
    {
      "auxiliary_loss_clip": 0.01191913,
      "auxiliary_loss_mlp": 0.01028861,
      "balance_loss_clip": 1.04992223,
      "balance_loss_mlp": 1.02099919,
      "epoch": 0.5284675043588047,
      "flos": 26760524163840.0,
      "grad_norm": 2.142992167236776,
      "language_loss": 0.74373066,
      "learning_rate": 1.9127828427810693e-06,
      "loss": 0.7659384,
      "num_input_tokens_seen": 94980515,
      "step": 4395,
      "time_per_iteration": 2.715075731277466
    },
    {
      "auxiliary_loss_clip": 0.01339537,
      "auxiliary_loss_mlp": 0.01023563,
      "balance_loss_clip": 1.04886317,
      "balance_loss_mlp": 1.01592731,
      "epoch": 0.5285877472494439,
      "flos": 19899898473600.0,
      "grad_norm": 2.0298545238946333,
      "language_loss": 0.80912155,
      "learning_rate": 1.9120046188096607e-06,
      "loss": 0.83275259,
      "num_input_tokens_seen": 94998560,
      "step": 4396,
      "time_per_iteration": 2.720177173614502
    },
    {
      "auxiliary_loss_clip": 0.01287982,
      "auxiliary_loss_mlp": 0.01031983,
      "balance_loss_clip": 1.05643344,
      "balance_loss_mlp": 1.02422881,
      "epoch": 0.528707990140083,
      "flos": 20011257613440.0,
      "grad_norm": 2.0737285155104286,
      "language_loss": 0.74596846,
      "learning_rate": 1.9112264081870804e-06,
      "loss": 0.76916814,
      "num_input_tokens_seen": 95016950,
      "step": 4397,
      "time_per_iteration": 2.686845064163208
    },
    {
      "auxiliary_loss_clip": 0.0133825,
      "auxiliary_loss_mlp": 0.01029815,
      "balance_loss_clip": 1.05553603,
      "balance_loss_mlp": 1.02197671,
      "epoch": 0.528828233030722,
      "flos": 20667956014080.0,
      "grad_norm": 2.2207256564153735,
      "language_loss": 0.75835323,
      "learning_rate": 1.9104482110313843e-06,
      "loss": 0.78203386,
      "num_input_tokens_seen": 95036540,
      "step": 4398,
      "time_per_iteration": 2.7054479122161865
    },
    {
      "auxiliary_loss_clip": 0.01232492,
      "auxiliary_loss_mlp": 0.0102959,
      "balance_loss_clip": 1.05141401,
      "balance_loss_mlp": 1.02197862,
      "epoch": 0.5289484759213612,
      "flos": 25192448956800.0,
      "grad_norm": 2.0857759582847404,
      "language_loss": 0.74337035,
      "learning_rate": 1.909670027460623e-06,
      "loss": 0.76599121,
      "num_input_tokens_seen": 95053840,
      "step": 4399,
      "time_per_iteration": 2.6722099781036377
    },
    {
      "auxiliary_loss_clip": 0.0123543,
      "auxiliary_loss_mlp": 0.01030242,
      "balance_loss_clip": 1.05461144,
      "balance_loss_mlp": 1.02191544,
      "epoch": 0.5290687188120002,
      "flos": 31139255715840.0,
      "grad_norm": 2.756565584566713,
      "language_loss": 0.71794248,
      "learning_rate": 1.908891857592847e-06,
      "loss": 0.74059916,
      "num_input_tokens_seen": 95074910,
      "step": 4400,
      "time_per_iteration": 3.7423958778381348
    },
    {
      "auxiliary_loss_clip": 0.01333694,
      "auxiliary_loss_mlp": 0.01024018,
      "balance_loss_clip": 1.05411553,
      "balance_loss_mlp": 1.01654959,
      "epoch": 0.5291889617026393,
      "flos": 20119851406080.0,
      "grad_norm": 2.2837454189663857,
      "language_loss": 0.90501851,
      "learning_rate": 1.9081137015461034e-06,
      "loss": 0.9285956,
      "num_input_tokens_seen": 95090985,
      "step": 4401,
      "time_per_iteration": 2.6742448806762695
    },
    {
      "auxiliary_loss_clip": 0.01376058,
      "auxiliary_loss_mlp": 0.01029815,
      "balance_loss_clip": 1.04969597,
      "balance_loss_mlp": 1.02179861,
      "epoch": 0.5293092045932785,
      "flos": 19643747610240.0,
      "grad_norm": 1.9460034158419945,
      "language_loss": 0.90556818,
      "learning_rate": 1.9073355594384383e-06,
      "loss": 0.92962694,
      "num_input_tokens_seen": 95109225,
      "step": 4402,
      "time_per_iteration": 3.6304914951324463
    },
    {
      "auxiliary_loss_clip": 0.01335616,
      "auxiliary_loss_mlp": 0.01028694,
      "balance_loss_clip": 1.05191314,
      "balance_loss_mlp": 1.02092791,
      "epoch": 0.5294294474839175,
      "flos": 24317736958080.0,
      "grad_norm": 1.8481395496933284,
      "language_loss": 0.80487388,
      "learning_rate": 1.906557431387895e-06,
      "loss": 0.82851696,
      "num_input_tokens_seen": 95128215,
      "step": 4403,
      "time_per_iteration": 2.7530767917633057
    },
    {
      "auxiliary_loss_clip": 0.01337377,
      "auxiliary_loss_mlp": 0.01032794,
      "balance_loss_clip": 1.05899882,
      "balance_loss_mlp": 1.02444029,
      "epoch": 0.5295496903745566,
      "flos": 18875941464960.0,
      "grad_norm": 1.8662470317671334,
      "language_loss": 0.78707546,
      "learning_rate": 1.905779317512516e-06,
      "loss": 0.81077719,
      "num_input_tokens_seen": 95145760,
      "step": 4404,
      "time_per_iteration": 2.727649450302124
    },
    {
      "auxiliary_loss_clip": 0.01232257,
      "auxiliary_loss_mlp": 0.01031049,
      "balance_loss_clip": 1.05264139,
      "balance_loss_mlp": 1.02370536,
      "epoch": 0.5296699332651957,
      "flos": 20923101296640.0,
      "grad_norm": 1.993246985336603,
      "language_loss": 0.80904454,
      "learning_rate": 1.9050012179303385e-06,
      "loss": 0.83167756,
      "num_input_tokens_seen": 95164270,
      "step": 4405,
      "time_per_iteration": 3.533539056777954
    },
    {
      "auxiliary_loss_clip": 0.01235257,
      "auxiliary_loss_mlp": 0.01022432,
      "balance_loss_clip": 1.04970407,
      "balance_loss_mlp": 1.01506507,
      "epoch": 0.5297901761558348,
      "flos": 22046745525120.0,
      "grad_norm": 2.3141382399903487,
      "language_loss": 0.68799257,
      "learning_rate": 1.904223132759401e-06,
      "loss": 0.7105695,
      "num_input_tokens_seen": 95182870,
      "step": 4406,
      "time_per_iteration": 2.7328834533691406
    },
    {
      "auxiliary_loss_clip": 0.0123597,
      "auxiliary_loss_mlp": 0.01028032,
      "balance_loss_clip": 1.05316055,
      "balance_loss_mlp": 1.02069497,
      "epoch": 0.5299104190464738,
      "flos": 21798495653760.0,
      "grad_norm": 2.4017844351390214,
      "language_loss": 0.69310069,
      "learning_rate": 1.9034450621177383e-06,
      "loss": 0.71574074,
      "num_input_tokens_seen": 95201190,
      "step": 4407,
      "time_per_iteration": 2.690855026245117
    },
    {
      "auxiliary_loss_clip": 0.0123552,
      "auxiliary_loss_mlp": 0.0102798,
      "balance_loss_clip": 1.05535436,
      "balance_loss_mlp": 1.0204246,
      "epoch": 0.530030661937113,
      "flos": 14720790119040.0,
      "grad_norm": 1.7973449600078506,
      "language_loss": 0.70299506,
      "learning_rate": 1.9026670061233824e-06,
      "loss": 0.72563004,
      "num_input_tokens_seen": 95218625,
      "step": 4408,
      "time_per_iteration": 3.6432478427886963
    },
    {
      "auxiliary_loss_clip": 0.01278348,
      "auxiliary_loss_mlp": 0.01031469,
      "balance_loss_clip": 1.05134296,
      "balance_loss_mlp": 1.02432823,
      "epoch": 0.5301509048277521,
      "flos": 21251504367360.0,
      "grad_norm": 1.8749033949663176,
      "language_loss": 0.80544186,
      "learning_rate": 1.901888964894365e-06,
      "loss": 0.82854003,
      "num_input_tokens_seen": 95237665,
      "step": 4409,
      "time_per_iteration": 2.6385996341705322
    },
    {
      "auxiliary_loss_clip": 0.01187763,
      "auxiliary_loss_mlp": 0.01025454,
      "balance_loss_clip": 1.05444288,
      "balance_loss_mlp": 1.01797998,
      "epoch": 0.5302711477183911,
      "flos": 25957058791680.0,
      "grad_norm": 2.328274124813815,
      "language_loss": 0.67797303,
      "learning_rate": 1.9011109385487134e-06,
      "loss": 0.70010525,
      "num_input_tokens_seen": 95258915,
      "step": 4410,
      "time_per_iteration": 2.7030937671661377
    },
    {
      "auxiliary_loss_clip": 0.01183491,
      "auxiliary_loss_mlp": 0.01026544,
      "balance_loss_clip": 1.05117822,
      "balance_loss_mlp": 1.01869369,
      "epoch": 0.5303913906090303,
      "flos": 22273126992000.0,
      "grad_norm": 3.572836870142444,
      "language_loss": 0.66284347,
      "learning_rate": 1.900332927204454e-06,
      "loss": 0.6849438,
      "num_input_tokens_seen": 95277365,
      "step": 4411,
      "time_per_iteration": 2.644507884979248
    },
    {
      "auxiliary_loss_clip": 0.01192884,
      "auxiliary_loss_mlp": 0.0102694,
      "balance_loss_clip": 1.05240774,
      "balance_loss_mlp": 1.01879764,
      "epoch": 0.5305116334996693,
      "flos": 24936010784640.0,
      "grad_norm": 1.962416289814965,
      "language_loss": 0.77031374,
      "learning_rate": 1.8995549309796097e-06,
      "loss": 0.792512,
      "num_input_tokens_seen": 95296670,
      "step": 4412,
      "time_per_iteration": 2.7139759063720703
    },
    {
      "auxiliary_loss_clip": 0.01142561,
      "auxiliary_loss_mlp": 0.01034981,
      "balance_loss_clip": 1.05387807,
      "balance_loss_mlp": 1.02759576,
      "epoch": 0.5306318763903084,
      "flos": 20189338266240.0,
      "grad_norm": 1.7148704070479583,
      "language_loss": 0.76322293,
      "learning_rate": 1.8987769499922028e-06,
      "loss": 0.78499836,
      "num_input_tokens_seen": 95315640,
      "step": 4413,
      "time_per_iteration": 2.615846633911133
    },
    {
      "auxiliary_loss_clip": 0.01232393,
      "auxiliary_loss_mlp": 0.02569418,
      "balance_loss_clip": 1.05216885,
      "balance_loss_mlp": 1.00033283,
      "epoch": 0.5307521192809476,
      "flos": 20266366982400.0,
      "grad_norm": 2.451933707398109,
      "language_loss": 0.71305704,
      "learning_rate": 1.897998984360252e-06,
      "loss": 0.75107515,
      "num_input_tokens_seen": 95334610,
      "step": 4414,
      "time_per_iteration": 2.644105911254883
    },
    {
      "auxiliary_loss_clip": 0.01282508,
      "auxiliary_loss_mlp": 0.01025881,
      "balance_loss_clip": 1.05069721,
      "balance_loss_mlp": 1.01862097,
      "epoch": 0.5308723621715866,
      "flos": 28844276976000.0,
      "grad_norm": 1.4405899041750962,
      "language_loss": 0.78390163,
      "learning_rate": 1.897221034201775e-06,
      "loss": 0.8069855,
      "num_input_tokens_seen": 95358350,
      "step": 4415,
      "time_per_iteration": 2.7365224361419678
    },
    {
      "auxiliary_loss_clip": 0.01234402,
      "auxiliary_loss_mlp": 0.01028851,
      "balance_loss_clip": 1.04821467,
      "balance_loss_mlp": 1.02211308,
      "epoch": 0.5309926050622257,
      "flos": 27457766040960.0,
      "grad_norm": 1.6650359898731233,
      "language_loss": 0.6668098,
      "learning_rate": 1.8964430996347842e-06,
      "loss": 0.68944228,
      "num_input_tokens_seen": 95379900,
      "step": 4416,
      "time_per_iteration": 2.781513214111328
    },
    {
      "auxiliary_loss_clip": 0.0128032,
      "auxiliary_loss_mlp": 0.0102768,
      "balance_loss_clip": 1.04844666,
      "balance_loss_mlp": 1.01982367,
      "epoch": 0.5311128479528648,
      "flos": 20514545026560.0,
      "grad_norm": 2.5482128225261467,
      "language_loss": 0.82303858,
      "learning_rate": 1.8956651807772931e-06,
      "loss": 0.84611857,
      "num_input_tokens_seen": 95397935,
      "step": 4417,
      "time_per_iteration": 2.63783860206604
    },
    {
      "auxiliary_loss_clip": 0.0123194,
      "auxiliary_loss_mlp": 0.01027937,
      "balance_loss_clip": 1.05372298,
      "balance_loss_mlp": 1.02066541,
      "epoch": 0.5312330908435039,
      "flos": 21397660807680.0,
      "grad_norm": 1.8349017987870098,
      "language_loss": 0.83970398,
      "learning_rate": 1.8948872777473115e-06,
      "loss": 0.86230278,
      "num_input_tokens_seen": 95415890,
      "step": 4418,
      "time_per_iteration": 2.6132283210754395
    },
    {
      "auxiliary_loss_clip": 0.0128452,
      "auxiliary_loss_mlp": 0.01026621,
      "balance_loss_clip": 1.05284858,
      "balance_loss_mlp": 1.01933193,
      "epoch": 0.531353333734143,
      "flos": 24717350741760.0,
      "grad_norm": 2.1121984100085873,
      "language_loss": 0.63418818,
      "learning_rate": 1.8941093906628458e-06,
      "loss": 0.65729952,
      "num_input_tokens_seen": 95433675,
      "step": 4419,
      "time_per_iteration": 2.7168147563934326
    },
    {
      "auxiliary_loss_clip": 0.01274532,
      "auxiliary_loss_mlp": 0.01029318,
      "balance_loss_clip": 1.04712641,
      "balance_loss_mlp": 1.02173066,
      "epoch": 0.531473576624782,
      "flos": 30480689808000.0,
      "grad_norm": 1.84051968648129,
      "language_loss": 0.71102995,
      "learning_rate": 1.893331519641902e-06,
      "loss": 0.73406845,
      "num_input_tokens_seen": 95455820,
      "step": 4420,
      "time_per_iteration": 2.7308335304260254
    },
    {
      "auxiliary_loss_clip": 0.01324801,
      "auxiliary_loss_mlp": 0.01026443,
      "balance_loss_clip": 1.04425144,
      "balance_loss_mlp": 1.01917171,
      "epoch": 0.5315938195154212,
      "flos": 23002975440000.0,
      "grad_norm": 2.4684487689074937,
      "language_loss": 0.73930788,
      "learning_rate": 1.8925536648024815e-06,
      "loss": 0.76282024,
      "num_input_tokens_seen": 95473240,
      "step": 4421,
      "time_per_iteration": 2.733017921447754
    },
    {
      "auxiliary_loss_clip": 0.01185747,
      "auxiliary_loss_mlp": 0.0102654,
      "balance_loss_clip": 1.05382848,
      "balance_loss_mlp": 1.01942956,
      "epoch": 0.5317140624060602,
      "flos": 22748584343040.0,
      "grad_norm": 2.8530276167130597,
      "language_loss": 0.75926602,
      "learning_rate": 1.8917758262625849e-06,
      "loss": 0.78138888,
      "num_input_tokens_seen": 95493480,
      "step": 4422,
      "time_per_iteration": 2.614306926727295
    },
    {
      "auxiliary_loss_clip": 0.01275069,
      "auxiliary_loss_mlp": 0.0102761,
      "balance_loss_clip": 1.05066037,
      "balance_loss_mlp": 1.02074623,
      "epoch": 0.5318343052966993,
      "flos": 22821087945600.0,
      "grad_norm": 1.7179753305538454,
      "language_loss": 0.80912864,
      "learning_rate": 1.8909980041402089e-06,
      "loss": 0.83215541,
      "num_input_tokens_seen": 95512075,
      "step": 4423,
      "time_per_iteration": 2.713853120803833
    },
    {
      "auxiliary_loss_clip": 0.01231037,
      "auxiliary_loss_mlp": 0.0102744,
      "balance_loss_clip": 1.05002475,
      "balance_loss_mlp": 1.01990294,
      "epoch": 0.5319545481873384,
      "flos": 13626089274240.0,
      "grad_norm": 2.954552652520029,
      "language_loss": 0.65331769,
      "learning_rate": 1.8902201985533494e-06,
      "loss": 0.67590249,
      "num_input_tokens_seen": 95529340,
      "step": 4424,
      "time_per_iteration": 2.5758488178253174
    },
    {
      "auxiliary_loss_clip": 0.01279161,
      "auxiliary_loss_mlp": 0.01024179,
      "balance_loss_clip": 1.05034637,
      "balance_loss_mlp": 1.01658547,
      "epoch": 0.5320747910779775,
      "flos": 22162522037760.0,
      "grad_norm": 1.7278943971376004,
      "language_loss": 0.74747902,
      "learning_rate": 1.8894424096199983e-06,
      "loss": 0.77051246,
      "num_input_tokens_seen": 95548545,
      "step": 4425,
      "time_per_iteration": 2.676978349685669
    },
    {
      "auxiliary_loss_clip": 0.01233542,
      "auxiliary_loss_mlp": 0.01027911,
      "balance_loss_clip": 1.05353498,
      "balance_loss_mlp": 1.02005553,
      "epoch": 0.5321950339686166,
      "flos": 18588081870720.0,
      "grad_norm": 2.3582776413376503,
      "language_loss": 0.85971999,
      "learning_rate": 1.8886646374581463e-06,
      "loss": 0.88233453,
      "num_input_tokens_seen": 95567770,
      "step": 4426,
      "time_per_iteration": 2.5816051959991455
    },
    {
      "auxiliary_loss_clip": 0.01232612,
      "auxiliary_loss_mlp": 0.0102908,
      "balance_loss_clip": 1.04971552,
      "balance_loss_mlp": 1.02184439,
      "epoch": 0.5323152768592557,
      "flos": 22856818999680.0,
      "grad_norm": 1.7382575540923493,
      "language_loss": 0.71058547,
      "learning_rate": 1.8878868821857795e-06,
      "loss": 0.73320246,
      "num_input_tokens_seen": 95587420,
      "step": 4427,
      "time_per_iteration": 3.639629602432251
    },
    {
      "auxiliary_loss_clip": 0.0138912,
      "auxiliary_loss_mlp": 0.01030532,
      "balance_loss_clip": 1.04870915,
      "balance_loss_mlp": 1.02187717,
      "epoch": 0.5324355197498948,
      "flos": 33948690998400.0,
      "grad_norm": 5.9573634477542345,
      "language_loss": 0.7541123,
      "learning_rate": 1.8871091439208838e-06,
      "loss": 0.77830881,
      "num_input_tokens_seen": 95609030,
      "step": 4428,
      "time_per_iteration": 2.8746516704559326
    },
    {
      "auxiliary_loss_clip": 0.01379223,
      "auxiliary_loss_mlp": 0.01033514,
      "balance_loss_clip": 1.04623222,
      "balance_loss_mlp": 1.02537179,
      "epoch": 0.5325557626405338,
      "flos": 23256720092160.0,
      "grad_norm": 2.0643864885502,
      "language_loss": 0.77679288,
      "learning_rate": 1.8863314227814414e-06,
      "loss": 0.80092019,
      "num_input_tokens_seen": 95627340,
      "step": 4429,
      "time_per_iteration": 3.6129109859466553
    },
    {
      "auxiliary_loss_clip": 0.01240028,
      "auxiliary_loss_mlp": 0.0103095,
      "balance_loss_clip": 1.05242968,
      "balance_loss_mlp": 1.02312934,
      "epoch": 0.532676005531173,
      "flos": 26718687797760.0,
      "grad_norm": 2.4008873478968735,
      "language_loss": 0.49043664,
      "learning_rate": 1.8855537188854313e-06,
      "loss": 0.5131464,
      "num_input_tokens_seen": 95646315,
      "step": 4430,
      "time_per_iteration": 2.668762445449829
    },
    {
      "auxiliary_loss_clip": 0.01234486,
      "auxiliary_loss_mlp": 0.01024015,
      "balance_loss_clip": 1.04832506,
      "balance_loss_mlp": 1.01664233,
      "epoch": 0.5327962484218121,
      "flos": 17894610921600.0,
      "grad_norm": 2.237363689320833,
      "language_loss": 0.78353643,
      "learning_rate": 1.8847760323508315e-06,
      "loss": 0.80612135,
      "num_input_tokens_seen": 95665220,
      "step": 4431,
      "time_per_iteration": 3.500570774078369
    },
    {
      "auxiliary_loss_clip": 0.01273079,
      "auxiliary_loss_mlp": 0.01033984,
      "balance_loss_clip": 1.04977298,
      "balance_loss_mlp": 1.0269084,
      "epoch": 0.5329164913124511,
      "flos": 17925385898880.0,
      "grad_norm": 2.095234196097736,
      "language_loss": 0.75783962,
      "learning_rate": 1.883998363295616e-06,
      "loss": 0.78091025,
      "num_input_tokens_seen": 95682700,
      "step": 4432,
      "time_per_iteration": 2.6286723613739014
    },
    {
      "auxiliary_loss_clip": 0.01177983,
      "auxiliary_loss_mlp": 0.01000131,
      "balance_loss_clip": 1.00955594,
      "balance_loss_mlp": 0.99929088,
      "epoch": 0.5330367342030903,
      "flos": 57254178781440.0,
      "grad_norm": 0.8799564695616859,
      "language_loss": 0.62592113,
      "learning_rate": 1.8832207118377565e-06,
      "loss": 0.64770222,
      "num_input_tokens_seen": 95738070,
      "step": 4433,
      "time_per_iteration": 3.1451516151428223
    },
    {
      "auxiliary_loss_clip": 0.01181565,
      "auxiliary_loss_mlp": 0.01025175,
      "balance_loss_clip": 1.05279422,
      "balance_loss_mlp": 1.01811218,
      "epoch": 0.5331569770937293,
      "flos": 17420518287360.0,
      "grad_norm": 2.168248543413643,
      "language_loss": 0.69501495,
      "learning_rate": 1.882443078095222e-06,
      "loss": 0.71708238,
      "num_input_tokens_seen": 95756950,
      "step": 4434,
      "time_per_iteration": 3.477083206176758
    },
    {
      "auxiliary_loss_clip": 0.01285994,
      "auxiliary_loss_mlp": 0.00998741,
      "balance_loss_clip": 1.01296687,
      "balance_loss_mlp": 0.99771559,
      "epoch": 0.5332772199843684,
      "flos": 56750783627520.0,
      "grad_norm": 0.8679342972089867,
      "language_loss": 0.66727257,
      "learning_rate": 1.8816654621859794e-06,
      "loss": 0.69011986,
      "num_input_tokens_seen": 95816615,
      "step": 4435,
      "time_per_iteration": 3.150754928588867
    },
    {
      "auxiliary_loss_clip": 0.01181802,
      "auxiliary_loss_mlp": 0.01026896,
      "balance_loss_clip": 1.05259717,
      "balance_loss_mlp": 1.01951146,
      "epoch": 0.5333974628750076,
      "flos": 18697753071360.0,
      "grad_norm": 2.9920142596969486,
      "language_loss": 0.72616208,
      "learning_rate": 1.8808878642279915e-06,
      "loss": 0.74824899,
      "num_input_tokens_seen": 95832020,
      "step": 4436,
      "time_per_iteration": 2.6199655532836914
    },
    {
      "auxiliary_loss_clip": 0.0133444,
      "auxiliary_loss_mlp": 0.01029574,
      "balance_loss_clip": 1.04364347,
      "balance_loss_mlp": 1.02204013,
      "epoch": 0.5335177057656466,
      "flos": 23805507058560.0,
      "grad_norm": 2.383317645569247,
      "language_loss": 0.6529274,
      "learning_rate": 1.8801102843392209e-06,
      "loss": 0.67656755,
      "num_input_tokens_seen": 95851425,
      "step": 4437,
      "time_per_iteration": 2.692514181137085
    },
    {
      "auxiliary_loss_clip": 0.01327756,
      "auxiliary_loss_mlp": 0.01032382,
      "balance_loss_clip": 1.04587352,
      "balance_loss_mlp": 1.02479768,
      "epoch": 0.5336379486562857,
      "flos": 25078683605760.0,
      "grad_norm": 1.648635795586077,
      "language_loss": 0.85297298,
      "learning_rate": 1.8793327226376238e-06,
      "loss": 0.8765744,
      "num_input_tokens_seen": 95870745,
      "step": 4438,
      "time_per_iteration": 2.7795395851135254
    },
    {
      "auxiliary_loss_clip": 0.01190682,
      "auxiliary_loss_mlp": 0.01028628,
      "balance_loss_clip": 1.04843211,
      "balance_loss_mlp": 1.02111435,
      "epoch": 0.5337581915469248,
      "flos": 21396691140480.0,
      "grad_norm": 1.840081934934768,
      "language_loss": 0.80241752,
      "learning_rate": 1.8785551792411569e-06,
      "loss": 0.82461059,
      "num_input_tokens_seen": 95889755,
      "step": 4439,
      "time_per_iteration": 2.646946907043457
    },
    {
      "auxiliary_loss_clip": 0.01283101,
      "auxiliary_loss_mlp": 0.01029977,
      "balance_loss_clip": 1.05275893,
      "balance_loss_mlp": 1.02329564,
      "epoch": 0.5338784344375639,
      "flos": 14865905064960.0,
      "grad_norm": 2.2743884325532417,
      "language_loss": 0.82838506,
      "learning_rate": 1.8777776542677733e-06,
      "loss": 0.85151589,
      "num_input_tokens_seen": 95907805,
      "step": 4440,
      "time_per_iteration": 2.664463520050049
    },
    {
      "auxiliary_loss_clip": 0.01329824,
      "auxiliary_loss_mlp": 0.01024842,
      "balance_loss_clip": 1.04469323,
      "balance_loss_mlp": 1.01752281,
      "epoch": 0.5339986773282029,
      "flos": 20813501923200.0,
      "grad_norm": 1.8779537978569611,
      "language_loss": 0.73376364,
      "learning_rate": 1.8770001478354216e-06,
      "loss": 0.75731027,
      "num_input_tokens_seen": 95927480,
      "step": 4441,
      "time_per_iteration": 2.6752665042877197
    },
    {
      "auxiliary_loss_clip": 0.0123362,
      "auxiliary_loss_mlp": 0.01031153,
      "balance_loss_clip": 1.05058801,
      "balance_loss_mlp": 1.02333295,
      "epoch": 0.5341189202188421,
      "flos": 17969089772160.0,
      "grad_norm": 2.1510530917278454,
      "language_loss": 0.83798563,
      "learning_rate": 1.8762226600620504e-06,
      "loss": 0.86063337,
      "num_input_tokens_seen": 95946095,
      "step": 4442,
      "time_per_iteration": 2.637486219406128
    },
    {
      "auxiliary_loss_clip": 0.01291935,
      "auxiliary_loss_mlp": 0.01027906,
      "balance_loss_clip": 1.04950392,
      "balance_loss_mlp": 1.01974034,
      "epoch": 0.5342391631094812,
      "flos": 11031866328960.0,
      "grad_norm": 2.396025824633304,
      "language_loss": 0.58787549,
      "learning_rate": 1.8754451910656031e-06,
      "loss": 0.61107385,
      "num_input_tokens_seen": 95959995,
      "step": 4443,
      "time_per_iteration": 2.5972797870635986
    },
    {
      "auxiliary_loss_clip": 0.01386649,
      "auxiliary_loss_mlp": 0.01030986,
      "balance_loss_clip": 1.04778862,
      "balance_loss_mlp": 1.02377415,
      "epoch": 0.5343594060001202,
      "flos": 15339135772800.0,
      "grad_norm": 1.8201204011138943,
      "language_loss": 0.8263858,
      "learning_rate": 1.8746677409640212e-06,
      "loss": 0.85056221,
      "num_input_tokens_seen": 95977095,
      "step": 4444,
      "time_per_iteration": 2.8030319213867188
    },
    {
      "auxiliary_loss_clip": 0.01240782,
      "auxiliary_loss_mlp": 0.01026178,
      "balance_loss_clip": 1.05538774,
      "balance_loss_mlp": 1.01889408,
      "epoch": 0.5344796488907594,
      "flos": 26900898514560.0,
      "grad_norm": 1.877989119785938,
      "language_loss": 0.84568352,
      "learning_rate": 1.8738903098752432e-06,
      "loss": 0.86835313,
      "num_input_tokens_seen": 95996225,
      "step": 4445,
      "time_per_iteration": 2.665954351425171
    },
    {
      "auxiliary_loss_clip": 0.01285127,
      "auxiliary_loss_mlp": 0.01026512,
      "balance_loss_clip": 1.04944932,
      "balance_loss_mlp": 1.0190165,
      "epoch": 0.5345998917813984,
      "flos": 25411216740480.0,
      "grad_norm": 2.2100647932407593,
      "language_loss": 0.73337996,
      "learning_rate": 1.8731128979172052e-06,
      "loss": 0.75649631,
      "num_input_tokens_seen": 96015425,
      "step": 4446,
      "time_per_iteration": 2.732459783554077
    },
    {
      "auxiliary_loss_clip": 0.01277291,
      "auxiliary_loss_mlp": 0.01023585,
      "balance_loss_clip": 1.04982626,
      "balance_loss_mlp": 1.01608062,
      "epoch": 0.5347201346720375,
      "flos": 32853379622400.0,
      "grad_norm": 2.2684301209695485,
      "language_loss": 0.67189038,
      "learning_rate": 1.8723355052078394e-06,
      "loss": 0.69489914,
      "num_input_tokens_seen": 96035460,
      "step": 4447,
      "time_per_iteration": 2.7272303104400635
    },
    {
      "auxiliary_loss_clip": 0.01235709,
      "auxiliary_loss_mlp": 0.01030463,
      "balance_loss_clip": 1.05191898,
      "balance_loss_mlp": 1.02230346,
      "epoch": 0.5348403775626767,
      "flos": 17967940536960.0,
      "grad_norm": 2.5287701583912825,
      "language_loss": 0.77675337,
      "learning_rate": 1.8715581318650765e-06,
      "loss": 0.79941511,
      "num_input_tokens_seen": 96054515,
      "step": 4448,
      "time_per_iteration": 2.6476898193359375
    },
    {
      "auxiliary_loss_clip": 0.01340867,
      "auxiliary_loss_mlp": 0.01033898,
      "balance_loss_clip": 1.04978836,
      "balance_loss_mlp": 1.0256424,
      "epoch": 0.5349606204533157,
      "flos": 17603339535360.0,
      "grad_norm": 3.3293249875912116,
      "language_loss": 0.82055134,
      "learning_rate": 1.8707807780068422e-06,
      "loss": 0.84429896,
      "num_input_tokens_seen": 96072330,
      "step": 4449,
      "time_per_iteration": 2.653832197189331
    },
    {
      "auxiliary_loss_clip": 0.0128224,
      "auxiliary_loss_mlp": 0.01028461,
      "balance_loss_clip": 1.05048192,
      "balance_loss_mlp": 1.02096534,
      "epoch": 0.5350808633439548,
      "flos": 29167831710720.0,
      "grad_norm": 2.458454015749908,
      "language_loss": 0.66511816,
      "learning_rate": 1.8700034437510611e-06,
      "loss": 0.68822521,
      "num_input_tokens_seen": 96092425,
      "step": 4450,
      "time_per_iteration": 2.6986446380615234
    },
    {
      "auxiliary_loss_clip": 0.01329006,
      "auxiliary_loss_mlp": 0.01030594,
      "balance_loss_clip": 1.04759598,
      "balance_loss_mlp": 1.02304244,
      "epoch": 0.5352011062345938,
      "flos": 19499997381120.0,
      "grad_norm": 2.4758487427762654,
      "language_loss": 0.81562459,
      "learning_rate": 1.8692261292156549e-06,
      "loss": 0.83922064,
      "num_input_tokens_seen": 96111660,
      "step": 4451,
      "time_per_iteration": 2.7031590938568115
    },
    {
      "auxiliary_loss_clip": 0.01182278,
      "auxiliary_loss_mlp": 0.01026737,
      "balance_loss_clip": 1.05501032,
      "balance_loss_mlp": 1.01920843,
      "epoch": 0.535321349125233,
      "flos": 23477642691840.0,
      "grad_norm": 2.0952953375274985,
      "language_loss": 0.81372982,
      "learning_rate": 1.8684488345185401e-06,
      "loss": 0.83581996,
      "num_input_tokens_seen": 96131835,
      "step": 4452,
      "time_per_iteration": 3.531764507293701
    },
    {
      "auxiliary_loss_clip": 0.0118612,
      "auxiliary_loss_mlp": 0.01028308,
      "balance_loss_clip": 1.05406499,
      "balance_loss_mlp": 1.02067327,
      "epoch": 0.535441592015872,
      "flos": 20478059786880.0,
      "grad_norm": 2.651664230588639,
      "language_loss": 0.79256016,
      "learning_rate": 1.8676715597776332e-06,
      "loss": 0.81470442,
      "num_input_tokens_seen": 96150180,
      "step": 4453,
      "time_per_iteration": 2.5915939807891846
    },
    {
      "auxiliary_loss_clip": 0.01369431,
      "auxiliary_loss_mlp": 0.0102492,
      "balance_loss_clip": 1.04292798,
      "balance_loss_mlp": 1.01729608,
      "epoch": 0.5355618349065111,
      "flos": 19573147428480.0,
      "grad_norm": 2.2575438676725015,
      "language_loss": 0.76202881,
      "learning_rate": 1.8668943051108455e-06,
      "loss": 0.78597224,
      "num_input_tokens_seen": 96167485,
      "step": 4454,
      "time_per_iteration": 2.7119991779327393
    },
    {
      "auxiliary_loss_clip": 0.01283186,
      "auxiliary_loss_mlp": 0.01026985,
      "balance_loss_clip": 1.04986429,
      "balance_loss_mlp": 1.0189085,
      "epoch": 0.5356820777971503,
      "flos": 24024633978240.0,
      "grad_norm": 2.002130354135141,
      "language_loss": 0.76859021,
      "learning_rate": 1.8661170706360856e-06,
      "loss": 0.7916919,
      "num_input_tokens_seen": 96186650,
      "step": 4455,
      "time_per_iteration": 3.665884256362915
    },
    {
      "auxiliary_loss_clip": 0.01227541,
      "auxiliary_loss_mlp": 0.01027977,
      "balance_loss_clip": 1.05181181,
      "balance_loss_mlp": 1.0208962,
      "epoch": 0.5358023206877893,
      "flos": 20884676722560.0,
      "grad_norm": 2.0759518817941722,
      "language_loss": 0.81333995,
      "learning_rate": 1.8653398564712594e-06,
      "loss": 0.83589518,
      "num_input_tokens_seen": 96205595,
      "step": 4456,
      "time_per_iteration": 2.68106746673584
    },
    {
      "auxiliary_loss_clip": 0.0123023,
      "auxiliary_loss_mlp": 0.01027946,
      "balance_loss_clip": 1.05139089,
      "balance_loss_mlp": 1.02055204,
      "epoch": 0.5359225635784284,
      "flos": 22418996123520.0,
      "grad_norm": 2.088769941457838,
      "language_loss": 0.82364392,
      "learning_rate": 1.8645626627342704e-06,
      "loss": 0.84622562,
      "num_input_tokens_seen": 96226360,
      "step": 4457,
      "time_per_iteration": 3.5616838932037354
    },
    {
      "auxiliary_loss_clip": 0.01237639,
      "auxiliary_loss_mlp": 0.01026638,
      "balance_loss_clip": 1.05225742,
      "balance_loss_mlp": 1.01932406,
      "epoch": 0.5360428064690675,
      "flos": 24097784025600.0,
      "grad_norm": 2.108636518134639,
      "language_loss": 0.81347299,
      "learning_rate": 1.8637854895430172e-06,
      "loss": 0.83611572,
      "num_input_tokens_seen": 96245625,
      "step": 4458,
      "time_per_iteration": 2.68507719039917
    },
    {
      "auxiliary_loss_clip": 0.0133553,
      "auxiliary_loss_mlp": 0.01030743,
      "balance_loss_clip": 1.05080569,
      "balance_loss_mlp": 1.02319694,
      "epoch": 0.5361630493597066,
      "flos": 21434505183360.0,
      "grad_norm": 2.308592280950981,
      "language_loss": 0.69802582,
      "learning_rate": 1.8630083370153978e-06,
      "loss": 0.72168857,
      "num_input_tokens_seen": 96265265,
      "step": 4459,
      "time_per_iteration": 2.680406093597412
    },
    {
      "auxiliary_loss_clip": 0.01340701,
      "auxiliary_loss_mlp": 0.00999757,
      "balance_loss_clip": 1.00906229,
      "balance_loss_mlp": 0.99872553,
      "epoch": 0.5362832922503457,
      "flos": 68888696520960.0,
      "grad_norm": 0.7424601735233134,
      "language_loss": 0.55357742,
      "learning_rate": 1.8622312052693041e-06,
      "loss": 0.57698202,
      "num_input_tokens_seen": 96326445,
      "step": 4460,
      "time_per_iteration": 4.30298924446106
    },
    {
      "auxiliary_loss_clip": 0.01228452,
      "auxiliary_loss_mlp": 0.01030988,
      "balance_loss_clip": 1.04871798,
      "balance_loss_mlp": 1.02350187,
      "epoch": 0.5364035351409848,
      "flos": 9793702563840.0,
      "grad_norm": 2.163201669690553,
      "language_loss": 0.72058737,
      "learning_rate": 1.8614540944226267e-06,
      "loss": 0.74318182,
      "num_input_tokens_seen": 96343115,
      "step": 4461,
      "time_per_iteration": 2.7324352264404297
    },
    {
      "auxiliary_loss_clip": 0.01280466,
      "auxiliary_loss_mlp": 0.01030302,
      "balance_loss_clip": 1.05351055,
      "balance_loss_mlp": 1.02326608,
      "epoch": 0.5365237780316239,
      "flos": 23290080848640.0,
      "grad_norm": 1.8662429531375802,
      "language_loss": 0.68007171,
      "learning_rate": 1.8606770045932537e-06,
      "loss": 0.70317942,
      "num_input_tokens_seen": 96362230,
      "step": 4462,
      "time_per_iteration": 2.6839284896850586
    },
    {
      "auxiliary_loss_clip": 0.0133058,
      "auxiliary_loss_mlp": 0.01029185,
      "balance_loss_clip": 1.04447007,
      "balance_loss_mlp": 1.02108169,
      "epoch": 0.5366440209222629,
      "flos": 26578133879040.0,
      "grad_norm": 1.8051391264652918,
      "language_loss": 0.81611609,
      "learning_rate": 1.859899935899068e-06,
      "loss": 0.83971369,
      "num_input_tokens_seen": 96382085,
      "step": 4463,
      "time_per_iteration": 2.7336862087249756
    },
    {
      "auxiliary_loss_clip": 0.01286476,
      "auxiliary_loss_mlp": 0.01024877,
      "balance_loss_clip": 1.05569458,
      "balance_loss_mlp": 1.01722074,
      "epoch": 0.5367642638129021,
      "flos": 19608052469760.0,
      "grad_norm": 1.668052983023667,
      "language_loss": 0.79300559,
      "learning_rate": 1.8591228884579506e-06,
      "loss": 0.81611907,
      "num_input_tokens_seen": 96400580,
      "step": 4464,
      "time_per_iteration": 2.621901512145996
    },
    {
      "auxiliary_loss_clip": 0.01334533,
      "auxiliary_loss_mlp": 0.01032528,
      "balance_loss_clip": 1.0492481,
      "balance_loss_mlp": 1.02501798,
      "epoch": 0.5368845067035412,
      "flos": 23915214172800.0,
      "grad_norm": 3.1392902232799496,
      "language_loss": 0.82464683,
      "learning_rate": 1.8583458623877795e-06,
      "loss": 0.8483175,
      "num_input_tokens_seen": 96419680,
      "step": 4465,
      "time_per_iteration": 2.7451260089874268
    },
    {
      "auxiliary_loss_clip": 0.01236215,
      "auxiliary_loss_mlp": 0.01031846,
      "balance_loss_clip": 1.0531311,
      "balance_loss_mlp": 1.02429414,
      "epoch": 0.5370047495941802,
      "flos": 16873131951360.0,
      "grad_norm": 1.9977581024357034,
      "language_loss": 0.74183148,
      "learning_rate": 1.8575688578064281e-06,
      "loss": 0.76451206,
      "num_input_tokens_seen": 96437805,
      "step": 4466,
      "time_per_iteration": 2.5836660861968994
    },
    {
      "auxiliary_loss_clip": 0.01238193,
      "auxiliary_loss_mlp": 0.01027027,
      "balance_loss_clip": 1.05459416,
      "balance_loss_mlp": 1.01903391,
      "epoch": 0.5371249924848194,
      "flos": 20740926493440.0,
      "grad_norm": 1.8722604868300439,
      "language_loss": 0.76916516,
      "learning_rate": 1.8567918748317674e-06,
      "loss": 0.79181737,
      "num_input_tokens_seen": 96457155,
      "step": 4467,
      "time_per_iteration": 2.6660170555114746
    },
    {
      "auxiliary_loss_clip": 0.01334079,
      "auxiliary_loss_mlp": 0.01035683,
      "balance_loss_clip": 1.04629922,
      "balance_loss_mlp": 1.02747583,
      "epoch": 0.5372452353754584,
      "flos": 17968120104960.0,
      "grad_norm": 1.8216586885843669,
      "language_loss": 0.82725358,
      "learning_rate": 1.8560149135816659e-06,
      "loss": 0.85095125,
      "num_input_tokens_seen": 96473990,
      "step": 4468,
      "time_per_iteration": 2.6919517517089844
    },
    {
      "auxiliary_loss_clip": 0.0123092,
      "auxiliary_loss_mlp": 0.0102783,
      "balance_loss_clip": 1.04974604,
      "balance_loss_mlp": 1.02005792,
      "epoch": 0.5373654782660975,
      "flos": 15377021642880.0,
      "grad_norm": 2.271618752112108,
      "language_loss": 0.84555787,
      "learning_rate": 1.8552379741739873e-06,
      "loss": 0.86814541,
      "num_input_tokens_seen": 96491335,
      "step": 4469,
      "time_per_iteration": 2.578207015991211
    },
    {
      "auxiliary_loss_clip": 0.01232019,
      "auxiliary_loss_mlp": 0.02507041,
      "balance_loss_clip": 1.01211643,
      "balance_loss_mlp": 1.00014198,
      "epoch": 0.5374857211567367,
      "flos": 69000091574400.0,
      "grad_norm": 0.8953746917126124,
      "language_loss": 0.55594438,
      "learning_rate": 1.8544610567265935e-06,
      "loss": 0.59333503,
      "num_input_tokens_seen": 96545275,
      "step": 4470,
      "time_per_iteration": 3.2288105487823486
    },
    {
      "auxiliary_loss_clip": 0.01280468,
      "auxiliary_loss_mlp": 0.0256823,
      "balance_loss_clip": 1.05268359,
      "balance_loss_mlp": 1.00017738,
      "epoch": 0.5376059640473757,
      "flos": 15085355207040.0,
      "grad_norm": 1.9516414112150426,
      "language_loss": 0.83214235,
      "learning_rate": 1.853684161357341e-06,
      "loss": 0.87062931,
      "num_input_tokens_seen": 96562935,
      "step": 4471,
      "time_per_iteration": 2.6232361793518066
    },
    {
      "auxiliary_loss_clip": 0.01234006,
      "auxiliary_loss_mlp": 0.02571642,
      "balance_loss_clip": 1.05478323,
      "balance_loss_mlp": 1.00018191,
      "epoch": 0.5377262069380148,
      "flos": 19792597570560.0,
      "grad_norm": 1.744406824581951,
      "language_loss": 0.77215898,
      "learning_rate": 1.852907288184085e-06,
      "loss": 0.81021547,
      "num_input_tokens_seen": 96581820,
      "step": 4472,
      "time_per_iteration": 2.6645655632019043
    },
    {
      "auxiliary_loss_clip": 0.01393296,
      "auxiliary_loss_mlp": 0.01025884,
      "balance_loss_clip": 1.05339622,
      "balance_loss_mlp": 1.01808476,
      "epoch": 0.5378464498286539,
      "flos": 30003077640960.0,
      "grad_norm": 1.9827958682681817,
      "language_loss": 0.70177197,
      "learning_rate": 1.8521304373246762e-06,
      "loss": 0.72596377,
      "num_input_tokens_seen": 96602865,
      "step": 4473,
      "time_per_iteration": 2.7705183029174805
    },
    {
      "auxiliary_loss_clip": 0.01237346,
      "auxiliary_loss_mlp": 0.01031154,
      "balance_loss_clip": 1.05104136,
      "balance_loss_mlp": 1.0230242,
      "epoch": 0.537966692719293,
      "flos": 21251217058560.0,
      "grad_norm": 3.9830663156830726,
      "language_loss": 0.89168549,
      "learning_rate": 1.8513536088969626e-06,
      "loss": 0.91437054,
      "num_input_tokens_seen": 96620530,
      "step": 4474,
      "time_per_iteration": 2.6913797855377197
    },
    {
      "auxiliary_loss_clip": 0.0123796,
      "auxiliary_loss_mlp": 0.01028596,
      "balance_loss_clip": 1.05624938,
      "balance_loss_mlp": 1.02066278,
      "epoch": 0.538086935609932,
      "flos": 21543170803200.0,
      "grad_norm": 1.8888907137139854,
      "language_loss": 0.80153477,
      "learning_rate": 1.8505768030187884e-06,
      "loss": 0.82420039,
      "num_input_tokens_seen": 96640660,
      "step": 4475,
      "time_per_iteration": 2.611132860183716
    },
    {
      "auxiliary_loss_clip": 0.01280675,
      "auxiliary_loss_mlp": 0.01032339,
      "balance_loss_clip": 1.05290806,
      "balance_loss_mlp": 1.02469754,
      "epoch": 0.5382071785005712,
      "flos": 22747219626240.0,
      "grad_norm": 1.6122809789026693,
      "language_loss": 0.79985315,
      "learning_rate": 1.849800019807995e-06,
      "loss": 0.82298326,
      "num_input_tokens_seen": 96661885,
      "step": 4476,
      "time_per_iteration": 2.760871171951294
    },
    {
      "auxiliary_loss_clip": 0.01333229,
      "auxiliary_loss_mlp": 0.01026469,
      "balance_loss_clip": 1.05125594,
      "balance_loss_mlp": 1.01845253,
      "epoch": 0.5383274213912103,
      "flos": 24934574240640.0,
      "grad_norm": 2.5351578031954416,
      "language_loss": 0.71024728,
      "learning_rate": 1.8490232593824186e-06,
      "loss": 0.73384428,
      "num_input_tokens_seen": 96678340,
      "step": 4477,
      "time_per_iteration": 2.666750192642212
    },
    {
      "auxiliary_loss_clip": 0.01277838,
      "auxiliary_loss_mlp": 0.0102779,
      "balance_loss_clip": 1.05238307,
      "balance_loss_mlp": 1.02010095,
      "epoch": 0.5384476642818493,
      "flos": 22310186849280.0,
      "grad_norm": 2.1199186884475947,
      "language_loss": 0.84909761,
      "learning_rate": 1.8482465218598935e-06,
      "loss": 0.87215388,
      "num_input_tokens_seen": 96698285,
      "step": 4478,
      "time_per_iteration": 3.93097186088562
    },
    {
      "auxiliary_loss_clip": 0.01335415,
      "auxiliary_loss_mlp": 0.01033104,
      "balance_loss_clip": 1.04879832,
      "balance_loss_mlp": 1.02499807,
      "epoch": 0.5385679071724885,
      "flos": 22711021695360.0,
      "grad_norm": 6.8765341992454445,
      "language_loss": 0.83740377,
      "learning_rate": 1.8474698073582508e-06,
      "loss": 0.86108899,
      "num_input_tokens_seen": 96719655,
      "step": 4479,
      "time_per_iteration": 2.7604830265045166
    },
    {
      "auxiliary_loss_clip": 0.01335198,
      "auxiliary_loss_mlp": 0.01028712,
      "balance_loss_clip": 1.04781747,
      "balance_loss_mlp": 1.02097583,
      "epoch": 0.5386881500631275,
      "flos": 15953746412160.0,
      "grad_norm": 84.01387600767887,
      "language_loss": 0.87722957,
      "learning_rate": 1.8466931159953166e-06,
      "loss": 0.90086871,
      "num_input_tokens_seen": 96736290,
      "step": 4480,
      "time_per_iteration": 2.697531223297119
    },
    {
      "auxiliary_loss_clip": 0.01286567,
      "auxiliary_loss_mlp": 0.01034512,
      "balance_loss_clip": 1.05371952,
      "balance_loss_mlp": 1.02592289,
      "epoch": 0.5388083929537666,
      "flos": 24060041809920.0,
      "grad_norm": 1.7187021731527345,
      "language_loss": 0.84615839,
      "learning_rate": 1.8459164478889158e-06,
      "loss": 0.86936915,
      "num_input_tokens_seen": 96757685,
      "step": 4481,
      "time_per_iteration": 3.5353281497955322
    },
    {
      "auxiliary_loss_clip": 0.01323668,
      "auxiliary_loss_mlp": 0.01025189,
      "balance_loss_clip": 1.04321766,
      "balance_loss_mlp": 1.01793492,
      "epoch": 0.5389286358444056,
      "flos": 22236893147520.0,
      "grad_norm": 1.934824225279793,
      "language_loss": 0.75817955,
      "learning_rate": 1.8451398031568663e-06,
      "loss": 0.78166819,
      "num_input_tokens_seen": 96777310,
      "step": 4482,
      "time_per_iteration": 2.759193181991577
    },
    {
      "auxiliary_loss_clip": 0.01329542,
      "auxiliary_loss_mlp": 0.01033483,
      "balance_loss_clip": 1.04790926,
      "balance_loss_mlp": 1.02526379,
      "epoch": 0.5390488787350448,
      "flos": 24281718595200.0,
      "grad_norm": 1.7603867054941869,
      "language_loss": 0.74447489,
      "learning_rate": 1.844363181916986e-06,
      "loss": 0.76810515,
      "num_input_tokens_seen": 96798035,
      "step": 4483,
      "time_per_iteration": 3.6051831245422363
    },
    {
      "auxiliary_loss_clip": 0.012364,
      "auxiliary_loss_mlp": 0.01027018,
      "balance_loss_clip": 1.05080962,
      "balance_loss_mlp": 1.01884627,
      "epoch": 0.5391691216256839,
      "flos": 16581393688320.0,
      "grad_norm": 1.8023594792739972,
      "language_loss": 0.8304863,
      "learning_rate": 1.8435865842870868e-06,
      "loss": 0.85312051,
      "num_input_tokens_seen": 96815975,
      "step": 4484,
      "time_per_iteration": 2.6228084564208984
    },
    {
      "auxiliary_loss_clip": 0.01279873,
      "auxiliary_loss_mlp": 0.02572097,
      "balance_loss_clip": 1.04685783,
      "balance_loss_mlp": 1.00013149,
      "epoch": 0.5392893645163229,
      "flos": 23330049707520.0,
      "grad_norm": 1.7860447664803163,
      "language_loss": 0.71614051,
      "learning_rate": 1.8428100103849787e-06,
      "loss": 0.75466025,
      "num_input_tokens_seen": 96835770,
      "step": 4485,
      "time_per_iteration": 2.659630298614502
    },
    {
      "auxiliary_loss_clip": 0.01279754,
      "auxiliary_loss_mlp": 0.01030758,
      "balance_loss_clip": 1.0531466,
      "balance_loss_mlp": 1.02278328,
      "epoch": 0.5394096074069621,
      "flos": 15669801400320.0,
      "grad_norm": 2.1943808600938386,
      "language_loss": 0.73252809,
      "learning_rate": 1.842033460328467e-06,
      "loss": 0.75563318,
      "num_input_tokens_seen": 96854490,
      "step": 4486,
      "time_per_iteration": 3.5680630207061768
    },
    {
      "auxiliary_loss_clip": 0.01285215,
      "auxiliary_loss_mlp": 0.02569911,
      "balance_loss_clip": 1.04815054,
      "balance_loss_mlp": 1.00008488,
      "epoch": 0.5395298502976011,
      "flos": 22893447893760.0,
      "grad_norm": 1.828554318849219,
      "language_loss": 0.75214338,
      "learning_rate": 1.8412569342353541e-06,
      "loss": 0.79069459,
      "num_input_tokens_seen": 96874645,
      "step": 4487,
      "time_per_iteration": 2.7275190353393555
    },
    {
      "auxiliary_loss_clip": 0.01288374,
      "auxiliary_loss_mlp": 0.01032049,
      "balance_loss_clip": 1.05279601,
      "balance_loss_mlp": 1.02309656,
      "epoch": 0.5396500931882402,
      "flos": 23842135952640.0,
      "grad_norm": 2.153080314040614,
      "language_loss": 0.85085917,
      "learning_rate": 1.840480432223438e-06,
      "loss": 0.87406337,
      "num_input_tokens_seen": 96893650,
      "step": 4488,
      "time_per_iteration": 2.754060983657837
    },
    {
      "auxiliary_loss_clip": 0.01285937,
      "auxiliary_loss_mlp": 0.01030779,
      "balance_loss_clip": 1.0503118,
      "balance_loss_mlp": 1.02334595,
      "epoch": 0.5397703360788794,
      "flos": 26322988596480.0,
      "grad_norm": 2.3353168442384624,
      "language_loss": 0.7787292,
      "learning_rate": 1.8397039544105131e-06,
      "loss": 0.80189633,
      "num_input_tokens_seen": 96912735,
      "step": 4489,
      "time_per_iteration": 2.742990732192993
    },
    {
      "auxiliary_loss_clip": 0.01282915,
      "auxiliary_loss_mlp": 0.01030281,
      "balance_loss_clip": 1.04760683,
      "balance_loss_mlp": 1.02228832,
      "epoch": 0.5398905789695184,
      "flos": 21214588164480.0,
      "grad_norm": 2.0294604865489307,
      "language_loss": 0.69619083,
      "learning_rate": 1.8389275009143711e-06,
      "loss": 0.7193228,
      "num_input_tokens_seen": 96932475,
      "step": 4490,
      "time_per_iteration": 2.720344305038452
    },
    {
      "auxiliary_loss_clip": 0.01183139,
      "auxiliary_loss_mlp": 0.01029915,
      "balance_loss_clip": 1.05277562,
      "balance_loss_mlp": 1.02271473,
      "epoch": 0.5400108218601575,
      "flos": 25080335631360.0,
      "grad_norm": 1.7068144331066497,
      "language_loss": 0.73812258,
      "learning_rate": 1.8381510718527988e-06,
      "loss": 0.76025307,
      "num_input_tokens_seen": 96952085,
      "step": 4491,
      "time_per_iteration": 2.622429609298706
    },
    {
      "auxiliary_loss_clip": 0.01285631,
      "auxiliary_loss_mlp": 0.0103519,
      "balance_loss_clip": 1.04764557,
      "balance_loss_mlp": 1.027233,
      "epoch": 0.5401310647507966,
      "flos": 26357498588160.0,
      "grad_norm": 2.1186050753799677,
      "language_loss": 0.63425952,
      "learning_rate": 1.8373746673435812e-06,
      "loss": 0.65746772,
      "num_input_tokens_seen": 96973110,
      "step": 4492,
      "time_per_iteration": 2.7408626079559326
    },
    {
      "auxiliary_loss_clip": 0.0118726,
      "auxiliary_loss_mlp": 0.01026415,
      "balance_loss_clip": 1.05554032,
      "balance_loss_mlp": 1.01832032,
      "epoch": 0.5402513076414357,
      "flos": 27855332749440.0,
      "grad_norm": 1.5734292605496722,
      "language_loss": 0.7904737,
      "learning_rate": 1.8365982875044964e-06,
      "loss": 0.81261039,
      "num_input_tokens_seen": 96993420,
      "step": 4493,
      "time_per_iteration": 2.6595115661621094
    },
    {
      "auxiliary_loss_clip": 0.01240675,
      "auxiliary_loss_mlp": 0.02574928,
      "balance_loss_clip": 1.0539763,
      "balance_loss_mlp": 1.00012183,
      "epoch": 0.5403715505320748,
      "flos": 22893771116160.0,
      "grad_norm": 2.403059831138205,
      "language_loss": 0.7599628,
      "learning_rate": 1.8358219324533217e-06,
      "loss": 0.79811883,
      "num_input_tokens_seen": 97013685,
      "step": 4494,
      "time_per_iteration": 2.6333398818969727
    },
    {
      "auxiliary_loss_clip": 0.01279005,
      "auxiliary_loss_mlp": 0.01025114,
      "balance_loss_clip": 1.04934728,
      "balance_loss_mlp": 1.01823545,
      "epoch": 0.5404917934227139,
      "flos": 30224143895040.0,
      "grad_norm": 1.6994184572004745,
      "language_loss": 0.70376778,
      "learning_rate": 1.8350456023078292e-06,
      "loss": 0.72680891,
      "num_input_tokens_seen": 97036060,
      "step": 4495,
      "time_per_iteration": 2.715528964996338
    },
    {
      "auxiliary_loss_clip": 0.01189006,
      "auxiliary_loss_mlp": 0.01029381,
      "balance_loss_clip": 1.05282462,
      "balance_loss_mlp": 1.02132237,
      "epoch": 0.540612036313353,
      "flos": 19938502615680.0,
      "grad_norm": 2.527437182760878,
      "language_loss": 0.78297204,
      "learning_rate": 1.8342692971857874e-06,
      "loss": 0.80515587,
      "num_input_tokens_seen": 97055260,
      "step": 4496,
      "time_per_iteration": 2.637469530105591
    },
    {
      "auxiliary_loss_clip": 0.01278419,
      "auxiliary_loss_mlp": 0.01027076,
      "balance_loss_clip": 1.04976058,
      "balance_loss_mlp": 1.01939893,
      "epoch": 0.540732279203992,
      "flos": 24279599692800.0,
      "grad_norm": 6.590252840968521,
      "language_loss": 0.71451843,
      "learning_rate": 1.833493017204962e-06,
      "loss": 0.73757339,
      "num_input_tokens_seen": 97075365,
      "step": 4497,
      "time_per_iteration": 2.6559832096099854
    },
    {
      "auxiliary_loss_clip": 0.01186252,
      "auxiliary_loss_mlp": 0.01027985,
      "balance_loss_clip": 1.05397558,
      "balance_loss_mlp": 1.01939631,
      "epoch": 0.5408525220946312,
      "flos": 20193216935040.0,
      "grad_norm": 6.049448983686824,
      "language_loss": 0.77720964,
      "learning_rate": 1.8327167624831134e-06,
      "loss": 0.79935193,
      "num_input_tokens_seen": 97093095,
      "step": 4498,
      "time_per_iteration": 2.6645667552948
    },
    {
      "auxiliary_loss_clip": 0.01187516,
      "auxiliary_loss_mlp": 0.01027365,
      "balance_loss_clip": 1.05645776,
      "balance_loss_mlp": 1.01919961,
      "epoch": 0.5409727649852702,
      "flos": 24134448833280.0,
      "grad_norm": 2.1629203208861334,
      "language_loss": 0.70788091,
      "learning_rate": 1.831940533137999e-06,
      "loss": 0.7300297,
      "num_input_tokens_seen": 97112000,
      "step": 4499,
      "time_per_iteration": 2.5630605220794678
    },
    {
      "auxiliary_loss_clip": 0.01238949,
      "auxiliary_loss_mlp": 0.01030086,
      "balance_loss_clip": 1.05666375,
      "balance_loss_mlp": 1.02241516,
      "epoch": 0.5410930078759093,
      "flos": 23912700220800.0,
      "grad_norm": 1.6920747243818073,
      "language_loss": 0.72545588,
      "learning_rate": 1.8311643292873718e-06,
      "loss": 0.74814624,
      "num_input_tokens_seen": 97130820,
      "step": 4500,
      "time_per_iteration": 2.749964714050293
    },
    {
      "auxiliary_loss_clip": 0.01227367,
      "auxiliary_loss_mlp": 0.01030112,
      "balance_loss_clip": 1.05212712,
      "balance_loss_mlp": 1.02289379,
      "epoch": 0.5412132507665485,
      "flos": 21105132445440.0,
      "grad_norm": 1.8655982564667606,
      "language_loss": 0.88017476,
      "learning_rate": 1.8303881510489818e-06,
      "loss": 0.90274954,
      "num_input_tokens_seen": 97149210,
      "step": 4501,
      "time_per_iteration": 2.6332592964172363
    },
    {
      "auxiliary_loss_clip": 0.01284191,
      "auxiliary_loss_mlp": 0.01025602,
      "balance_loss_clip": 1.05249977,
      "balance_loss_mlp": 1.01756692,
      "epoch": 0.5413334936571875,
      "flos": 30227340205440.0,
      "grad_norm": 7.174409592603019,
      "language_loss": 0.69060445,
      "learning_rate": 1.829611998540574e-06,
      "loss": 0.71370238,
      "num_input_tokens_seen": 97170415,
      "step": 4502,
      "time_per_iteration": 2.731602191925049
    },
    {
      "auxiliary_loss_clip": 0.01232185,
      "auxiliary_loss_mlp": 0.02571935,
      "balance_loss_clip": 1.05078101,
      "balance_loss_mlp": 1.00010061,
      "epoch": 0.5414537365478266,
      "flos": 24279635606400.0,
      "grad_norm": 1.6719564317558226,
      "language_loss": 0.7986939,
      "learning_rate": 1.8288358718798914e-06,
      "loss": 0.83673513,
      "num_input_tokens_seen": 97189605,
      "step": 4503,
      "time_per_iteration": 2.6927952766418457
    },
    {
      "auxiliary_loss_clip": 0.01229492,
      "auxiliary_loss_mlp": 0.02568765,
      "balance_loss_clip": 1.05172086,
      "balance_loss_mlp": 1.00001073,
      "epoch": 0.5415739794384657,
      "flos": 16654543735680.0,
      "grad_norm": 1.6520341686985553,
      "language_loss": 0.72547799,
      "learning_rate": 1.8280597711846703e-06,
      "loss": 0.76346052,
      "num_input_tokens_seen": 97207845,
      "step": 4504,
      "time_per_iteration": 3.5981485843658447
    },
    {
      "auxiliary_loss_clip": 0.01239058,
      "auxiliary_loss_mlp": 0.01029801,
      "balance_loss_clip": 1.0572989,
      "balance_loss_mlp": 1.02239847,
      "epoch": 0.5416942223291048,
      "flos": 23185724860800.0,
      "grad_norm": 1.8969772841744423,
      "language_loss": 0.83687192,
      "learning_rate": 1.8272836965726455e-06,
      "loss": 0.85956049,
      "num_input_tokens_seen": 97226780,
      "step": 4505,
      "time_per_iteration": 2.6601197719573975
    },
    {
      "auxiliary_loss_clip": 0.01434436,
      "auxiliary_loss_mlp": 0.01030774,
      "balance_loss_clip": 1.0441308,
      "balance_loss_mlp": 1.02287006,
      "epoch": 0.5418144652197439,
      "flos": 20303247271680.0,
      "grad_norm": 1.8435670840164178,
      "language_loss": 0.78607178,
      "learning_rate": 1.8265076481615461e-06,
      "loss": 0.81072384,
      "num_input_tokens_seen": 97246695,
      "step": 4506,
      "time_per_iteration": 2.8759567737579346
    },
    {
      "auxiliary_loss_clip": 0.01283053,
      "auxiliary_loss_mlp": 0.01031091,
      "balance_loss_clip": 1.05370307,
      "balance_loss_mlp": 1.02329183,
      "epoch": 0.541934708110383,
      "flos": 12458633431680.0,
      "grad_norm": 2.130627239378055,
      "language_loss": 0.87794524,
      "learning_rate": 1.8257316260690987e-06,
      "loss": 0.90108669,
      "num_input_tokens_seen": 97264480,
      "step": 4507,
      "time_per_iteration": 3.5824501514434814
    },
    {
      "auxiliary_loss_clip": 0.01137866,
      "auxiliary_loss_mlp": 0.01024355,
      "balance_loss_clip": 1.05399561,
      "balance_loss_mlp": 1.01718497,
      "epoch": 0.5420549510010221,
      "flos": 21253802837760.0,
      "grad_norm": 1.4897702558176484,
      "language_loss": 0.76170838,
      "learning_rate": 1.8249556304130254e-06,
      "loss": 0.78333056,
      "num_input_tokens_seen": 97285760,
      "step": 4508,
      "time_per_iteration": 2.612170457839966
    },
    {
      "auxiliary_loss_clip": 0.01276188,
      "auxiliary_loss_mlp": 0.01030319,
      "balance_loss_clip": 1.04818344,
      "balance_loss_mlp": 1.02239788,
      "epoch": 0.5421751938916611,
      "flos": 29490524519040.0,
      "grad_norm": 18.734975432679548,
      "language_loss": 0.68405402,
      "learning_rate": 1.824179661311044e-06,
      "loss": 0.70711911,
      "num_input_tokens_seen": 97304510,
      "step": 4509,
      "time_per_iteration": 3.590867042541504
    },
    {
      "auxiliary_loss_clip": 0.01379585,
      "auxiliary_loss_mlp": 0.01028302,
      "balance_loss_clip": 1.042009,
      "balance_loss_mlp": 1.02106047,
      "epoch": 0.5422954367823003,
      "flos": 18734238311040.0,
      "grad_norm": 1.9151834614788033,
      "language_loss": 0.7952081,
      "learning_rate": 1.823403718880868e-06,
      "loss": 0.81928694,
      "num_input_tokens_seen": 97323270,
      "step": 4510,
      "time_per_iteration": 2.7538390159606934
    },
    {
      "auxiliary_loss_clip": 0.01285111,
      "auxiliary_loss_mlp": 0.01031238,
      "balance_loss_clip": 1.04706228,
      "balance_loss_mlp": 1.02322769,
      "epoch": 0.5424156796729394,
      "flos": 39969006940800.0,
      "grad_norm": 2.231119656455754,
      "language_loss": 0.66278648,
      "learning_rate": 1.822627803240207e-06,
      "loss": 0.68594998,
      "num_input_tokens_seen": 97345600,
      "step": 4511,
      "time_per_iteration": 2.8428750038146973
    },
    {
      "auxiliary_loss_clip": 0.01334292,
      "auxiliary_loss_mlp": 0.01028313,
      "balance_loss_clip": 1.04843974,
      "balance_loss_mlp": 1.02116346,
      "epoch": 0.5425359225635784,
      "flos": 11546538353280.0,
      "grad_norm": 2.207829462459169,
      "language_loss": 0.85388541,
      "learning_rate": 1.8218519145067675e-06,
      "loss": 0.87751144,
      "num_input_tokens_seen": 97361220,
      "step": 4512,
      "time_per_iteration": 3.561455726623535
    },
    {
      "auxiliary_loss_clip": 0.01328903,
      "auxiliary_loss_mlp": 0.01026957,
      "balance_loss_clip": 1.04729211,
      "balance_loss_mlp": 1.01933408,
      "epoch": 0.5426561654542175,
      "flos": 20229702174720.0,
      "grad_norm": 2.0436150692627417,
      "language_loss": 0.89546764,
      "learning_rate": 1.8210760527982508e-06,
      "loss": 0.9190262,
      "num_input_tokens_seen": 97381505,
      "step": 4513,
      "time_per_iteration": 2.7052371501922607
    },
    {
      "auxiliary_loss_clip": 0.01285174,
      "auxiliary_loss_mlp": 0.02567924,
      "balance_loss_clip": 1.05340958,
      "balance_loss_mlp": 0.99999392,
      "epoch": 0.5427764083448566,
      "flos": 21871681614720.0,
      "grad_norm": 4.559731883494288,
      "language_loss": 0.75159764,
      "learning_rate": 1.8203002182323552e-06,
      "loss": 0.79012871,
      "num_input_tokens_seen": 97399060,
      "step": 4514,
      "time_per_iteration": 2.727151393890381
    },
    {
      "auxiliary_loss_clip": 0.01291444,
      "auxiliary_loss_mlp": 0.01029084,
      "balance_loss_clip": 1.0534302,
      "balance_loss_mlp": 1.02112079,
      "epoch": 0.5428966512354957,
      "flos": 19640946349440.0,
      "grad_norm": 1.7681589357186471,
      "language_loss": 0.75639832,
      "learning_rate": 1.819524410926773e-06,
      "loss": 0.77960366,
      "num_input_tokens_seen": 97416740,
      "step": 4515,
      "time_per_iteration": 2.658545732498169
    },
    {
      "auxiliary_loss_clip": 0.01428535,
      "auxiliary_loss_mlp": 0.01026859,
      "balance_loss_clip": 1.04891884,
      "balance_loss_mlp": 1.01939631,
      "epoch": 0.5430168941261347,
      "flos": 22382187661440.0,
      "grad_norm": 1.495903969937153,
      "language_loss": 0.77149403,
      "learning_rate": 1.8187486309991944e-06,
      "loss": 0.79604793,
      "num_input_tokens_seen": 97437620,
      "step": 4516,
      "time_per_iteration": 2.849848985671997
    },
    {
      "auxiliary_loss_clip": 0.01141279,
      "auxiliary_loss_mlp": 0.01033135,
      "balance_loss_clip": 1.05400169,
      "balance_loss_mlp": 1.02585697,
      "epoch": 0.5431371370167739,
      "flos": 18764187275520.0,
      "grad_norm": 2.2904099498481565,
      "language_loss": 0.77523255,
      "learning_rate": 1.817972878567304e-06,
      "loss": 0.79697669,
      "num_input_tokens_seen": 97456275,
      "step": 4517,
      "time_per_iteration": 2.7956297397613525
    },
    {
      "auxiliary_loss_clip": 0.01289402,
      "auxiliary_loss_mlp": 0.01033158,
      "balance_loss_clip": 1.05074358,
      "balance_loss_mlp": 1.02535558,
      "epoch": 0.543257379907413,
      "flos": 18806023641600.0,
      "grad_norm": 1.8034055347374058,
      "language_loss": 0.76341069,
      "learning_rate": 1.8171971537487834e-06,
      "loss": 0.78663635,
      "num_input_tokens_seen": 97474925,
      "step": 4518,
      "time_per_iteration": 2.659496784210205
    },
    {
      "auxiliary_loss_clip": 0.01187936,
      "auxiliary_loss_mlp": 0.01026874,
      "balance_loss_clip": 1.05330682,
      "balance_loss_mlp": 1.01934588,
      "epoch": 0.543377622798052,
      "flos": 17493381025920.0,
      "grad_norm": 1.790860051453592,
      "language_loss": 0.80629349,
      "learning_rate": 1.8164214566613093e-06,
      "loss": 0.8284415,
      "num_input_tokens_seen": 97493550,
      "step": 4519,
      "time_per_iteration": 2.6098196506500244
    },
    {
      "auxiliary_loss_clip": 0.01182227,
      "auxiliary_loss_mlp": 0.01029875,
      "balance_loss_clip": 1.05164289,
      "balance_loss_mlp": 1.02226305,
      "epoch": 0.5434978656886912,
      "flos": 18989311766400.0,
      "grad_norm": 22.717279894417217,
      "language_loss": 0.66310996,
      "learning_rate": 1.8156457874225547e-06,
      "loss": 0.68523097,
      "num_input_tokens_seen": 97512010,
      "step": 4520,
      "time_per_iteration": 2.5993218421936035
    },
    {
      "auxiliary_loss_clip": 0.01273562,
      "auxiliary_loss_mlp": 0.01027522,
      "balance_loss_clip": 1.04960048,
      "balance_loss_mlp": 1.020787,
      "epoch": 0.5436181085793302,
      "flos": 17274936464640.0,
      "grad_norm": 2.1268214146062587,
      "language_loss": 0.80624598,
      "learning_rate": 1.814870146150187e-06,
      "loss": 0.82925683,
      "num_input_tokens_seen": 97530120,
      "step": 4521,
      "time_per_iteration": 2.632239580154419
    },
    {
      "auxiliary_loss_clip": 0.01189802,
      "auxiliary_loss_mlp": 0.01031256,
      "balance_loss_clip": 1.04858398,
      "balance_loss_mlp": 1.02362084,
      "epoch": 0.5437383514699693,
      "flos": 19098587917440.0,
      "grad_norm": 1.9880651742824034,
      "language_loss": 0.78589112,
      "learning_rate": 1.814094532961871e-06,
      "loss": 0.80810165,
      "num_input_tokens_seen": 97548695,
      "step": 4522,
      "time_per_iteration": 2.664611577987671
    },
    {
      "auxiliary_loss_clip": 0.01385493,
      "auxiliary_loss_mlp": 0.01029213,
      "balance_loss_clip": 1.0449785,
      "balance_loss_mlp": 1.02141118,
      "epoch": 0.5438585943606085,
      "flos": 22602715211520.0,
      "grad_norm": 1.846198873535851,
      "language_loss": 0.83866107,
      "learning_rate": 1.8133189479752666e-06,
      "loss": 0.86280811,
      "num_input_tokens_seen": 97567625,
      "step": 4523,
      "time_per_iteration": 2.8159639835357666
    },
    {
      "auxiliary_loss_clip": 0.011806,
      "auxiliary_loss_mlp": 0.01032773,
      "balance_loss_clip": 1.05122495,
      "balance_loss_mlp": 1.02554929,
      "epoch": 0.5439788372512475,
      "flos": 21798495653760.0,
      "grad_norm": 9.38377310482567,
      "language_loss": 0.82213652,
      "learning_rate": 1.8125433913080292e-06,
      "loss": 0.84427023,
      "num_input_tokens_seen": 97585325,
      "step": 4524,
      "time_per_iteration": 2.5612564086914062
    },
    {
      "auxiliary_loss_clip": 0.01583549,
      "auxiliary_loss_mlp": 0.01030194,
      "balance_loss_clip": 1.04400575,
      "balance_loss_mlp": 1.02289844,
      "epoch": 0.5440990801418866,
      "flos": 16399362539520.0,
      "grad_norm": 2.67320244953315,
      "language_loss": 0.82473689,
      "learning_rate": 1.811767863077811e-06,
      "loss": 0.8508743,
      "num_input_tokens_seen": 97604275,
      "step": 4525,
      "time_per_iteration": 2.988917827606201
    },
    {
      "auxiliary_loss_clip": 0.01422784,
      "auxiliary_loss_mlp": 0.01028875,
      "balance_loss_clip": 1.04622662,
      "balance_loss_mlp": 1.02184796,
      "epoch": 0.5442193230325257,
      "flos": 21615638492160.0,
      "grad_norm": 1.5842614954494652,
      "language_loss": 0.78483033,
      "learning_rate": 1.8109923634022577e-06,
      "loss": 0.80934691,
      "num_input_tokens_seen": 97624300,
      "step": 4526,
      "time_per_iteration": 2.9779064655303955
    },
    {
      "auxiliary_loss_clip": 0.01186187,
      "auxiliary_loss_mlp": 0.01026552,
      "balance_loss_clip": 1.05318046,
      "balance_loss_mlp": 1.01920557,
      "epoch": 0.5443395659231648,
      "flos": 15481198062720.0,
      "grad_norm": 2.1603266645149946,
      "language_loss": 0.86411786,
      "learning_rate": 1.8102168923990128e-06,
      "loss": 0.88624519,
      "num_input_tokens_seen": 97637845,
      "step": 4527,
      "time_per_iteration": 2.7571797370910645
    },
    {
      "auxiliary_loss_clip": 0.01237,
      "auxiliary_loss_mlp": 0.02568195,
      "balance_loss_clip": 1.0537076,
      "balance_loss_mlp": 1.00003242,
      "epoch": 0.5444598088138038,
      "flos": 18770436241920.0,
      "grad_norm": 2.997992636863713,
      "language_loss": 0.7961989,
      "learning_rate": 1.809441450185714e-06,
      "loss": 0.83425081,
      "num_input_tokens_seen": 97656330,
      "step": 4528,
      "time_per_iteration": 2.7207083702087402
    },
    {
      "auxiliary_loss_clip": 0.012863,
      "auxiliary_loss_mlp": 0.01025833,
      "balance_loss_clip": 1.04706907,
      "balance_loss_mlp": 1.0180006,
      "epoch": 0.544580051704443,
      "flos": 21142335957120.0,
      "grad_norm": 2.3716861460386442,
      "language_loss": 0.73596263,
      "learning_rate": 1.8086660368799958e-06,
      "loss": 0.75908399,
      "num_input_tokens_seen": 97674380,
      "step": 4529,
      "time_per_iteration": 2.667907953262329
    },
    {
      "auxiliary_loss_clip": 0.01283593,
      "auxiliary_loss_mlp": 0.01029788,
      "balance_loss_clip": 1.05099499,
      "balance_loss_mlp": 1.02139592,
      "epoch": 0.5447002945950821,
      "flos": 32491508054400.0,
      "grad_norm": 1.745474619489347,
      "language_loss": 0.77647173,
      "learning_rate": 1.807890652599488e-06,
      "loss": 0.79960549,
      "num_input_tokens_seen": 97698765,
      "step": 4530,
      "time_per_iteration": 4.2622387409210205
    },
    {
      "auxiliary_loss_clip": 0.01181446,
      "auxiliary_loss_mlp": 0.01027939,
      "balance_loss_clip": 1.05260086,
      "balance_loss_mlp": 1.02097154,
      "epoch": 0.5448205374857211,
      "flos": 11798307757440.0,
      "grad_norm": 2.0374502603550955,
      "language_loss": 0.82604313,
      "learning_rate": 1.8071152974618156e-06,
      "loss": 0.84813696,
      "num_input_tokens_seen": 97716565,
      "step": 4531,
      "time_per_iteration": 2.622394561767578
    },
    {
      "auxiliary_loss_clip": 0.01332129,
      "auxiliary_loss_mlp": 0.02569175,
      "balance_loss_clip": 1.04577947,
      "balance_loss_mlp": 1.00003767,
      "epoch": 0.5449407803763603,
      "flos": 24133766474880.0,
      "grad_norm": 2.2034734964559073,
      "language_loss": 0.78263915,
      "learning_rate": 1.806339971584599e-06,
      "loss": 0.82165217,
      "num_input_tokens_seen": 97733225,
      "step": 4532,
      "time_per_iteration": 2.681488513946533
    },
    {
      "auxiliary_loss_clip": 0.01184245,
      "auxiliary_loss_mlp": 0.01025221,
      "balance_loss_clip": 1.05327892,
      "balance_loss_mlp": 1.01825368,
      "epoch": 0.5450610232669993,
      "flos": 23258551685760.0,
      "grad_norm": 1.8577325122815846,
      "language_loss": 0.85463971,
      "learning_rate": 1.8055646750854546e-06,
      "loss": 0.87673438,
      "num_input_tokens_seen": 97752735,
      "step": 4533,
      "time_per_iteration": 3.44612717628479
    },
    {
      "auxiliary_loss_clip": 0.01287062,
      "auxiliary_loss_mlp": 0.01033081,
      "balance_loss_clip": 1.05132365,
      "balance_loss_mlp": 1.02547503,
      "epoch": 0.5451812661576384,
      "flos": 17785083375360.0,
      "grad_norm": 2.404428848962965,
      "language_loss": 0.81787121,
      "learning_rate": 1.8047894080819945e-06,
      "loss": 0.84107268,
      "num_input_tokens_seen": 97769985,
      "step": 4534,
      "time_per_iteration": 2.6227357387542725
    },
    {
      "auxiliary_loss_clip": 0.01071526,
      "auxiliary_loss_mlp": 0.01007921,
      "balance_loss_clip": 1.01297879,
      "balance_loss_mlp": 1.00700307,
      "epoch": 0.5453015090482776,
      "flos": 71062586513280.0,
      "grad_norm": 0.722843420769958,
      "language_loss": 0.63172281,
      "learning_rate": 1.8040141706918258e-06,
      "loss": 0.65251732,
      "num_input_tokens_seen": 97831225,
      "step": 4535,
      "time_per_iteration": 4.191967487335205
    },
    {
      "auxiliary_loss_clip": 0.01184263,
      "auxiliary_loss_mlp": 0.01025745,
      "balance_loss_clip": 1.05089355,
      "balance_loss_mlp": 1.01859236,
      "epoch": 0.5454217519389166,
      "flos": 25552201622400.0,
      "grad_norm": 1.9239383213910626,
      "language_loss": 0.77068847,
      "learning_rate": 1.8032389630325525e-06,
      "loss": 0.79278857,
      "num_input_tokens_seen": 97849975,
      "step": 4536,
      "time_per_iteration": 2.7548232078552246
    },
    {
      "auxiliary_loss_clip": 0.01280904,
      "auxiliary_loss_mlp": 0.01027561,
      "balance_loss_clip": 1.04486597,
      "balance_loss_mlp": 1.02066135,
      "epoch": 0.5455419948295557,
      "flos": 23658345037440.0,
      "grad_norm": 1.9732184564176583,
      "language_loss": 0.75688046,
      "learning_rate": 1.8024637852217707e-06,
      "loss": 0.77996516,
      "num_input_tokens_seen": 97869700,
      "step": 4537,
      "time_per_iteration": 2.675246000289917
    },
    {
      "auxiliary_loss_clip": 0.01280984,
      "auxiliary_loss_mlp": 0.01027753,
      "balance_loss_clip": 1.05081511,
      "balance_loss_mlp": 1.02078819,
      "epoch": 0.5456622377201948,
      "flos": 23403989854080.0,
      "grad_norm": 1.927195178386429,
      "language_loss": 0.84745532,
      "learning_rate": 1.8016886373770766e-06,
      "loss": 0.87054271,
      "num_input_tokens_seen": 97888215,
      "step": 4538,
      "time_per_iteration": 2.7440197467803955
    },
    {
      "auxiliary_loss_clip": 0.01283694,
      "auxiliary_loss_mlp": 0.01027614,
      "balance_loss_clip": 1.04930031,
      "balance_loss_mlp": 1.02009165,
      "epoch": 0.5457824806108339,
      "flos": 23988040997760.0,
      "grad_norm": 1.7333558321307618,
      "language_loss": 0.79244781,
      "learning_rate": 1.8009135196160579e-06,
      "loss": 0.81556094,
      "num_input_tokens_seen": 97907090,
      "step": 4539,
      "time_per_iteration": 3.633702516555786
    },
    {
      "auxiliary_loss_clip": 0.01328266,
      "auxiliary_loss_mlp": 0.01031144,
      "balance_loss_clip": 1.04686546,
      "balance_loss_mlp": 1.02390516,
      "epoch": 0.545902723501473,
      "flos": 22565870835840.0,
      "grad_norm": 2.1583400820763705,
      "language_loss": 0.84296107,
      "learning_rate": 1.8001384320563e-06,
      "loss": 0.86655509,
      "num_input_tokens_seen": 97927345,
      "step": 4540,
      "time_per_iteration": 2.7704291343688965
    },
    {
      "auxiliary_loss_clip": 0.01071155,
      "auxiliary_loss_mlp": 0.01004808,
      "balance_loss_clip": 1.01249921,
      "balance_loss_mlp": 1.00386071,
      "epoch": 0.5460229663921121,
      "flos": 55198399685760.0,
      "grad_norm": 0.7712229241989073,
      "language_loss": 0.57726693,
      "learning_rate": 1.7993633748153833e-06,
      "loss": 0.59802657,
      "num_input_tokens_seen": 97981950,
      "step": 4541,
      "time_per_iteration": 3.054633378982544
    },
    {
      "auxiliary_loss_clip": 0.01238785,
      "auxiliary_loss_mlp": 0.01024681,
      "balance_loss_clip": 1.05315721,
      "balance_loss_mlp": 1.01764131,
      "epoch": 0.5461432092827512,
      "flos": 15413866018560.0,
      "grad_norm": 1.9407270689064575,
      "language_loss": 0.72759199,
      "learning_rate": 1.7985883480108834e-06,
      "loss": 0.75022662,
      "num_input_tokens_seen": 97999585,
      "step": 4542,
      "time_per_iteration": 2.593532085418701
    },
    {
      "auxiliary_loss_clip": 0.01234194,
      "auxiliary_loss_mlp": 0.0102539,
      "balance_loss_clip": 1.05098093,
      "balance_loss_mlp": 1.01839209,
      "epoch": 0.5462634521733902,
      "flos": 24024921287040.0,
      "grad_norm": 2.6504388740649856,
      "language_loss": 0.72464907,
      "learning_rate": 1.797813351760371e-06,
      "loss": 0.74724489,
      "num_input_tokens_seen": 98021290,
      "step": 4543,
      "time_per_iteration": 2.7103240489959717
    },
    {
      "auxiliary_loss_clip": 0.01184304,
      "auxiliary_loss_mlp": 0.01028163,
      "balance_loss_clip": 1.05220294,
      "balance_loss_mlp": 1.02083755,
      "epoch": 0.5463836950640293,
      "flos": 22820944291200.0,
      "grad_norm": 2.028097284284308,
      "language_loss": 0.78036577,
      "learning_rate": 1.7970383861814116e-06,
      "loss": 0.80249047,
      "num_input_tokens_seen": 98041060,
      "step": 4544,
      "time_per_iteration": 2.6060426235198975
    },
    {
      "auxiliary_loss_clip": 0.01232552,
      "auxiliary_loss_mlp": 0.01028406,
      "balance_loss_clip": 1.05304623,
      "balance_loss_mlp": 1.02028155,
      "epoch": 0.5465039379546685,
      "flos": 20448290390400.0,
      "grad_norm": 2.1866419856603345,
      "language_loss": 0.74155116,
      "learning_rate": 1.7962634513915684e-06,
      "loss": 0.76416069,
      "num_input_tokens_seen": 98058410,
      "step": 4545,
      "time_per_iteration": 2.667836904525757
    },
    {
      "auxiliary_loss_clip": 0.01182284,
      "auxiliary_loss_mlp": 0.0102707,
      "balance_loss_clip": 1.0525347,
      "balance_loss_mlp": 1.02000701,
      "epoch": 0.5466241808453075,
      "flos": 17343310003200.0,
      "grad_norm": 1.5617020868691744,
      "language_loss": 0.79374468,
      "learning_rate": 1.7954885475083969e-06,
      "loss": 0.81583822,
      "num_input_tokens_seen": 98076080,
      "step": 4546,
      "time_per_iteration": 2.608644962310791
    },
    {
      "auxiliary_loss_clip": 0.01186926,
      "auxiliary_loss_mlp": 0.01025106,
      "balance_loss_clip": 1.05423212,
      "balance_loss_mlp": 1.01738429,
      "epoch": 0.5467444237359466,
      "flos": 21617039122560.0,
      "grad_norm": 2.709724757275978,
      "language_loss": 0.73025602,
      "learning_rate": 1.7947136746494513e-06,
      "loss": 0.75237632,
      "num_input_tokens_seen": 98096995,
      "step": 4547,
      "time_per_iteration": 2.647827625274658
    },
    {
      "auxiliary_loss_clip": 0.01231108,
      "auxiliary_loss_mlp": 0.0102524,
      "balance_loss_clip": 1.0529213,
      "balance_loss_mlp": 1.01813793,
      "epoch": 0.5468646666265857,
      "flos": 24170467196160.0,
      "grad_norm": 1.977722154551884,
      "language_loss": 0.88379419,
      "learning_rate": 1.793938832932277e-06,
      "loss": 0.90635765,
      "num_input_tokens_seen": 98115105,
      "step": 4548,
      "time_per_iteration": 2.73384428024292
    },
    {
      "auxiliary_loss_clip": 0.01182334,
      "auxiliary_loss_mlp": 0.01026735,
      "balance_loss_clip": 1.05122972,
      "balance_loss_mlp": 1.01933777,
      "epoch": 0.5469849095172248,
      "flos": 27527001505920.0,
      "grad_norm": 1.9389726926407547,
      "language_loss": 0.70361549,
      "learning_rate": 1.7931640224744185e-06,
      "loss": 0.72570622,
      "num_input_tokens_seen": 98135655,
      "step": 4549,
      "time_per_iteration": 2.6903538703918457
    },
    {
      "auxiliary_loss_clip": 0.01321598,
      "auxiliary_loss_mlp": 0.01028925,
      "balance_loss_clip": 1.0408783,
      "balance_loss_mlp": 1.0213728,
      "epoch": 0.5471051524078638,
      "flos": 27964680727680.0,
      "grad_norm": 1.612362214963138,
      "language_loss": 0.73541135,
      "learning_rate": 1.7923892433934127e-06,
      "loss": 0.75891662,
      "num_input_tokens_seen": 98156730,
      "step": 4550,
      "time_per_iteration": 2.7857728004455566
    },
    {
      "auxiliary_loss_clip": 0.01286035,
      "auxiliary_loss_mlp": 0.02572371,
      "balance_loss_clip": 1.05168438,
      "balance_loss_mlp": 1.00001192,
      "epoch": 0.547225395298503,
      "flos": 18150510389760.0,
      "grad_norm": 2.1548236682635062,
      "language_loss": 0.78824866,
      "learning_rate": 1.7916144958067939e-06,
      "loss": 0.82683277,
      "num_input_tokens_seen": 98174590,
      "step": 4551,
      "time_per_iteration": 2.6917479038238525
    },
    {
      "auxiliary_loss_clip": 0.01231756,
      "auxiliary_loss_mlp": 0.01026825,
      "balance_loss_clip": 1.05020177,
      "balance_loss_mlp": 1.01969051,
      "epoch": 0.5473456381891421,
      "flos": 21361498790400.0,
      "grad_norm": 1.956477709628978,
      "language_loss": 0.78853166,
      "learning_rate": 1.7908397798320905e-06,
      "loss": 0.81111747,
      "num_input_tokens_seen": 98194325,
      "step": 4552,
      "time_per_iteration": 2.717228412628174
    },
    {
      "auxiliary_loss_clip": 0.01233161,
      "auxiliary_loss_mlp": 0.02570686,
      "balance_loss_clip": 1.04976487,
      "balance_loss_mlp": 0.99998915,
      "epoch": 0.5474658810797811,
      "flos": 19932145908480.0,
      "grad_norm": 1.953683815224385,
      "language_loss": 0.7482565,
      "learning_rate": 1.7900650955868265e-06,
      "loss": 0.786295,
      "num_input_tokens_seen": 98213970,
      "step": 4553,
      "time_per_iteration": 2.634089469909668
    },
    {
      "auxiliary_loss_clip": 0.01230538,
      "auxiliary_loss_mlp": 0.02568502,
      "balance_loss_clip": 1.05175233,
      "balance_loss_mlp": 0.99999541,
      "epoch": 0.5475861239704203,
      "flos": 50476217264640.0,
      "grad_norm": 1.429820852738912,
      "language_loss": 0.76358128,
      "learning_rate": 1.7892904431885202e-06,
      "loss": 0.80157173,
      "num_input_tokens_seen": 98241145,
      "step": 4554,
      "time_per_iteration": 2.971467971801758
    },
    {
      "auxiliary_loss_clip": 0.0137719,
      "auxiliary_loss_mlp": 0.0103775,
      "balance_loss_clip": 1.0429678,
      "balance_loss_mlp": 1.03045464,
      "epoch": 0.5477063668610593,
      "flos": 20705123612160.0,
      "grad_norm": 3.4296478902106786,
      "language_loss": 0.75548542,
      "learning_rate": 1.788515822754686e-06,
      "loss": 0.77963477,
      "num_input_tokens_seen": 98261565,
      "step": 4555,
      "time_per_iteration": 2.786411762237549
    },
    {
      "auxiliary_loss_clip": 0.01336255,
      "auxiliary_loss_mlp": 0.01025055,
      "balance_loss_clip": 1.0455333,
      "balance_loss_mlp": 1.01732421,
      "epoch": 0.5478266097516984,
      "flos": 19609740408960.0,
      "grad_norm": 2.7209734079417354,
      "language_loss": 0.78468239,
      "learning_rate": 1.7877412344028335e-06,
      "loss": 0.80829543,
      "num_input_tokens_seen": 98281370,
      "step": 4556,
      "time_per_iteration": 3.6621856689453125
    },
    {
      "auxiliary_loss_clip": 0.01234482,
      "auxiliary_loss_mlp": 0.01039401,
      "balance_loss_clip": 1.05068612,
      "balance_loss_mlp": 1.03191435,
      "epoch": 0.5479468526423376,
      "flos": 12896599962240.0,
      "grad_norm": 2.2328470229357387,
      "language_loss": 0.7738322,
      "learning_rate": 1.7869666782504668e-06,
      "loss": 0.79657102,
      "num_input_tokens_seen": 98297950,
      "step": 4557,
      "time_per_iteration": 2.6051809787750244
    },
    {
      "auxiliary_loss_clip": 0.01274134,
      "auxiliary_loss_mlp": 0.01027028,
      "balance_loss_clip": 1.04632258,
      "balance_loss_mlp": 1.01992583,
      "epoch": 0.5480670955329766,
      "flos": 18588800142720.0,
      "grad_norm": 1.763582687951691,
      "language_loss": 0.68646902,
      "learning_rate": 1.7861921544150867e-06,
      "loss": 0.70948058,
      "num_input_tokens_seen": 98316800,
      "step": 4558,
      "time_per_iteration": 2.6647815704345703
    },
    {
      "auxiliary_loss_clip": 0.01420766,
      "auxiliary_loss_mlp": 0.02566921,
      "balance_loss_clip": 1.04587042,
      "balance_loss_mlp": 0.99998569,
      "epoch": 0.5481873384236157,
      "flos": 15954608338560.0,
      "grad_norm": 2.6158473172140715,
      "language_loss": 0.76695633,
      "learning_rate": 1.7854176630141856e-06,
      "loss": 0.80683315,
      "num_input_tokens_seen": 98333935,
      "step": 4559,
      "time_per_iteration": 3.6588268280029297
    },
    {
      "auxiliary_loss_clip": 0.01187977,
      "auxiliary_loss_mlp": 0.01029373,
      "balance_loss_clip": 1.05519128,
      "balance_loss_mlp": 1.02154672,
      "epoch": 0.5483075813142548,
      "flos": 22783812606720.0,
      "grad_norm": 2.403270253988939,
      "language_loss": 0.84664434,
      "learning_rate": 1.784643204165255e-06,
      "loss": 0.86881787,
      "num_input_tokens_seen": 98353255,
      "step": 4560,
      "time_per_iteration": 2.6644489765167236
    },
    {
      "auxiliary_loss_clip": 0.01227845,
      "auxiliary_loss_mlp": 0.01023994,
      "balance_loss_clip": 1.05223608,
      "balance_loss_mlp": 1.01677537,
      "epoch": 0.5484278242048939,
      "flos": 19317212046720.0,
      "grad_norm": 1.8351761355650127,
      "language_loss": 0.7749095,
      "learning_rate": 1.7838687779857783e-06,
      "loss": 0.79742789,
      "num_input_tokens_seen": 98371130,
      "step": 4561,
      "time_per_iteration": 3.598186492919922
    },
    {
      "auxiliary_loss_clip": 0.01278825,
      "auxiliary_loss_mlp": 0.01027723,
      "balance_loss_clip": 1.0482316,
      "balance_loss_mlp": 1.02014756,
      "epoch": 0.5485480670955329,
      "flos": 22816024128000.0,
      "grad_norm": 2.206789147246992,
      "language_loss": 0.64496541,
      "learning_rate": 1.7830943845932366e-06,
      "loss": 0.66803098,
      "num_input_tokens_seen": 98390455,
      "step": 4562,
      "time_per_iteration": 2.74924373626709
    },
    {
      "auxiliary_loss_clip": 0.01288911,
      "auxiliary_loss_mlp": 0.0102924,
      "balance_loss_clip": 1.05283606,
      "balance_loss_mlp": 1.02163458,
      "epoch": 0.5486683099861721,
      "flos": 22671304231680.0,
      "grad_norm": 2.7234820120898715,
      "language_loss": 0.75220937,
      "learning_rate": 1.7823200241051044e-06,
      "loss": 0.77539086,
      "num_input_tokens_seen": 98409370,
      "step": 4563,
      "time_per_iteration": 2.6385128498077393
    },
    {
      "auxiliary_loss_clip": 0.01186259,
      "auxiliary_loss_mlp": 0.01035216,
      "balance_loss_clip": 1.05574226,
      "balance_loss_mlp": 1.02739549,
      "epoch": 0.5487885528768112,
      "flos": 23149383275520.0,
      "grad_norm": 2.162138675944536,
      "language_loss": 0.80351186,
      "learning_rate": 1.7815456966388513e-06,
      "loss": 0.82572663,
      "num_input_tokens_seen": 98428465,
      "step": 4564,
      "time_per_iteration": 2.634920120239258
    },
    {
      "auxiliary_loss_clip": 0.012354,
      "auxiliary_loss_mlp": 0.01025486,
      "balance_loss_clip": 1.0469749,
      "balance_loss_mlp": 1.01849401,
      "epoch": 0.5489087957674502,
      "flos": 22053928245120.0,
      "grad_norm": 5.703596231249841,
      "language_loss": 0.8094312,
      "learning_rate": 1.780771402311943e-06,
      "loss": 0.83204007,
      "num_input_tokens_seen": 98447300,
      "step": 4565,
      "time_per_iteration": 3.554806709289551
    },
    {
      "auxiliary_loss_clip": 0.01285985,
      "auxiliary_loss_mlp": 0.01033624,
      "balance_loss_clip": 1.05234861,
      "balance_loss_mlp": 1.02543974,
      "epoch": 0.5490290386580894,
      "flos": 24315977191680.0,
      "grad_norm": 2.199132603250437,
      "language_loss": 0.78660667,
      "learning_rate": 1.7799971412418374e-06,
      "loss": 0.80980277,
      "num_input_tokens_seen": 98468695,
      "step": 4566,
      "time_per_iteration": 2.71284556388855
    },
    {
      "auxiliary_loss_clip": 0.01236602,
      "auxiliary_loss_mlp": 0.01028901,
      "balance_loss_clip": 1.04817593,
      "balance_loss_mlp": 1.02041292,
      "epoch": 0.5491492815487284,
      "flos": 18294942977280.0,
      "grad_norm": 3.3116210034817164,
      "language_loss": 0.74298841,
      "learning_rate": 1.7792229135459918e-06,
      "loss": 0.76564342,
      "num_input_tokens_seen": 98485345,
      "step": 4567,
      "time_per_iteration": 2.566453695297241
    },
    {
      "auxiliary_loss_clip": 0.01297039,
      "auxiliary_loss_mlp": 0.01001359,
      "balance_loss_clip": 1.02930641,
      "balance_loss_mlp": 1.00036931,
      "epoch": 0.5492695244393675,
      "flos": 64550257050240.0,
      "grad_norm": 0.7366953679895234,
      "language_loss": 0.61591882,
      "learning_rate": 1.7784487193418538e-06,
      "loss": 0.63890278,
      "num_input_tokens_seen": 98543195,
      "step": 4568,
      "time_per_iteration": 3.097522258758545
    },
    {
      "auxiliary_loss_clip": 0.01320614,
      "auxiliary_loss_mlp": 0.01028311,
      "balance_loss_clip": 1.04266357,
      "balance_loss_mlp": 1.02032447,
      "epoch": 0.5493897673300067,
      "flos": 17379579761280.0,
      "grad_norm": 2.3707279325351176,
      "language_loss": 0.60995948,
      "learning_rate": 1.7776745587468698e-06,
      "loss": 0.63344872,
      "num_input_tokens_seen": 98560620,
      "step": 4569,
      "time_per_iteration": 2.5947883129119873
    },
    {
      "auxiliary_loss_clip": 0.01183921,
      "auxiliary_loss_mlp": 0.01026625,
      "balance_loss_clip": 1.05203688,
      "balance_loss_mlp": 1.01950264,
      "epoch": 0.5495100102206457,
      "flos": 19901765980800.0,
      "grad_norm": 2.8430824723016763,
      "language_loss": 0.82338929,
      "learning_rate": 1.7769004318784776e-06,
      "loss": 0.84549487,
      "num_input_tokens_seen": 98578265,
      "step": 4570,
      "time_per_iteration": 2.704746961593628
    },
    {
      "auxiliary_loss_clip": 0.01231271,
      "auxiliary_loss_mlp": 0.01025451,
      "balance_loss_clip": 1.04989791,
      "balance_loss_mlp": 1.01835489,
      "epoch": 0.5496302531112848,
      "flos": 16727190992640.0,
      "grad_norm": 1.6934261350444662,
      "language_loss": 0.80537617,
      "learning_rate": 1.776126338854113e-06,
      "loss": 0.82794338,
      "num_input_tokens_seen": 98596055,
      "step": 4571,
      "time_per_iteration": 2.5802130699157715
    },
    {
      "auxiliary_loss_clip": 0.01238288,
      "auxiliary_loss_mlp": 0.01027132,
      "balance_loss_clip": 1.05776596,
      "balance_loss_mlp": 1.01962459,
      "epoch": 0.5497504960019239,
      "flos": 24572343536640.0,
      "grad_norm": 1.7851914431886786,
      "language_loss": 0.84453171,
      "learning_rate": 1.7753522797912044e-06,
      "loss": 0.86718589,
      "num_input_tokens_seen": 98616140,
      "step": 4572,
      "time_per_iteration": 2.6681549549102783
    },
    {
      "auxiliary_loss_clip": 0.01289196,
      "auxiliary_loss_mlp": 0.01028152,
      "balance_loss_clip": 1.04980206,
      "balance_loss_mlp": 1.02014112,
      "epoch": 0.549870738892563,
      "flos": 15450494912640.0,
      "grad_norm": 2.260560481052079,
      "language_loss": 0.70017695,
      "learning_rate": 1.7745782548071765e-06,
      "loss": 0.72335041,
      "num_input_tokens_seen": 98633035,
      "step": 4573,
      "time_per_iteration": 2.643233299255371
    },
    {
      "auxiliary_loss_clip": 0.01335236,
      "auxiliary_loss_mlp": 0.01025006,
      "balance_loss_clip": 1.05563998,
      "balance_loss_mlp": 1.01778483,
      "epoch": 0.549990981783202,
      "flos": 21069114082560.0,
      "grad_norm": 1.6912043240157393,
      "language_loss": 0.74257016,
      "learning_rate": 1.7738042640194482e-06,
      "loss": 0.76617259,
      "num_input_tokens_seen": 98652700,
      "step": 4574,
      "time_per_iteration": 2.688095808029175
    },
    {
      "auxiliary_loss_clip": 0.01183492,
      "auxiliary_loss_mlp": 0.01028838,
      "balance_loss_clip": 1.05232441,
      "balance_loss_mlp": 1.02102959,
      "epoch": 0.5501112246738411,
      "flos": 21395901041280.0,
      "grad_norm": 1.509259475590616,
      "language_loss": 0.70195031,
      "learning_rate": 1.7730303075454335e-06,
      "loss": 0.72407359,
      "num_input_tokens_seen": 98671590,
      "step": 4575,
      "time_per_iteration": 2.577312707901001
    },
    {
      "auxiliary_loss_clip": 0.01336363,
      "auxiliary_loss_mlp": 0.0102851,
      "balance_loss_clip": 1.04796553,
      "balance_loss_mlp": 1.02122593,
      "epoch": 0.5502314675644803,
      "flos": 17456931699840.0,
      "grad_norm": 1.9385994768995296,
      "language_loss": 0.84782892,
      "learning_rate": 1.7722563855025402e-06,
      "loss": 0.8714776,
      "num_input_tokens_seen": 98689620,
      "step": 4576,
      "time_per_iteration": 2.616804599761963
    },
    {
      "auxiliary_loss_clip": 0.01283625,
      "auxiliary_loss_mlp": 0.01022733,
      "balance_loss_clip": 1.04624486,
      "balance_loss_mlp": 1.01521957,
      "epoch": 0.5503517104551193,
      "flos": 24310410583680.0,
      "grad_norm": 2.527740756005029,
      "language_loss": 0.70783651,
      "learning_rate": 1.7714824980081721e-06,
      "loss": 0.73090011,
      "num_input_tokens_seen": 98708915,
      "step": 4577,
      "time_per_iteration": 2.648982286453247
    },
    {
      "auxiliary_loss_clip": 0.01230736,
      "auxiliary_loss_mlp": 0.01022201,
      "balance_loss_clip": 1.05358207,
      "balance_loss_mlp": 1.01505113,
      "epoch": 0.5504719533457584,
      "flos": 22419427086720.0,
      "grad_norm": 5.021141381762656,
      "language_loss": 0.73714876,
      "learning_rate": 1.7707086451797276e-06,
      "loss": 0.75967813,
      "num_input_tokens_seen": 98729790,
      "step": 4578,
      "time_per_iteration": 2.6209757328033447
    },
    {
      "auxiliary_loss_clip": 0.01230321,
      "auxiliary_loss_mlp": 0.010006,
      "balance_loss_clip": 1.00923371,
      "balance_loss_mlp": 0.99950927,
      "epoch": 0.5505921962363975,
      "flos": 67294155968640.0,
      "grad_norm": 0.7148361405147315,
      "language_loss": 0.52288961,
      "learning_rate": 1.7699348271345993e-06,
      "loss": 0.54519892,
      "num_input_tokens_seen": 98792415,
      "step": 4579,
      "time_per_iteration": 3.1932172775268555
    },
    {
      "auxiliary_loss_clip": 0.01292067,
      "auxiliary_loss_mlp": 0.00998735,
      "balance_loss_clip": 1.01070595,
      "balance_loss_mlp": 0.99767399,
      "epoch": 0.5507124391270366,
      "flos": 45685125578880.0,
      "grad_norm": 0.7071810800813072,
      "language_loss": 0.54426432,
      "learning_rate": 1.7691610439901753e-06,
      "loss": 0.56717229,
      "num_input_tokens_seen": 98855350,
      "step": 4580,
      "time_per_iteration": 3.328946828842163
    },
    {
      "auxiliary_loss_clip": 0.01235601,
      "auxiliary_loss_mlp": 0.01026215,
      "balance_loss_clip": 1.05187166,
      "balance_loss_mlp": 1.01912534,
      "epoch": 0.5508326820176757,
      "flos": 22273845264000.0,
      "grad_norm": 1.8088972977191407,
      "language_loss": 0.75374925,
      "learning_rate": 1.7683872958638367e-06,
      "loss": 0.77636737,
      "num_input_tokens_seen": 98874230,
      "step": 4581,
      "time_per_iteration": 2.602073907852173
    },
    {
      "auxiliary_loss_clip": 0.01281298,
      "auxiliary_loss_mlp": 0.01029888,
      "balance_loss_clip": 1.04812503,
      "balance_loss_mlp": 1.02235353,
      "epoch": 0.5509529249083148,
      "flos": 20012442762240.0,
      "grad_norm": 2.2097415108079685,
      "language_loss": 0.84065425,
      "learning_rate": 1.7676135828729614e-06,
      "loss": 0.86376619,
      "num_input_tokens_seen": 98893940,
      "step": 4582,
      "time_per_iteration": 3.5674266815185547
    },
    {
      "auxiliary_loss_clip": 0.0123105,
      "auxiliary_loss_mlp": 0.0102732,
      "balance_loss_clip": 1.05130231,
      "balance_loss_mlp": 1.01997924,
      "epoch": 0.5510731677989539,
      "flos": 21834801325440.0,
      "grad_norm": 2.217713858330521,
      "language_loss": 0.83121961,
      "learning_rate": 1.7668399051349205e-06,
      "loss": 0.85380328,
      "num_input_tokens_seen": 98913620,
      "step": 4583,
      "time_per_iteration": 2.808623790740967
    },
    {
      "auxiliary_loss_clip": 0.01332972,
      "auxiliary_loss_mlp": 0.01023862,
      "balance_loss_clip": 1.04965162,
      "balance_loss_mlp": 1.01614904,
      "epoch": 0.5511934106895929,
      "flos": 21467901853440.0,
      "grad_norm": 1.8908481794429806,
      "language_loss": 0.83432186,
      "learning_rate": 1.766066262767081e-06,
      "loss": 0.85789019,
      "num_input_tokens_seen": 98931460,
      "step": 4584,
      "time_per_iteration": 3.6763291358947754
    },
    {
      "auxiliary_loss_clip": 0.01276185,
      "auxiliary_loss_mlp": 0.01026947,
      "balance_loss_clip": 1.04996133,
      "balance_loss_mlp": 1.0195117,
      "epoch": 0.5513136535802321,
      "flos": 21068934514560.0,
      "grad_norm": 2.128123405659414,
      "language_loss": 0.77010506,
      "learning_rate": 1.765292655886803e-06,
      "loss": 0.79313636,
      "num_input_tokens_seen": 98950105,
      "step": 4585,
      "time_per_iteration": 2.6600000858306885
    },
    {
      "auxiliary_loss_clip": 0.01338911,
      "auxiliary_loss_mlp": 0.01028288,
      "balance_loss_clip": 1.04756963,
      "balance_loss_mlp": 1.02055407,
      "epoch": 0.5514338964708712,
      "flos": 27815004754560.0,
      "grad_norm": 3.6720807319292916,
      "language_loss": 0.70712692,
      "learning_rate": 1.764519084611443e-06,
      "loss": 0.7307989,
      "num_input_tokens_seen": 98970560,
      "step": 4586,
      "time_per_iteration": 2.7389307022094727
    },
    {
      "auxiliary_loss_clip": 0.01283338,
      "auxiliary_loss_mlp": 0.01024156,
      "balance_loss_clip": 1.04944062,
      "balance_loss_mlp": 1.01630056,
      "epoch": 0.5515541393615102,
      "flos": 21908525990400.0,
      "grad_norm": 2.0478188633420316,
      "language_loss": 0.78095233,
      "learning_rate": 1.7637455490583505e-06,
      "loss": 0.80402732,
      "num_input_tokens_seen": 98989885,
      "step": 4587,
      "time_per_iteration": 3.559061288833618
    },
    {
      "auxiliary_loss_clip": 0.01235237,
      "auxiliary_loss_mlp": 0.01027283,
      "balance_loss_clip": 1.05341852,
      "balance_loss_mlp": 1.0203836,
      "epoch": 0.5516743822521494,
      "flos": 20485422074880.0,
      "grad_norm": 2.1955105760240534,
      "language_loss": 0.7750935,
      "learning_rate": 1.7629720493448701e-06,
      "loss": 0.7977187,
      "num_input_tokens_seen": 99007180,
      "step": 4588,
      "time_per_iteration": 2.61098313331604
    },
    {
      "auxiliary_loss_clip": 0.01290539,
      "auxiliary_loss_mlp": 0.01026211,
      "balance_loss_clip": 1.04974961,
      "balance_loss_mlp": 1.01895165,
      "epoch": 0.5517946251427884,
      "flos": 14940383915520.0,
      "grad_norm": 1.8161021040661782,
      "language_loss": 0.8503474,
      "learning_rate": 1.7621985855883418e-06,
      "loss": 0.87351489,
      "num_input_tokens_seen": 99023880,
      "step": 4589,
      "time_per_iteration": 2.7922823429107666
    },
    {
      "auxiliary_loss_clip": 0.01275385,
      "auxiliary_loss_mlp": 0.01028398,
      "balance_loss_clip": 1.04781747,
      "balance_loss_mlp": 1.02100968,
      "epoch": 0.5519148680334275,
      "flos": 18404865573120.0,
      "grad_norm": 2.2286114814836293,
      "language_loss": 0.72150129,
      "learning_rate": 1.7614251579060983e-06,
      "loss": 0.74453908,
      "num_input_tokens_seen": 99042475,
      "step": 4590,
      "time_per_iteration": 3.622863531112671
    },
    {
      "auxiliary_loss_clip": 0.01232624,
      "auxiliary_loss_mlp": 0.01027211,
      "balance_loss_clip": 1.048141,
      "balance_loss_mlp": 1.01979637,
      "epoch": 0.5520351109240667,
      "flos": 25113337251840.0,
      "grad_norm": 1.8416988980400224,
      "language_loss": 0.84881127,
      "learning_rate": 1.76065176641547e-06,
      "loss": 0.87140965,
      "num_input_tokens_seen": 99065185,
      "step": 4591,
      "time_per_iteration": 2.771068811416626
    },
    {
      "auxiliary_loss_clip": 0.0123162,
      "auxiliary_loss_mlp": 0.01022859,
      "balance_loss_clip": 1.04757953,
      "balance_loss_mlp": 1.0154804,
      "epoch": 0.5521553538147057,
      "flos": 21069545045760.0,
      "grad_norm": 1.9072031534947504,
      "language_loss": 0.7813319,
      "learning_rate": 1.759878411233777e-06,
      "loss": 0.8038767,
      "num_input_tokens_seen": 99083645,
      "step": 4592,
      "time_per_iteration": 2.675658702850342
    },
    {
      "auxiliary_loss_clip": 0.01232721,
      "auxiliary_loss_mlp": 0.010318,
      "balance_loss_clip": 1.05220127,
      "balance_loss_mlp": 1.02446544,
      "epoch": 0.5522755967053448,
      "flos": 18879999701760.0,
      "grad_norm": 2.2654592697140687,
      "language_loss": 0.76516283,
      "learning_rate": 1.7591050924783388e-06,
      "loss": 0.78780806,
      "num_input_tokens_seen": 99100835,
      "step": 4593,
      "time_per_iteration": 2.568741798400879
    },
    {
      "auxiliary_loss_clip": 0.01277512,
      "auxiliary_loss_mlp": 0.01005862,
      "balance_loss_clip": 1.00898504,
      "balance_loss_mlp": 1.00483119,
      "epoch": 0.5523958395959839,
      "flos": 64675622494080.0,
      "grad_norm": 2.0976446927768464,
      "language_loss": 0.57864106,
      "learning_rate": 1.7583318102664661e-06,
      "loss": 0.60147488,
      "num_input_tokens_seen": 99168400,
      "step": 4594,
      "time_per_iteration": 3.356858491897583
    },
    {
      "auxiliary_loss_clip": 0.01236042,
      "auxiliary_loss_mlp": 0.01024194,
      "balance_loss_clip": 1.04795313,
      "balance_loss_mlp": 1.01660085,
      "epoch": 0.552516082486623,
      "flos": 10889732211840.0,
      "grad_norm": 1.9252315158910271,
      "language_loss": 0.79389113,
      "learning_rate": 1.757558564715466e-06,
      "loss": 0.81649351,
      "num_input_tokens_seen": 99186475,
      "step": 4595,
      "time_per_iteration": 2.5958051681518555
    },
    {
      "auxiliary_loss_clip": 0.01235232,
      "auxiliary_loss_mlp": 0.01030965,
      "balance_loss_clip": 1.05102575,
      "balance_loss_mlp": 1.02326679,
      "epoch": 0.552636325377262,
      "flos": 22199797376640.0,
      "grad_norm": 2.7147134546213865,
      "language_loss": 0.73973954,
      "learning_rate": 1.7567853559426386e-06,
      "loss": 0.76240152,
      "num_input_tokens_seen": 99203525,
      "step": 4596,
      "time_per_iteration": 2.6043500900268555
    },
    {
      "auxiliary_loss_clip": 0.01236613,
      "auxiliary_loss_mlp": 0.01030497,
      "balance_loss_clip": 1.05240512,
      "balance_loss_mlp": 1.02332091,
      "epoch": 0.5527565682679012,
      "flos": 23988184652160.0,
      "grad_norm": 2.950695040998139,
      "language_loss": 0.75328052,
      "learning_rate": 1.7560121840652797e-06,
      "loss": 0.77595162,
      "num_input_tokens_seen": 99222910,
      "step": 4597,
      "time_per_iteration": 2.717092752456665
    },
    {
      "auxiliary_loss_clip": 0.01322061,
      "auxiliary_loss_mlp": 0.01027472,
      "balance_loss_clip": 1.04667473,
      "balance_loss_mlp": 1.01961017,
      "epoch": 0.5528768111585403,
      "flos": 19719267955200.0,
      "grad_norm": 1.909875245096945,
      "language_loss": 0.69231892,
      "learning_rate": 1.7552390492006782e-06,
      "loss": 0.71581429,
      "num_input_tokens_seen": 99241230,
      "step": 4598,
      "time_per_iteration": 2.6672170162200928
    },
    {
      "auxiliary_loss_clip": 0.01388209,
      "auxiliary_loss_mlp": 0.02572355,
      "balance_loss_clip": 1.0440619,
      "balance_loss_mlp": 0.99999082,
      "epoch": 0.5529970540491793,
      "flos": 26215975002240.0,
      "grad_norm": 2.088140350309763,
      "language_loss": 0.64853007,
      "learning_rate": 1.7544659514661184e-06,
      "loss": 0.68813574,
      "num_input_tokens_seen": 99264320,
      "step": 4599,
      "time_per_iteration": 2.7999765872955322
    },
    {
      "auxiliary_loss_clip": 0.01278701,
      "auxiliary_loss_mlp": 0.01027726,
      "balance_loss_clip": 1.04581499,
      "balance_loss_mlp": 1.02039754,
      "epoch": 0.5531172969398185,
      "flos": 24425971614720.0,
      "grad_norm": 2.044326664767469,
      "language_loss": 0.79761249,
      "learning_rate": 1.7536928909788786e-06,
      "loss": 0.8206768,
      "num_input_tokens_seen": 99283625,
      "step": 4600,
      "time_per_iteration": 2.6741347312927246
    },
    {
      "auxiliary_loss_clip": 0.01185131,
      "auxiliary_loss_mlp": 0.01002928,
      "balance_loss_clip": 1.01003587,
      "balance_loss_mlp": 1.00190914,
      "epoch": 0.5532375398304575,
      "flos": 64907316195840.0,
      "grad_norm": 0.8788862221009657,
      "language_loss": 0.61936533,
      "learning_rate": 1.752919867856231e-06,
      "loss": 0.64124584,
      "num_input_tokens_seen": 99335270,
      "step": 4601,
      "time_per_iteration": 3.145951747894287
    },
    {
      "auxiliary_loss_clip": 0.0127482,
      "auxiliary_loss_mlp": 0.01029756,
      "balance_loss_clip": 1.04664063,
      "balance_loss_mlp": 1.02234745,
      "epoch": 0.5533577827210966,
      "flos": 19683105937920.0,
      "grad_norm": 1.771566795645043,
      "language_loss": 0.78894484,
      "learning_rate": 1.7521468822154436e-06,
      "loss": 0.81199062,
      "num_input_tokens_seen": 99354185,
      "step": 4602,
      "time_per_iteration": 2.6239748001098633
    },
    {
      "auxiliary_loss_clip": 0.01276727,
      "auxiliary_loss_mlp": 0.01025697,
      "balance_loss_clip": 1.05110717,
      "balance_loss_mlp": 1.01855624,
      "epoch": 0.5534780256117358,
      "flos": 32306496076800.0,
      "grad_norm": 2.4641819182355658,
      "language_loss": 0.75349146,
      "learning_rate": 1.751373934173777e-06,
      "loss": 0.77651572,
      "num_input_tokens_seen": 99376930,
      "step": 4603,
      "time_per_iteration": 2.730943441390991
    },
    {
      "auxiliary_loss_clip": 0.01184125,
      "auxiliary_loss_mlp": 0.01023418,
      "balance_loss_clip": 1.05093706,
      "balance_loss_mlp": 1.01603293,
      "epoch": 0.5535982685023748,
      "flos": 23222425582080.0,
      "grad_norm": 1.6760280858645769,
      "language_loss": 0.7351684,
      "learning_rate": 1.750601023848487e-06,
      "loss": 0.75724387,
      "num_input_tokens_seen": 99397655,
      "step": 4604,
      "time_per_iteration": 2.620096445083618
    },
    {
      "auxiliary_loss_clip": 0.01181776,
      "auxiliary_loss_mlp": 0.02568561,
      "balance_loss_clip": 1.05286217,
      "balance_loss_mlp": 1.00001252,
      "epoch": 0.5537185113930139,
      "flos": 24352534258560.0,
      "grad_norm": 2.0717294825212202,
      "language_loss": 0.74182379,
      "learning_rate": 1.749828151356823e-06,
      "loss": 0.77932715,
      "num_input_tokens_seen": 99417850,
      "step": 4605,
      "time_per_iteration": 2.6339786052703857
    },
    {
      "auxiliary_loss_clip": 0.01283453,
      "auxiliary_loss_mlp": 0.01029421,
      "balance_loss_clip": 1.05067563,
      "balance_loss_mlp": 1.02258742,
      "epoch": 0.553838754283653,
      "flos": 23549068886400.0,
      "grad_norm": 1.7188970466643212,
      "language_loss": 0.75727987,
      "learning_rate": 1.7490553168160297e-06,
      "loss": 0.78040862,
      "num_input_tokens_seen": 99438920,
      "step": 4606,
      "time_per_iteration": 2.645874500274658
    },
    {
      "auxiliary_loss_clip": 0.01282672,
      "auxiliary_loss_mlp": 0.01029457,
      "balance_loss_clip": 1.04875743,
      "balance_loss_mlp": 1.02234936,
      "epoch": 0.5539589971742921,
      "flos": 17275044205440.0,
      "grad_norm": 2.258578803181166,
      "language_loss": 0.76319629,
      "learning_rate": 1.748282520343345e-06,
      "loss": 0.78631759,
      "num_input_tokens_seen": 99457950,
      "step": 4607,
      "time_per_iteration": 2.6670987606048584
    },
    {
      "auxiliary_loss_clip": 0.01241606,
      "auxiliary_loss_mlp": 0.01027349,
      "balance_loss_clip": 1.05173945,
      "balance_loss_mlp": 1.01950538,
      "epoch": 0.5540792400649311,
      "flos": 27564169104000.0,
      "grad_norm": 2.245526154493969,
      "language_loss": 0.78669131,
      "learning_rate": 1.7475097620560023e-06,
      "loss": 0.80938089,
      "num_input_tokens_seen": 99478015,
      "step": 4608,
      "time_per_iteration": 3.649614095687866
    },
    {
      "auxiliary_loss_clip": 0.01183915,
      "auxiliary_loss_mlp": 0.01028948,
      "balance_loss_clip": 1.05322611,
      "balance_loss_mlp": 1.02134287,
      "epoch": 0.5541994829555702,
      "flos": 23878657105920.0,
      "grad_norm": 2.2809867914893207,
      "language_loss": 0.70971668,
      "learning_rate": 1.746737042071228e-06,
      "loss": 0.73184526,
      "num_input_tokens_seen": 99496520,
      "step": 4609,
      "time_per_iteration": 2.6223933696746826
    },
    {
      "auxiliary_loss_clip": 0.01273625,
      "auxiliary_loss_mlp": 0.01028577,
      "balance_loss_clip": 1.04935086,
      "balance_loss_mlp": 1.02122509,
      "epoch": 0.5543197258462094,
      "flos": 20115721342080.0,
      "grad_norm": 2.9668012067239125,
      "language_loss": 0.79102975,
      "learning_rate": 1.7459643605062424e-06,
      "loss": 0.81405175,
      "num_input_tokens_seen": 99513780,
      "step": 4610,
      "time_per_iteration": 3.5975637435913086
    },
    {
      "auxiliary_loss_clip": 0.01375754,
      "auxiliary_loss_mlp": 0.01029461,
      "balance_loss_clip": 1.04833913,
      "balance_loss_mlp": 1.02201891,
      "epoch": 0.5544399687368484,
      "flos": 20916565021440.0,
      "grad_norm": 3.5795488261394373,
      "language_loss": 0.80638444,
      "learning_rate": 1.745191717478262e-06,
      "loss": 0.83043659,
      "num_input_tokens_seen": 99532360,
      "step": 4611,
      "time_per_iteration": 2.77559232711792
    },
    {
      "auxiliary_loss_clip": 0.01284747,
      "auxiliary_loss_mlp": 0.01032397,
      "balance_loss_clip": 1.05256796,
      "balance_loss_mlp": 1.02535152,
      "epoch": 0.5545602116274875,
      "flos": 25518661297920.0,
      "grad_norm": 2.1804098760504633,
      "language_loss": 0.79651022,
      "learning_rate": 1.7444191131044948e-06,
      "loss": 0.81968164,
      "num_input_tokens_seen": 99552635,
      "step": 4612,
      "time_per_iteration": 2.7006454467773438
    },
    {
      "auxiliary_loss_clip": 0.01281095,
      "auxiliary_loss_mlp": 0.01027196,
      "balance_loss_clip": 1.0508945,
      "balance_loss_mlp": 1.01933396,
      "epoch": 0.5546804545181266,
      "flos": 20995568985600.0,
      "grad_norm": 2.1170997143835835,
      "language_loss": 0.72867382,
      "learning_rate": 1.7436465475021456e-06,
      "loss": 0.75175667,
      "num_input_tokens_seen": 99572685,
      "step": 4613,
      "time_per_iteration": 2.777911424636841
    },
    {
      "auxiliary_loss_clip": 0.01324994,
      "auxiliary_loss_mlp": 0.01030577,
      "balance_loss_clip": 1.0481894,
      "balance_loss_mlp": 1.02321613,
      "epoch": 0.5548006974087657,
      "flos": 26833638297600.0,
      "grad_norm": 2.4249659414913034,
      "language_loss": 0.71457011,
      "learning_rate": 1.7428740207884111e-06,
      "loss": 0.7381258,
      "num_input_tokens_seen": 99593565,
      "step": 4614,
      "time_per_iteration": 3.6802845001220703
    },
    {
      "auxiliary_loss_clip": 0.01285062,
      "auxiliary_loss_mlp": 0.0102707,
      "balance_loss_clip": 1.0479939,
      "balance_loss_mlp": 1.01973248,
      "epoch": 0.5549209402994048,
      "flos": 33656414031360.0,
      "grad_norm": 1.605776057227853,
      "language_loss": 0.61057639,
      "learning_rate": 1.7421015330804833e-06,
      "loss": 0.63369769,
      "num_input_tokens_seen": 99613485,
      "step": 4615,
      "time_per_iteration": 2.9002208709716797
    },
    {
      "auxiliary_loss_clip": 0.01182592,
      "auxiliary_loss_mlp": 0.01026041,
      "balance_loss_clip": 1.05117118,
      "balance_loss_mlp": 1.01842999,
      "epoch": 0.5550411831900439,
      "flos": 23769524609280.0,
      "grad_norm": 1.9911530438905423,
      "language_loss": 0.72632152,
      "learning_rate": 1.7413290844955475e-06,
      "loss": 0.7484079,
      "num_input_tokens_seen": 99633515,
      "step": 4616,
      "time_per_iteration": 3.5866050720214844
    },
    {
      "auxiliary_loss_clip": 0.0122906,
      "auxiliary_loss_mlp": 0.01032885,
      "balance_loss_clip": 1.05109692,
      "balance_loss_mlp": 1.0256846,
      "epoch": 0.555161426080683,
      "flos": 21651189978240.0,
      "grad_norm": 1.8771519641287169,
      "language_loss": 0.78017861,
      "learning_rate": 1.7405566751507843e-06,
      "loss": 0.80279803,
      "num_input_tokens_seen": 99651560,
      "step": 4617,
      "time_per_iteration": 2.5815443992614746
    },
    {
      "auxiliary_loss_clip": 0.01328748,
      "auxiliary_loss_mlp": 0.01030318,
      "balance_loss_clip": 1.04720891,
      "balance_loss_mlp": 1.02318931,
      "epoch": 0.555281668971322,
      "flos": 49563116605440.0,
      "grad_norm": 1.486885687801532,
      "language_loss": 0.67416346,
      "learning_rate": 1.7397843051633668e-06,
      "loss": 0.69775414,
      "num_input_tokens_seen": 99674255,
      "step": 4618,
      "time_per_iteration": 2.9624972343444824
    },
    {
      "auxiliary_loss_clip": 0.01230279,
      "auxiliary_loss_mlp": 0.01030175,
      "balance_loss_clip": 1.05186629,
      "balance_loss_mlp": 1.02240229,
      "epoch": 0.5554019118619612,
      "flos": 20741608851840.0,
      "grad_norm": 1.6952314365326349,
      "language_loss": 0.7139436,
      "learning_rate": 1.739011974650464e-06,
      "loss": 0.73654813,
      "num_input_tokens_seen": 99693585,
      "step": 4619,
      "time_per_iteration": 2.5826756954193115
    },
    {
      "auxiliary_loss_clip": 0.01381963,
      "auxiliary_loss_mlp": 0.0102563,
      "balance_loss_clip": 1.04642737,
      "balance_loss_mlp": 1.01754785,
      "epoch": 0.5555221547526003,
      "flos": 25483217552640.0,
      "grad_norm": 1.9855115091093647,
      "language_loss": 0.76956093,
      "learning_rate": 1.7382396837292365e-06,
      "loss": 0.79363686,
      "num_input_tokens_seen": 99714045,
      "step": 4620,
      "time_per_iteration": 2.773322820663452
    },
    {
      "auxiliary_loss_clip": 0.01183577,
      "auxiliary_loss_mlp": 0.0103249,
      "balance_loss_clip": 1.0527215,
      "balance_loss_mlp": 1.02445579,
      "epoch": 0.5556423976432393,
      "flos": 21762513204480.0,
      "grad_norm": 2.29905138215585,
      "language_loss": 0.73556507,
      "learning_rate": 1.737467432516841e-06,
      "loss": 0.75772572,
      "num_input_tokens_seen": 99734145,
      "step": 4621,
      "time_per_iteration": 2.607388734817505
    },
    {
      "auxiliary_loss_clip": 0.01282279,
      "auxiliary_loss_mlp": 0.01026685,
      "balance_loss_clip": 1.046296,
      "balance_loss_mlp": 1.01907325,
      "epoch": 0.5557626405338785,
      "flos": 24900171989760.0,
      "grad_norm": 3.4753835554769714,
      "language_loss": 0.74679452,
      "learning_rate": 1.7366952211304274e-06,
      "loss": 0.76988423,
      "num_input_tokens_seen": 99751990,
      "step": 4622,
      "time_per_iteration": 2.668647050857544
    },
    {
      "auxiliary_loss_clip": 0.0127583,
      "auxiliary_loss_mlp": 0.01025084,
      "balance_loss_clip": 1.04861283,
      "balance_loss_mlp": 1.01797307,
      "epoch": 0.5558828834245175,
      "flos": 18697501676160.0,
      "grad_norm": 2.7817891683496003,
      "language_loss": 0.83466876,
      "learning_rate": 1.735923049687139e-06,
      "loss": 0.85767794,
      "num_input_tokens_seen": 99768565,
      "step": 4623,
      "time_per_iteration": 2.618623971939087
    },
    {
      "auxiliary_loss_clip": 0.01278129,
      "auxiliary_loss_mlp": 0.01029634,
      "balance_loss_clip": 1.04580212,
      "balance_loss_mlp": 1.0219394,
      "epoch": 0.5560031263151566,
      "flos": 27272179445760.0,
      "grad_norm": 1.5203832898271694,
      "language_loss": 0.73956919,
      "learning_rate": 1.7351509183041144e-06,
      "loss": 0.76264679,
      "num_input_tokens_seen": 99788895,
      "step": 4624,
      "time_per_iteration": 2.712900161743164
    },
    {
      "auxiliary_loss_clip": 0.01186175,
      "auxiliary_loss_mlp": 0.01028498,
      "balance_loss_clip": 1.05372369,
      "balance_loss_mlp": 1.02072561,
      "epoch": 0.5561233692057957,
      "flos": 23403738458880.0,
      "grad_norm": 1.8114967153717172,
      "language_loss": 0.71930814,
      "learning_rate": 1.7343788270984852e-06,
      "loss": 0.74145484,
      "num_input_tokens_seen": 99808035,
      "step": 4625,
      "time_per_iteration": 2.61588454246521
    },
    {
      "auxiliary_loss_clip": 0.01281217,
      "auxiliary_loss_mlp": 0.01021129,
      "balance_loss_clip": 1.05168366,
      "balance_loss_mlp": 1.01360035,
      "epoch": 0.5562436120964348,
      "flos": 37670867804160.0,
      "grad_norm": 1.9595745358462602,
      "language_loss": 0.74957156,
      "learning_rate": 1.7336067761873764e-06,
      "loss": 0.77259493,
      "num_input_tokens_seen": 99830460,
      "step": 4626,
      "time_per_iteration": 2.8278305530548096
    },
    {
      "auxiliary_loss_clip": 0.01238717,
      "auxiliary_loss_mlp": 0.01032531,
      "balance_loss_clip": 1.05077291,
      "balance_loss_mlp": 1.02431726,
      "epoch": 0.5563638549870739,
      "flos": 25155245445120.0,
      "grad_norm": 1.8352517639817474,
      "language_loss": 0.76375127,
      "learning_rate": 1.7328347656879076e-06,
      "loss": 0.78646374,
      "num_input_tokens_seen": 99850320,
      "step": 4627,
      "time_per_iteration": 2.641871213912964
    },
    {
      "auxiliary_loss_clip": 0.01333469,
      "auxiliary_loss_mlp": 0.01026631,
      "balance_loss_clip": 1.04497838,
      "balance_loss_mlp": 1.01938021,
      "epoch": 0.556484097877713,
      "flos": 13581810783360.0,
      "grad_norm": 4.499199770203564,
      "language_loss": 0.68640804,
      "learning_rate": 1.7320627957171927e-06,
      "loss": 0.7100091,
      "num_input_tokens_seen": 99864980,
      "step": 4628,
      "time_per_iteration": 2.643577814102173
    },
    {
      "auxiliary_loss_clip": 0.01184262,
      "auxiliary_loss_mlp": 0.01026117,
      "balance_loss_clip": 1.05455232,
      "balance_loss_mlp": 1.01828504,
      "epoch": 0.5566043407683521,
      "flos": 24681368292480.0,
      "grad_norm": 2.9036872126386633,
      "language_loss": 0.81494492,
      "learning_rate": 1.7312908663923382e-06,
      "loss": 0.83704865,
      "num_input_tokens_seen": 99881155,
      "step": 4629,
      "time_per_iteration": 2.647935628890991
    },
    {
      "auxiliary_loss_clip": 0.01225841,
      "auxiliary_loss_mlp": 0.01024729,
      "balance_loss_clip": 1.04697347,
      "balance_loss_mlp": 1.0170517,
      "epoch": 0.5567245836589911,
      "flos": 20588161950720.0,
      "grad_norm": 2.756942277045708,
      "language_loss": 0.67861199,
      "learning_rate": 1.7305189778304463e-06,
      "loss": 0.70111763,
      "num_input_tokens_seen": 99899330,
      "step": 4630,
      "time_per_iteration": 2.5871338844299316
    },
    {
      "auxiliary_loss_clip": 0.01281891,
      "auxiliary_loss_mlp": 0.0102837,
      "balance_loss_clip": 1.05186605,
      "balance_loss_mlp": 1.02117562,
      "epoch": 0.5568448265496303,
      "flos": 20704189858560.0,
      "grad_norm": 1.8372563623035347,
      "language_loss": 0.800623,
      "learning_rate": 1.729747130148611e-06,
      "loss": 0.82372564,
      "num_input_tokens_seen": 99918525,
      "step": 4631,
      "time_per_iteration": 2.657994031906128
    },
    {
      "auxiliary_loss_clip": 0.01334805,
      "auxiliary_loss_mlp": 0.01026817,
      "balance_loss_clip": 1.04736495,
      "balance_loss_mlp": 1.01832628,
      "epoch": 0.5569650694402694,
      "flos": 25302910256640.0,
      "grad_norm": 1.9403013848577986,
      "language_loss": 0.76982647,
      "learning_rate": 1.7289753234639208e-06,
      "loss": 0.79344273,
      "num_input_tokens_seen": 99937500,
      "step": 4632,
      "time_per_iteration": 2.7456836700439453
    },
    {
      "auxiliary_loss_clip": 0.0123661,
      "auxiliary_loss_mlp": 0.01029433,
      "balance_loss_clip": 1.05145121,
      "balance_loss_mlp": 1.02158344,
      "epoch": 0.5570853123309084,
      "flos": 19712623939200.0,
      "grad_norm": 2.2537286677341184,
      "language_loss": 0.76653457,
      "learning_rate": 1.7282035578934592e-06,
      "loss": 0.789195,
      "num_input_tokens_seen": 99955665,
      "step": 4633,
      "time_per_iteration": 2.653689384460449
    },
    {
      "auxiliary_loss_clip": 0.01274655,
      "auxiliary_loss_mlp": 0.01029777,
      "balance_loss_clip": 1.05029702,
      "balance_loss_mlp": 1.02181363,
      "epoch": 0.5572055552215476,
      "flos": 16108091153280.0,
      "grad_norm": 1.8241874152798876,
      "language_loss": 0.7920717,
      "learning_rate": 1.727431833554301e-06,
      "loss": 0.81511593,
      "num_input_tokens_seen": 99974140,
      "step": 4634,
      "time_per_iteration": 3.549848794937134
    },
    {
      "auxiliary_loss_clip": 0.01339411,
      "auxiliary_loss_mlp": 0.01028095,
      "balance_loss_clip": 1.04276419,
      "balance_loss_mlp": 1.02076352,
      "epoch": 0.5573257981121866,
      "flos": 17128815937920.0,
      "grad_norm": 2.1164139413523437,
      "language_loss": 0.7776686,
      "learning_rate": 1.7266601505635175e-06,
      "loss": 0.80134362,
      "num_input_tokens_seen": 99991480,
      "step": 4635,
      "time_per_iteration": 2.6800999641418457
    },
    {
      "auxiliary_loss_clip": 0.0123185,
      "auxiliary_loss_mlp": 0.01033281,
      "balance_loss_clip": 1.05145025,
      "balance_loss_mlp": 1.0258007,
      "epoch": 0.5574460410028257,
      "flos": 18807029222400.0,
      "grad_norm": 3.1189186649040668,
      "language_loss": 0.75888985,
      "learning_rate": 1.7258885090381717e-06,
      "loss": 0.78154117,
      "num_input_tokens_seen": 100009520,
      "step": 4636,
      "time_per_iteration": 2.6703035831451416
    },
    {
      "auxiliary_loss_clip": 0.01285851,
      "auxiliary_loss_mlp": 0.01028033,
      "balance_loss_clip": 1.04968286,
      "balance_loss_mlp": 1.02054393,
      "epoch": 0.5575662838934649,
      "flos": 29642678530560.0,
      "grad_norm": 1.8935357211592778,
      "language_loss": 0.78752142,
      "learning_rate": 1.7251169090953213e-06,
      "loss": 0.8106603,
      "num_input_tokens_seen": 100029995,
      "step": 4637,
      "time_per_iteration": 3.7283318042755127
    },
    {
      "auxiliary_loss_clip": 0.01228387,
      "auxiliary_loss_mlp": 0.01026141,
      "balance_loss_clip": 1.04929137,
      "balance_loss_mlp": 1.01868403,
      "epoch": 0.5576865267841039,
      "flos": 22054466949120.0,
      "grad_norm": 3.7337356271998283,
      "language_loss": 0.76413727,
      "learning_rate": 1.7243453508520168e-06,
      "loss": 0.78668255,
      "num_input_tokens_seen": 100046980,
      "step": 4638,
      "time_per_iteration": 2.71344256401062
    },
    {
      "auxiliary_loss_clip": 0.01282653,
      "auxiliary_loss_mlp": 0.01033313,
      "balance_loss_clip": 1.04839945,
      "balance_loss_mlp": 1.02540684,
      "epoch": 0.557806769674743,
      "flos": 17196040241280.0,
      "grad_norm": 1.9499342363012724,
      "language_loss": 0.84583902,
      "learning_rate": 1.7235738344253038e-06,
      "loss": 0.86899865,
      "num_input_tokens_seen": 100060610,
      "step": 4639,
      "time_per_iteration": 2.6307010650634766
    },
    {
      "auxiliary_loss_clip": 0.01237735,
      "auxiliary_loss_mlp": 0.01026838,
      "balance_loss_clip": 1.05596423,
      "balance_loss_mlp": 1.01892841,
      "epoch": 0.557927012565382,
      "flos": 24712717887360.0,
      "grad_norm": 1.988011514503129,
      "language_loss": 0.82837552,
      "learning_rate": 1.72280235993222e-06,
      "loss": 0.85102123,
      "num_input_tokens_seen": 100078915,
      "step": 4640,
      "time_per_iteration": 3.596297264099121
    },
    {
      "auxiliary_loss_clip": 0.01231687,
      "auxiliary_loss_mlp": 0.02573475,
      "balance_loss_clip": 1.05292463,
      "balance_loss_mlp": 1.0000391,
      "epoch": 0.5580472554560212,
      "flos": 16983090460800.0,
      "grad_norm": 2.4213848103777327,
      "language_loss": 0.69776511,
      "learning_rate": 1.722030927489798e-06,
      "loss": 0.73581672,
      "num_input_tokens_seen": 100096195,
      "step": 4641,
      "time_per_iteration": 3.5228207111358643
    },
    {
      "auxiliary_loss_clip": 0.01332229,
      "auxiliary_loss_mlp": 0.01029436,
      "balance_loss_clip": 1.04988527,
      "balance_loss_mlp": 1.02099657,
      "epoch": 0.5581674983466602,
      "flos": 23509100027520.0,
      "grad_norm": 2.324881766587197,
      "language_loss": 0.74025601,
      "learning_rate": 1.7212595372150634e-06,
      "loss": 0.76387268,
      "num_input_tokens_seen": 100116175,
      "step": 4642,
      "time_per_iteration": 2.6850671768188477
    },
    {
      "auxiliary_loss_clip": 0.01180601,
      "auxiliary_loss_mlp": 0.01029287,
      "balance_loss_clip": 1.05148935,
      "balance_loss_mlp": 1.02183008,
      "epoch": 0.5582877412372993,
      "flos": 13480291969920.0,
      "grad_norm": 2.1310174329459106,
      "language_loss": 0.7307834,
      "learning_rate": 1.720488189225035e-06,
      "loss": 0.7528823,
      "num_input_tokens_seen": 100133875,
      "step": 4643,
      "time_per_iteration": 2.5933361053466797
    },
    {
      "auxiliary_loss_clip": 0.012352,
      "auxiliary_loss_mlp": 0.01023942,
      "balance_loss_clip": 1.0506165,
      "balance_loss_mlp": 1.01636016,
      "epoch": 0.5584079841279385,
      "flos": 21903605827200.0,
      "grad_norm": 2.3637434539954802,
      "language_loss": 0.79030955,
      "learning_rate": 1.7197168836367265e-06,
      "loss": 0.8129009,
      "num_input_tokens_seen": 100150685,
      "step": 4644,
      "time_per_iteration": 2.614452600479126
    },
    {
      "auxiliary_loss_clip": 0.01228032,
      "auxiliary_loss_mlp": 0.02568718,
      "balance_loss_clip": 1.04883265,
      "balance_loss_mlp": 1.00001359,
      "epoch": 0.5585282270185775,
      "flos": 18843550375680.0,
      "grad_norm": 2.1452355535023044,
      "language_loss": 0.81934577,
      "learning_rate": 1.7189456205671433e-06,
      "loss": 0.85731328,
      "num_input_tokens_seen": 100169530,
      "step": 4645,
      "time_per_iteration": 2.6258387565612793
    },
    {
      "auxiliary_loss_clip": 0.01139764,
      "auxiliary_loss_mlp": 0.01030478,
      "balance_loss_clip": 1.05206752,
      "balance_loss_mlp": 1.02211523,
      "epoch": 0.5586484699092166,
      "flos": 21868449390720.0,
      "grad_norm": 1.9522779551608997,
      "language_loss": 0.82291847,
      "learning_rate": 1.7181744001332866e-06,
      "loss": 0.84462088,
      "num_input_tokens_seen": 100188140,
      "step": 4646,
      "time_per_iteration": 2.6066930294036865
    },
    {
      "auxiliary_loss_clip": 0.01181779,
      "auxiliary_loss_mlp": 0.01027845,
      "balance_loss_clip": 1.0532968,
      "balance_loss_mlp": 1.02060318,
      "epoch": 0.5587687127998557,
      "flos": 22893232412160.0,
      "grad_norm": 1.941043804896725,
      "language_loss": 0.63564718,
      "learning_rate": 1.7174032224521493e-06,
      "loss": 0.65774345,
      "num_input_tokens_seen": 100206850,
      "step": 4647,
      "time_per_iteration": 2.6045773029327393
    },
    {
      "auxiliary_loss_clip": 0.01231789,
      "auxiliary_loss_mlp": 0.01029787,
      "balance_loss_clip": 1.05070364,
      "balance_loss_mlp": 1.0223093,
      "epoch": 0.5588889556904948,
      "flos": 20303067703680.0,
      "grad_norm": 1.8025397438608308,
      "language_loss": 0.69516063,
      "learning_rate": 1.7166320876407184e-06,
      "loss": 0.71777642,
      "num_input_tokens_seen": 100226270,
      "step": 4648,
      "time_per_iteration": 2.579118251800537
    },
    {
      "auxiliary_loss_clip": 0.01184809,
      "auxiliary_loss_mlp": 0.02572379,
      "balance_loss_clip": 1.05292201,
      "balance_loss_mlp": 1.00003338,
      "epoch": 0.5590091985811338,
      "flos": 16472153450880.0,
      "grad_norm": 1.9055282525413095,
      "language_loss": 0.6774435,
      "learning_rate": 1.7158609958159742e-06,
      "loss": 0.71501535,
      "num_input_tokens_seen": 100243675,
      "step": 4649,
      "time_per_iteration": 2.6062703132629395
    },
    {
      "auxiliary_loss_clip": 0.01352273,
      "auxiliary_loss_mlp": 0.01029042,
      "balance_loss_clip": 1.05060911,
      "balance_loss_mlp": 1.02115631,
      "epoch": 0.559129441471773,
      "flos": 14532186781440.0,
      "grad_norm": 2.140975441452478,
      "language_loss": 0.78457433,
      "learning_rate": 1.7150899470948911e-06,
      "loss": 0.80838752,
      "num_input_tokens_seen": 100258940,
      "step": 4650,
      "time_per_iteration": 2.778404712677002
    },
    {
      "auxiliary_loss_clip": 0.01175333,
      "auxiliary_loss_mlp": 0.01012218,
      "balance_loss_clip": 1.01083279,
      "balance_loss_mlp": 1.01122212,
      "epoch": 0.5592496843624121,
      "flos": 60521009852160.0,
      "grad_norm": 0.8036619941416573,
      "language_loss": 0.56622702,
      "learning_rate": 1.7143189415944365e-06,
      "loss": 0.58810252,
      "num_input_tokens_seen": 100323400,
      "step": 4651,
      "time_per_iteration": 3.252281665802002
    },
    {
      "auxiliary_loss_clip": 0.01228853,
      "auxiliary_loss_mlp": 0.01029665,
      "balance_loss_clip": 1.05189455,
      "balance_loss_mlp": 1.02201223,
      "epoch": 0.5593699272530511,
      "flos": 20886256920960.0,
      "grad_norm": 1.8019906175585627,
      "language_loss": 0.76631767,
      "learning_rate": 1.7135479794315714e-06,
      "loss": 0.78890288,
      "num_input_tokens_seen": 100340355,
      "step": 4652,
      "time_per_iteration": 2.6411983966827393
    },
    {
      "auxiliary_loss_clip": 0.01326746,
      "auxiliary_loss_mlp": 0.01029165,
      "balance_loss_clip": 1.04812968,
      "balance_loss_mlp": 1.02196157,
      "epoch": 0.5594901701436903,
      "flos": 12896743616640.0,
      "grad_norm": 3.443036862145537,
      "language_loss": 0.78918296,
      "learning_rate": 1.7127770607232502e-06,
      "loss": 0.81274211,
      "num_input_tokens_seen": 100358900,
      "step": 4653,
      "time_per_iteration": 2.619926929473877
    },
    {
      "auxiliary_loss_clip": 0.01335633,
      "auxiliary_loss_mlp": 0.01025404,
      "balance_loss_clip": 1.04651523,
      "balance_loss_mlp": 1.01837111,
      "epoch": 0.5596104130343293,
      "flos": 23112107936640.0,
      "grad_norm": 4.770354925760323,
      "language_loss": 0.79737961,
      "learning_rate": 1.7120061855864204e-06,
      "loss": 0.82099003,
      "num_input_tokens_seen": 100378910,
      "step": 4654,
      "time_per_iteration": 2.747706174850464
    },
    {
      "auxiliary_loss_clip": 0.0123117,
      "auxiliary_loss_mlp": 0.01030956,
      "balance_loss_clip": 1.05188608,
      "balance_loss_mlp": 1.02367282,
      "epoch": 0.5597306559249684,
      "flos": 25957812977280.0,
      "grad_norm": 1.9998948507211953,
      "language_loss": 0.71432221,
      "learning_rate": 1.7112353541380233e-06,
      "loss": 0.73694342,
      "num_input_tokens_seen": 100398770,
      "step": 4655,
      "time_per_iteration": 2.7289676666259766
    },
    {
      "auxiliary_loss_clip": 0.01280903,
      "auxiliary_loss_mlp": 0.01035243,
      "balance_loss_clip": 1.05080879,
      "balance_loss_mlp": 1.02807307,
      "epoch": 0.5598508988156076,
      "flos": 22492289825280.0,
      "grad_norm": 1.4554971271600567,
      "language_loss": 0.72495371,
      "learning_rate": 1.7104645664949931e-06,
      "loss": 0.74811518,
      "num_input_tokens_seen": 100421240,
      "step": 4656,
      "time_per_iteration": 2.8559632301330566
    },
    {
      "auxiliary_loss_clip": 0.0127771,
      "auxiliary_loss_mlp": 0.01029477,
      "balance_loss_clip": 1.04540801,
      "balance_loss_mlp": 1.02130556,
      "epoch": 0.5599711417062466,
      "flos": 23112538899840.0,
      "grad_norm": 1.8418306429117164,
      "language_loss": 0.71602023,
      "learning_rate": 1.7096938227742584e-06,
      "loss": 0.73909211,
      "num_input_tokens_seen": 100442370,
      "step": 4657,
      "time_per_iteration": 2.662653684616089
    },
    {
      "auxiliary_loss_clip": 0.01182643,
      "auxiliary_loss_mlp": 0.01028347,
      "balance_loss_clip": 1.05259192,
      "balance_loss_mlp": 1.0205332,
      "epoch": 0.5600913845968857,
      "flos": 22339345714560.0,
      "grad_norm": 1.8499586851300305,
      "language_loss": 0.84233725,
      "learning_rate": 1.70892312309274e-06,
      "loss": 0.86444718,
      "num_input_tokens_seen": 100460260,
      "step": 4658,
      "time_per_iteration": 2.54872465133667
    },
    {
      "auxiliary_loss_clip": 0.01280643,
      "auxiliary_loss_mlp": 0.01024743,
      "balance_loss_clip": 1.04306233,
      "balance_loss_mlp": 1.01711941,
      "epoch": 0.5602116274875248,
      "flos": 17633791290240.0,
      "grad_norm": 2.5275863332042934,
      "language_loss": 0.68290979,
      "learning_rate": 1.7081524675673523e-06,
      "loss": 0.70596361,
      "num_input_tokens_seen": 100475750,
      "step": 4659,
      "time_per_iteration": 2.668945789337158
    },
    {
      "auxiliary_loss_clip": 0.01178797,
      "auxiliary_loss_mlp": 0.01001901,
      "balance_loss_clip": 1.01051688,
      "balance_loss_mlp": 1.00095892,
      "epoch": 0.5603318703781639,
      "flos": 70115945529600.0,
      "grad_norm": 0.7765296955693761,
      "language_loss": 0.59574258,
      "learning_rate": 1.7073818563150026e-06,
      "loss": 0.61754954,
      "num_input_tokens_seen": 100537830,
      "step": 4660,
      "time_per_iteration": 4.205419063568115
    },
    {
      "auxiliary_loss_clip": 0.01231048,
      "auxiliary_loss_mlp": 0.01026662,
      "balance_loss_clip": 1.04844224,
      "balance_loss_mlp": 1.01829314,
      "epoch": 0.560452113268803,
      "flos": 18545850455040.0,
      "grad_norm": 2.7674134060749975,
      "language_loss": 0.86828279,
      "learning_rate": 1.7066112894525935e-06,
      "loss": 0.89085984,
      "num_input_tokens_seen": 100555910,
      "step": 4661,
      "time_per_iteration": 2.65032958984375
    },
    {
      "auxiliary_loss_clip": 0.01274972,
      "auxiliary_loss_mlp": 0.01033722,
      "balance_loss_clip": 1.04596639,
      "balance_loss_mlp": 1.02563953,
      "epoch": 0.5605723561594421,
      "flos": 25264665250560.0,
      "grad_norm": 2.255901216324383,
      "language_loss": 0.72723269,
      "learning_rate": 1.7058407670970177e-06,
      "loss": 0.75031966,
      "num_input_tokens_seen": 100577385,
      "step": 4662,
      "time_per_iteration": 2.693502426147461
    },
    {
      "auxiliary_loss_clip": 0.01138393,
      "auxiliary_loss_mlp": 0.01031612,
      "balance_loss_clip": 1.05013776,
      "balance_loss_mlp": 1.02397633,
      "epoch": 0.5606925990500812,
      "flos": 20594949621120.0,
      "grad_norm": 2.712008108302704,
      "language_loss": 0.614223,
      "learning_rate": 1.7050702893651643e-06,
      "loss": 0.63592303,
      "num_input_tokens_seen": 100596965,
      "step": 4663,
      "time_per_iteration": 3.478433132171631
    },
    {
      "auxiliary_loss_clip": 0.01232093,
      "auxiliary_loss_mlp": 0.01032615,
      "balance_loss_clip": 1.05217409,
      "balance_loss_mlp": 1.02510476,
      "epoch": 0.5608128419407202,
      "flos": 35006044677120.0,
      "grad_norm": 5.566435952659172,
      "language_loss": 0.75625336,
      "learning_rate": 1.7042998563739134e-06,
      "loss": 0.7789005,
      "num_input_tokens_seen": 100615315,
      "step": 4664,
      "time_per_iteration": 2.7922983169555664
    },
    {
      "auxiliary_loss_clip": 0.01288865,
      "auxiliary_loss_mlp": 0.01030156,
      "balance_loss_clip": 1.04789448,
      "balance_loss_mlp": 1.02208591,
      "epoch": 0.5609330848313594,
      "flos": 24639819235200.0,
      "grad_norm": 2.0382645362397733,
      "language_loss": 0.7170819,
      "learning_rate": 1.703529468240139e-06,
      "loss": 0.74027205,
      "num_input_tokens_seen": 100634185,
      "step": 4665,
      "time_per_iteration": 3.594461679458618
    },
    {
      "auxiliary_loss_clip": 0.01275824,
      "auxiliary_loss_mlp": 0.01028777,
      "balance_loss_clip": 1.04950833,
      "balance_loss_mlp": 1.02045608,
      "epoch": 0.5610533277219985,
      "flos": 18762894385920.0,
      "grad_norm": 2.5684094060693066,
      "language_loss": 0.73723954,
      "learning_rate": 1.7027591250807088e-06,
      "loss": 0.76028562,
      "num_input_tokens_seen": 100651360,
      "step": 4666,
      "time_per_iteration": 2.6891298294067383
    },
    {
      "auxiliary_loss_clip": 0.01183357,
      "auxiliary_loss_mlp": 0.01025575,
      "balance_loss_clip": 1.05176854,
      "balance_loss_mlp": 1.01812708,
      "epoch": 0.5611735706126375,
      "flos": 15012384727680.0,
      "grad_norm": 4.527439790597977,
      "language_loss": 0.84441602,
      "learning_rate": 1.7019888270124825e-06,
      "loss": 0.86650527,
      "num_input_tokens_seen": 100668525,
      "step": 4667,
      "time_per_iteration": 2.609100341796875
    },
    {
      "auxiliary_loss_clip": 0.01233213,
      "auxiliary_loss_mlp": 0.01031937,
      "balance_loss_clip": 1.05086374,
      "balance_loss_mlp": 1.02515149,
      "epoch": 0.5612938135032767,
      "flos": 16468167041280.0,
      "grad_norm": 2.139555887559749,
      "language_loss": 0.82388103,
      "learning_rate": 1.7012185741523147e-06,
      "loss": 0.84653252,
      "num_input_tokens_seen": 100684850,
      "step": 4668,
      "time_per_iteration": 3.48781418800354
    },
    {
      "auxiliary_loss_clip": 0.01184551,
      "auxiliary_loss_mlp": 0.01026453,
      "balance_loss_clip": 1.05373824,
      "balance_loss_mlp": 1.01868951,
      "epoch": 0.5614140563939157,
      "flos": 25666433850240.0,
      "grad_norm": 3.1512086763689324,
      "language_loss": 0.6303333,
      "learning_rate": 1.7004483666170514e-06,
      "loss": 0.65244341,
      "num_input_tokens_seen": 100705345,
      "step": 4669,
      "time_per_iteration": 2.658010721206665
    },
    {
      "auxiliary_loss_clip": 0.01225372,
      "auxiliary_loss_mlp": 0.01029865,
      "balance_loss_clip": 1.04705572,
      "balance_loss_mlp": 1.02256596,
      "epoch": 0.5615342992845548,
      "flos": 24717566223360.0,
      "grad_norm": 1.9576714254078929,
      "language_loss": 0.80472094,
      "learning_rate": 1.699678204523533e-06,
      "loss": 0.82727325,
      "num_input_tokens_seen": 100725210,
      "step": 4670,
      "time_per_iteration": 2.6189095973968506
    },
    {
      "auxiliary_loss_clip": 0.01282686,
      "auxiliary_loss_mlp": 0.01031352,
      "balance_loss_clip": 1.05229354,
      "balance_loss_mlp": 1.02298915,
      "epoch": 0.5616545421751938,
      "flos": 22015934634240.0,
      "grad_norm": 4.330300691770355,
      "language_loss": 0.68810105,
      "learning_rate": 1.6989080879885918e-06,
      "loss": 0.71124136,
      "num_input_tokens_seen": 100743070,
      "step": 4671,
      "time_per_iteration": 2.6749045848846436
    },
    {
      "auxiliary_loss_clip": 0.0122647,
      "auxiliary_loss_mlp": 0.01001655,
      "balance_loss_clip": 1.01111853,
      "balance_loss_mlp": 1.00070155,
      "epoch": 0.561774785065833,
      "flos": 53760358690560.0,
      "grad_norm": 0.9017038453871216,
      "language_loss": 0.60913855,
      "learning_rate": 1.6981380171290544e-06,
      "loss": 0.63141984,
      "num_input_tokens_seen": 100804095,
      "step": 4672,
      "time_per_iteration": 3.240976572036743
    },
    {
      "auxiliary_loss_clip": 0.01179461,
      "auxiliary_loss_mlp": 0.01028836,
      "balance_loss_clip": 1.04478621,
      "balance_loss_mlp": 1.0205152,
      "epoch": 0.5618950279564721,
      "flos": 19750007018880.0,
      "grad_norm": 1.9512698036326073,
      "language_loss": 0.74541628,
      "learning_rate": 1.6973679920617396e-06,
      "loss": 0.76749921,
      "num_input_tokens_seen": 100821630,
      "step": 4673,
      "time_per_iteration": 2.66615891456604
    },
    {
      "auxiliary_loss_clip": 0.0128069,
      "auxiliary_loss_mlp": 0.01022279,
      "balance_loss_clip": 1.05287766,
      "balance_loss_mlp": 1.01480508,
      "epoch": 0.5620152708471111,
      "flos": 16800592435200.0,
      "grad_norm": 2.808089972360874,
      "language_loss": 0.85596716,
      "learning_rate": 1.6965980129034603e-06,
      "loss": 0.87899685,
      "num_input_tokens_seen": 100839015,
      "step": 4674,
      "time_per_iteration": 2.6055829524993896
    },
    {
      "auxiliary_loss_clip": 0.01281279,
      "auxiliary_loss_mlp": 0.01024722,
      "balance_loss_clip": 1.05031753,
      "balance_loss_mlp": 1.01740301,
      "epoch": 0.5621355137377503,
      "flos": 26797799502720.0,
      "grad_norm": 1.7941612388210035,
      "language_loss": 0.7662794,
      "learning_rate": 1.6958280797710209e-06,
      "loss": 0.78933942,
      "num_input_tokens_seen": 100860940,
      "step": 4675,
      "time_per_iteration": 2.7271995544433594
    },
    {
      "auxiliary_loss_clip": 0.01169866,
      "auxiliary_loss_mlp": 0.00995839,
      "balance_loss_clip": 1.0104413,
      "balance_loss_mlp": 0.9948799,
      "epoch": 0.5622557566283893,
      "flos": 61207046686080.0,
      "grad_norm": 0.7130211753556019,
      "language_loss": 0.54751605,
      "learning_rate": 1.6950581927812198e-06,
      "loss": 0.5691731,
      "num_input_tokens_seen": 100920510,
      "step": 4676,
      "time_per_iteration": 3.107309579849243
    },
    {
      "auxiliary_loss_clip": 0.01231912,
      "auxiliary_loss_mlp": 0.01026716,
      "balance_loss_clip": 1.04997468,
      "balance_loss_mlp": 1.01946211,
      "epoch": 0.5623759995190284,
      "flos": 26468534505600.0,
      "grad_norm": 3.6765822988844357,
      "language_loss": 0.79359126,
      "learning_rate": 1.6942883520508486e-06,
      "loss": 0.81617749,
      "num_input_tokens_seen": 100939245,
      "step": 4677,
      "time_per_iteration": 2.649785041809082
    },
    {
      "auxiliary_loss_clip": 0.01134995,
      "auxiliary_loss_mlp": 0.01026436,
      "balance_loss_clip": 1.05162096,
      "balance_loss_mlp": 1.0193578,
      "epoch": 0.5624962424096676,
      "flos": 19390900798080.0,
      "grad_norm": 5.234084230818149,
      "language_loss": 0.77151108,
      "learning_rate": 1.693518557696691e-06,
      "loss": 0.79312539,
      "num_input_tokens_seen": 100958385,
      "step": 4678,
      "time_per_iteration": 2.668201208114624
    },
    {
      "auxiliary_loss_clip": 0.01228871,
      "auxiliary_loss_mlp": 0.01025598,
      "balance_loss_clip": 1.04681706,
      "balance_loss_mlp": 1.01789761,
      "epoch": 0.5626164853003066,
      "flos": 20667345482880.0,
      "grad_norm": 4.373260342334784,
      "language_loss": 0.89272559,
      "learning_rate": 1.6927488098355252e-06,
      "loss": 0.91527021,
      "num_input_tokens_seen": 100976015,
      "step": 4679,
      "time_per_iteration": 2.5730412006378174
    },
    {
      "auxiliary_loss_clip": 0.0119205,
      "auxiliary_loss_mlp": 0.00997553,
      "balance_loss_clip": 1.01012266,
      "balance_loss_mlp": 0.99659389,
      "epoch": 0.5627367281909457,
      "flos": 62766071665920.0,
      "grad_norm": 0.9146476598368409,
      "language_loss": 0.6318121,
      "learning_rate": 1.6919791085841201e-06,
      "loss": 0.65370822,
      "num_input_tokens_seen": 101033425,
      "step": 4680,
      "time_per_iteration": 3.2836501598358154
    },
    {
      "auxiliary_loss_clip": 0.01229185,
      "auxiliary_loss_mlp": 0.01024392,
      "balance_loss_clip": 1.04669976,
      "balance_loss_mlp": 1.01653063,
      "epoch": 0.5628569710815848,
      "flos": 12787144243200.0,
      "grad_norm": 2.5717088151973484,
      "language_loss": 0.78960043,
      "learning_rate": 1.6912094540592396e-06,
      "loss": 0.81213617,
      "num_input_tokens_seen": 101048945,
      "step": 4681,
      "time_per_iteration": 2.5542104244232178
    },
    {
      "auxiliary_loss_clip": 0.01227957,
      "auxiliary_loss_mlp": 0.01031893,
      "balance_loss_clip": 1.04979849,
      "balance_loss_mlp": 1.02438867,
      "epoch": 0.5629772139722239,
      "flos": 13762082165760.0,
      "grad_norm": 2.7073232235888414,
      "language_loss": 0.81776285,
      "learning_rate": 1.6904398463776393e-06,
      "loss": 0.84036136,
      "num_input_tokens_seen": 101062745,
      "step": 4682,
      "time_per_iteration": 2.610830068588257
    },
    {
      "auxiliary_loss_clip": 0.01234636,
      "auxiliary_loss_mlp": 0.01023576,
      "balance_loss_clip": 1.04955316,
      "balance_loss_mlp": 1.01622939,
      "epoch": 0.5630974568628629,
      "flos": 21467830026240.0,
      "grad_norm": 2.3524799253325135,
      "language_loss": 0.72755492,
      "learning_rate": 1.6896702856560683e-06,
      "loss": 0.75013697,
      "num_input_tokens_seen": 101081840,
      "step": 4683,
      "time_per_iteration": 2.601748466491699
    },
    {
      "auxiliary_loss_clip": 0.01325958,
      "auxiliary_loss_mlp": 0.01029405,
      "balance_loss_clip": 1.04162431,
      "balance_loss_mlp": 1.02182364,
      "epoch": 0.5632176997535021,
      "flos": 14245907385600.0,
      "grad_norm": 3.531567475944299,
      "language_loss": 0.69412905,
      "learning_rate": 1.6889007720112677e-06,
      "loss": 0.71768272,
      "num_input_tokens_seen": 101099585,
      "step": 4684,
      "time_per_iteration": 2.730217695236206
    },
    {
      "auxiliary_loss_clip": 0.01232737,
      "auxiliary_loss_mlp": 0.01023521,
      "balance_loss_clip": 1.05011964,
      "balance_loss_mlp": 1.01574826,
      "epoch": 0.5633379426441412,
      "flos": 20812244947200.0,
      "grad_norm": 1.649058852890975,
      "language_loss": 0.77249432,
      "learning_rate": 1.6881313055599734e-06,
      "loss": 0.79505694,
      "num_input_tokens_seen": 101119515,
      "step": 4685,
      "time_per_iteration": 2.657871961593628
    },
    {
      "auxiliary_loss_clip": 0.0127332,
      "auxiliary_loss_mlp": 0.01032263,
      "balance_loss_clip": 1.04474604,
      "balance_loss_mlp": 1.02379906,
      "epoch": 0.5634581855347802,
      "flos": 22600883617920.0,
      "grad_norm": 2.425841398581771,
      "language_loss": 0.82448572,
      "learning_rate": 1.6873618864189117e-06,
      "loss": 0.84754157,
      "num_input_tokens_seen": 101135285,
      "step": 4686,
      "time_per_iteration": 3.6981089115142822
    },
    {
      "auxiliary_loss_clip": 0.0122895,
      "auxiliary_loss_mlp": 0.01026774,
      "balance_loss_clip": 1.04831648,
      "balance_loss_mlp": 1.01932597,
      "epoch": 0.5635784284254194,
      "flos": 21506972872320.0,
      "grad_norm": 3.6422782039701285,
      "language_loss": 0.7790885,
      "learning_rate": 1.686592514704803e-06,
      "loss": 0.80164576,
      "num_input_tokens_seen": 101152680,
      "step": 4687,
      "time_per_iteration": 2.6639463901519775
    },
    {
      "auxiliary_loss_clip": 0.0127949,
      "auxiliary_loss_mlp": 0.01025295,
      "balance_loss_clip": 1.05100131,
      "balance_loss_mlp": 1.0181452,
      "epoch": 0.5636986713160584,
      "flos": 19827466698240.0,
      "grad_norm": 4.0972591405213725,
      "language_loss": 0.70670587,
      "learning_rate": 1.685823190534361e-06,
      "loss": 0.72975367,
      "num_input_tokens_seen": 101170920,
      "step": 4688,
      "time_per_iteration": 2.6268081665039062
    },
    {
      "auxiliary_loss_clip": 0.01184544,
      "auxiliary_loss_mlp": 0.01028977,
      "balance_loss_clip": 1.05038559,
      "balance_loss_mlp": 1.02096653,
      "epoch": 0.5638189142066975,
      "flos": 19792453916160.0,
      "grad_norm": 1.980051542105391,
      "language_loss": 0.83885252,
      "learning_rate": 1.6850539140242907e-06,
      "loss": 0.86098772,
      "num_input_tokens_seen": 101190180,
      "step": 4689,
      "time_per_iteration": 3.5100197792053223
    },
    {
      "auxiliary_loss_clip": 0.01135303,
      "auxiliary_loss_mlp": 0.01028939,
      "balance_loss_clip": 1.04996407,
      "balance_loss_mlp": 1.02122617,
      "epoch": 0.5639391570973367,
      "flos": 22893771116160.0,
      "grad_norm": 1.9922590216665472,
      "language_loss": 0.82043463,
      "learning_rate": 1.684284685291292e-06,
      "loss": 0.84207708,
      "num_input_tokens_seen": 101211825,
      "step": 4690,
      "time_per_iteration": 2.6841273307800293
    },
    {
      "auxiliary_loss_clip": 0.01183124,
      "auxiliary_loss_mlp": 0.01026597,
      "balance_loss_clip": 1.05174899,
      "balance_loss_mlp": 1.01865149,
      "epoch": 0.5640593999879757,
      "flos": 23727077712000.0,
      "grad_norm": 2.033110568566966,
      "language_loss": 0.81385946,
      "learning_rate": 1.683515504452055e-06,
      "loss": 0.83595669,
      "num_input_tokens_seen": 101229200,
      "step": 4691,
      "time_per_iteration": 3.450983762741089
    },
    {
      "auxiliary_loss_clip": 0.01322503,
      "auxiliary_loss_mlp": 0.01029144,
      "balance_loss_clip": 1.04455781,
      "balance_loss_mlp": 1.02097201,
      "epoch": 0.5641796428786148,
      "flos": 22710123855360.0,
      "grad_norm": 1.6035310502849736,
      "language_loss": 0.66584229,
      "learning_rate": 1.6827463716232648e-06,
      "loss": 0.68935883,
      "num_input_tokens_seen": 101249860,
      "step": 4692,
      "time_per_iteration": 2.7288155555725098
    },
    {
      "auxiliary_loss_clip": 0.01231856,
      "auxiliary_loss_mlp": 0.02568757,
      "balance_loss_clip": 1.04845953,
      "balance_loss_mlp": 1.00007129,
      "epoch": 0.5642998857692539,
      "flos": 19791987039360.0,
      "grad_norm": 1.843466453379911,
      "language_loss": 0.75746787,
      "learning_rate": 1.6819772869215972e-06,
      "loss": 0.79547393,
      "num_input_tokens_seen": 101268940,
      "step": 4693,
      "time_per_iteration": 2.599470615386963
    },
    {
      "auxiliary_loss_clip": 0.0128717,
      "auxiliary_loss_mlp": 0.01026992,
      "balance_loss_clip": 1.05018401,
      "balance_loss_mlp": 1.01963651,
      "epoch": 0.564420128659893,
      "flos": 23185904428800.0,
      "grad_norm": 1.6840945608597229,
      "language_loss": 0.8209182,
      "learning_rate": 1.6812082504637228e-06,
      "loss": 0.84405982,
      "num_input_tokens_seen": 101290260,
      "step": 4694,
      "time_per_iteration": 3.590618848800659
    },
    {
      "auxiliary_loss_clip": 0.01228047,
      "auxiliary_loss_mlp": 0.0102421,
      "balance_loss_clip": 1.05165887,
      "balance_loss_mlp": 1.01715553,
      "epoch": 0.564540371550532,
      "flos": 23258264376960.0,
      "grad_norm": 1.4921170230097676,
      "language_loss": 0.74489367,
      "learning_rate": 1.6804392623663025e-06,
      "loss": 0.7674163,
      "num_input_tokens_seen": 101311465,
      "step": 4695,
      "time_per_iteration": 2.6966826915740967
    },
    {
      "auxiliary_loss_clip": 0.01225576,
      "auxiliary_loss_mlp": 0.01022924,
      "balance_loss_clip": 1.0486846,
      "balance_loss_mlp": 1.01560438,
      "epoch": 0.5646606144411712,
      "flos": 25010058672000.0,
      "grad_norm": 1.824233541456398,
      "language_loss": 0.78414363,
      "learning_rate": 1.6796703227459935e-06,
      "loss": 0.80662864,
      "num_input_tokens_seen": 101329420,
      "step": 4696,
      "time_per_iteration": 2.6232736110687256
    },
    {
      "auxiliary_loss_clip": 0.01373189,
      "auxiliary_loss_mlp": 0.01027794,
      "balance_loss_clip": 1.04246616,
      "balance_loss_mlp": 1.02000988,
      "epoch": 0.5647808573318103,
      "flos": 36539645806080.0,
      "grad_norm": 1.9825329287984166,
      "language_loss": 0.75894904,
      "learning_rate": 1.6789014317194407e-06,
      "loss": 0.78295887,
      "num_input_tokens_seen": 101350900,
      "step": 4697,
      "time_per_iteration": 2.8789963722229004
    },
    {
      "auxiliary_loss_clip": 0.01188601,
      "auxiliary_loss_mlp": 0.01030787,
      "balance_loss_clip": 1.04987526,
      "balance_loss_mlp": 1.02182293,
      "epoch": 0.5649011002224493,
      "flos": 22528451842560.0,
      "grad_norm": 2.646647085873654,
      "language_loss": 0.73006546,
      "learning_rate": 1.6781325894032853e-06,
      "loss": 0.75225937,
      "num_input_tokens_seen": 101369860,
      "step": 4698,
      "time_per_iteration": 2.6647844314575195
    },
    {
      "auxiliary_loss_clip": 0.01281907,
      "auxiliary_loss_mlp": 0.01037256,
      "balance_loss_clip": 1.05253637,
      "balance_loss_mlp": 1.0298059,
      "epoch": 0.5650213431130885,
      "flos": 18515147304960.0,
      "grad_norm": 3.2959709536054587,
      "language_loss": 0.92130446,
      "learning_rate": 1.6773637959141608e-06,
      "loss": 0.94449604,
      "num_input_tokens_seen": 101386835,
      "step": 4699,
      "time_per_iteration": 2.6307921409606934
    },
    {
      "auxiliary_loss_clip": 0.01276514,
      "auxiliary_loss_mlp": 0.01027453,
      "balance_loss_clip": 1.04816806,
      "balance_loss_mlp": 1.02007365,
      "epoch": 0.5651415860037275,
      "flos": 17526310819200.0,
      "grad_norm": 2.3929731708763375,
      "language_loss": 0.664846,
      "learning_rate": 1.6765950513686915e-06,
      "loss": 0.6878857,
      "num_input_tokens_seen": 101404945,
      "step": 4700,
      "time_per_iteration": 2.637324571609497
    },
    {
      "auxiliary_loss_clip": 0.01280724,
      "auxiliary_loss_mlp": 0.01025011,
      "balance_loss_clip": 1.04278123,
      "balance_loss_mlp": 1.01696408,
      "epoch": 0.5652618288943666,
      "flos": 25520026014720.0,
      "grad_norm": 6.03203248891467,
      "language_loss": 0.76383317,
      "learning_rate": 1.675826355883496e-06,
      "loss": 0.78689057,
      "num_input_tokens_seen": 101424160,
      "step": 4701,
      "time_per_iteration": 2.783923625946045
    },
    {
      "auxiliary_loss_clip": 0.01276535,
      "auxiliary_loss_mlp": 0.01025246,
      "balance_loss_clip": 1.05049276,
      "balance_loss_mlp": 1.01746488,
      "epoch": 0.5653820717850057,
      "flos": 19683105937920.0,
      "grad_norm": 2.197176249053272,
      "language_loss": 0.79145443,
      "learning_rate": 1.6750577095751848e-06,
      "loss": 0.81447226,
      "num_input_tokens_seen": 101443270,
      "step": 4702,
      "time_per_iteration": 2.6167078018188477
    },
    {
      "auxiliary_loss_clip": 0.01180372,
      "auxiliary_loss_mlp": 0.01027787,
      "balance_loss_clip": 1.05182016,
      "balance_loss_mlp": 1.02035403,
      "epoch": 0.5655023146756448,
      "flos": 26979722910720.0,
      "grad_norm": 1.6730662366314633,
      "language_loss": 0.73080349,
      "learning_rate": 1.6742891125603605e-06,
      "loss": 0.75288498,
      "num_input_tokens_seen": 101464175,
      "step": 4703,
      "time_per_iteration": 2.6359941959381104
    },
    {
      "auxiliary_loss_clip": 0.01230644,
      "auxiliary_loss_mlp": 0.01023222,
      "balance_loss_clip": 1.05044568,
      "balance_loss_mlp": 1.01530623,
      "epoch": 0.5656225575662839,
      "flos": 27669351104640.0,
      "grad_norm": 2.0248688320972748,
      "language_loss": 0.7241081,
      "learning_rate": 1.6735205649556185e-06,
      "loss": 0.74664676,
      "num_input_tokens_seen": 101484045,
      "step": 4704,
      "time_per_iteration": 2.6490819454193115
    },
    {
      "auxiliary_loss_clip": 0.01233289,
      "auxiliary_loss_mlp": 0.01026673,
      "balance_loss_clip": 1.04736328,
      "balance_loss_mlp": 1.0195142,
      "epoch": 0.5657428004569229,
      "flos": 24349732997760.0,
      "grad_norm": 1.9860805306156282,
      "language_loss": 0.85010183,
      "learning_rate": 1.6727520668775476e-06,
      "loss": 0.87270141,
      "num_input_tokens_seen": 101504330,
      "step": 4705,
      "time_per_iteration": 2.6958301067352295
    },
    {
      "auxiliary_loss_clip": 0.01184141,
      "auxiliary_loss_mlp": 0.0102419,
      "balance_loss_clip": 1.0518024,
      "balance_loss_mlp": 1.01609254,
      "epoch": 0.5658630433475621,
      "flos": 21944041562880.0,
      "grad_norm": 1.7528722096283198,
      "language_loss": 0.75604194,
      "learning_rate": 1.6719836184427275e-06,
      "loss": 0.77812523,
      "num_input_tokens_seen": 101524635,
      "step": 4706,
      "time_per_iteration": 2.5875251293182373
    },
    {
      "auxiliary_loss_clip": 0.01274911,
      "auxiliary_loss_mlp": 0.01020466,
      "balance_loss_clip": 1.04589415,
      "balance_loss_mlp": 1.01339364,
      "epoch": 0.5659832862382012,
      "flos": 30409012218240.0,
      "grad_norm": 1.7687027827728092,
      "language_loss": 0.64248878,
      "learning_rate": 1.671215219767733e-06,
      "loss": 0.66544259,
      "num_input_tokens_seen": 101544095,
      "step": 4707,
      "time_per_iteration": 2.905208110809326
    },
    {
      "auxiliary_loss_clip": 0.01384729,
      "auxiliary_loss_mlp": 0.01029098,
      "balance_loss_clip": 1.04485345,
      "balance_loss_mlp": 1.02125454,
      "epoch": 0.5661035291288402,
      "flos": 13188194570880.0,
      "grad_norm": 1.9350664604565126,
      "language_loss": 0.76118815,
      "learning_rate": 1.670446870969127e-06,
      "loss": 0.78532642,
      "num_input_tokens_seen": 101561760,
      "step": 4708,
      "time_per_iteration": 2.783738374710083
    },
    {
      "auxiliary_loss_clip": 0.01187144,
      "auxiliary_loss_mlp": 0.01024037,
      "balance_loss_clip": 1.05129743,
      "balance_loss_mlp": 1.01580548,
      "epoch": 0.5662237720194794,
      "flos": 16143032108160.0,
      "grad_norm": 2.3527862832707496,
      "language_loss": 0.79792035,
      "learning_rate": 1.6696785721634685e-06,
      "loss": 0.82003218,
      "num_input_tokens_seen": 101576245,
      "step": 4709,
      "time_per_iteration": 2.6452348232269287
    },
    {
      "auxiliary_loss_clip": 0.01234386,
      "auxiliary_loss_mlp": 0.01025039,
      "balance_loss_clip": 1.05035305,
      "balance_loss_mlp": 1.0170908,
      "epoch": 0.5663440149101184,
      "flos": 17676848718720.0,
      "grad_norm": 1.9024627975701927,
      "language_loss": 0.73521495,
      "learning_rate": 1.6689103234673086e-06,
      "loss": 0.75780916,
      "num_input_tokens_seen": 101594565,
      "step": 4710,
      "time_per_iteration": 2.6219842433929443
    },
    {
      "auxiliary_loss_clip": 0.01276932,
      "auxiliary_loss_mlp": 0.01033042,
      "balance_loss_clip": 1.04852462,
      "balance_loss_mlp": 1.02520728,
      "epoch": 0.5664642578007575,
      "flos": 23368330627200.0,
      "grad_norm": 1.926486180993532,
      "language_loss": 0.77101666,
      "learning_rate": 1.668142124997189e-06,
      "loss": 0.79411638,
      "num_input_tokens_seen": 101614225,
      "step": 4711,
      "time_per_iteration": 2.6872847080230713
    },
    {
      "auxiliary_loss_clip": 0.01162531,
      "auxiliary_loss_mlp": 0.01012715,
      "balance_loss_clip": 1.00840223,
      "balance_loss_mlp": 1.01169622,
      "epoch": 0.5665845006913967,
      "flos": 65516470945920.0,
      "grad_norm": 0.7348412636497615,
      "language_loss": 0.59740937,
      "learning_rate": 1.6673739768696453e-06,
      "loss": 0.61916184,
      "num_input_tokens_seen": 101680795,
      "step": 4712,
      "time_per_iteration": 4.234415054321289
    },
    {
      "auxiliary_loss_clip": 0.01286881,
      "auxiliary_loss_mlp": 0.01025057,
      "balance_loss_clip": 1.04747808,
      "balance_loss_mlp": 1.01708817,
      "epoch": 0.5667047435820357,
      "flos": 26140885620480.0,
      "grad_norm": 1.672834526337638,
      "language_loss": 0.77430654,
      "learning_rate": 1.6666058792012052e-06,
      "loss": 0.79742599,
      "num_input_tokens_seen": 101701680,
      "step": 4713,
      "time_per_iteration": 2.7485244274139404
    },
    {
      "auxiliary_loss_clip": 0.0112479,
      "auxiliary_loss_mlp": 0.01003468,
      "balance_loss_clip": 1.01072264,
      "balance_loss_mlp": 1.00249064,
      "epoch": 0.5668249864726748,
      "flos": 71866949725440.0,
      "grad_norm": 0.8766652000769987,
      "language_loss": 0.68780035,
      "learning_rate": 1.6658378321083878e-06,
      "loss": 0.70908296,
      "num_input_tokens_seen": 101766010,
      "step": 4714,
      "time_per_iteration": 3.2573964595794678
    },
    {
      "auxiliary_loss_clip": 0.01333834,
      "auxiliary_loss_mlp": 0.01030111,
      "balance_loss_clip": 1.04602265,
      "balance_loss_mlp": 1.02259529,
      "epoch": 0.5669452293633139,
      "flos": 22195667312640.0,
      "grad_norm": 2.1279814622783078,
      "language_loss": 0.82416368,
      "learning_rate": 1.6650698357077055e-06,
      "loss": 0.84780312,
      "num_input_tokens_seen": 101783055,
      "step": 4715,
      "time_per_iteration": 3.5946412086486816
    },
    {
      "auxiliary_loss_clip": 0.01285041,
      "auxiliary_loss_mlp": 0.01030702,
      "balance_loss_clip": 1.04857123,
      "balance_loss_mlp": 1.02252448,
      "epoch": 0.567065472253953,
      "flos": 18223193560320.0,
      "grad_norm": 2.1348224791753005,
      "language_loss": 0.80991089,
      "learning_rate": 1.6643018901156632e-06,
      "loss": 0.83306831,
      "num_input_tokens_seen": 101802150,
      "step": 4716,
      "time_per_iteration": 2.6212716102600098
    },
    {
      "auxiliary_loss_clip": 0.012838,
      "auxiliary_loss_mlp": 0.01028561,
      "balance_loss_clip": 1.04792452,
      "balance_loss_mlp": 1.02116418,
      "epoch": 0.567185715144592,
      "flos": 20371548983040.0,
      "grad_norm": 3.6230044664303396,
      "language_loss": 0.79498303,
      "learning_rate": 1.6635339954487566e-06,
      "loss": 0.81810665,
      "num_input_tokens_seen": 101818025,
      "step": 4717,
      "time_per_iteration": 3.5175905227661133
    },
    {
      "auxiliary_loss_clip": 0.01285173,
      "auxiliary_loss_mlp": 0.01032906,
      "balance_loss_clip": 1.05040729,
      "balance_loss_mlp": 1.02507114,
      "epoch": 0.5673059580352312,
      "flos": 23221348174080.0,
      "grad_norm": 1.717969033922128,
      "language_loss": 0.82374787,
      "learning_rate": 1.6627661518234765e-06,
      "loss": 0.84692872,
      "num_input_tokens_seen": 101837280,
      "step": 4718,
      "time_per_iteration": 2.7227799892425537
    },
    {
      "auxiliary_loss_clip": 0.01388534,
      "auxiliary_loss_mlp": 0.01032872,
      "balance_loss_clip": 1.05158341,
      "balance_loss_mlp": 1.025141,
      "epoch": 0.5674262009258703,
      "flos": 21719599430400.0,
      "grad_norm": 2.183539477077222,
      "language_loss": 0.85408044,
      "learning_rate": 1.661998359356302e-06,
      "loss": 0.87829447,
      "num_input_tokens_seen": 101856310,
      "step": 4719,
      "time_per_iteration": 2.732398509979248
    },
    {
      "auxiliary_loss_clip": 0.01068073,
      "auxiliary_loss_mlp": 0.01001556,
      "balance_loss_clip": 1.01003444,
      "balance_loss_mlp": 1.00059021,
      "epoch": 0.5675464438165093,
      "flos": 67470369114240.0,
      "grad_norm": 0.7446859585483963,
      "language_loss": 0.55785942,
      "learning_rate": 1.6612306181637077e-06,
      "loss": 0.5785557,
      "num_input_tokens_seen": 101915635,
      "step": 4720,
      "time_per_iteration": 4.031984329223633
    },
    {
      "auxiliary_loss_clip": 0.0132986,
      "auxiliary_loss_mlp": 0.01022855,
      "balance_loss_clip": 1.04655886,
      "balance_loss_mlp": 1.01517773,
      "epoch": 0.5676666867071485,
      "flos": 18879173688960.0,
      "grad_norm": 2.768132014296347,
      "language_loss": 0.66014767,
      "learning_rate": 1.6604629283621598e-06,
      "loss": 0.68367481,
      "num_input_tokens_seen": 101933565,
      "step": 4721,
      "time_per_iteration": 2.7024054527282715
    },
    {
      "auxiliary_loss_clip": 0.01186506,
      "auxiliary_loss_mlp": 0.01026381,
      "balance_loss_clip": 1.0534029,
      "balance_loss_mlp": 1.01834655,
      "epoch": 0.5677869295977875,
      "flos": 33546778744320.0,
      "grad_norm": 3.019931851807437,
      "language_loss": 0.74390626,
      "learning_rate": 1.6596952900681152e-06,
      "loss": 0.7660352,
      "num_input_tokens_seen": 101954325,
      "step": 4722,
      "time_per_iteration": 2.714780569076538
    },
    {
      "auxiliary_loss_clip": 0.01382465,
      "auxiliary_loss_mlp": 0.01031604,
      "balance_loss_clip": 1.04989052,
      "balance_loss_mlp": 1.02372384,
      "epoch": 0.5679071724884266,
      "flos": 28037256157440.0,
      "grad_norm": 2.0566151128570125,
      "language_loss": 0.81963181,
      "learning_rate": 1.658927703398025e-06,
      "loss": 0.84377253,
      "num_input_tokens_seen": 101974390,
      "step": 4723,
      "time_per_iteration": 2.856614589691162
    },
    {
      "auxiliary_loss_clip": 0.01377219,
      "auxiliary_loss_mlp": 0.01029312,
      "balance_loss_clip": 1.03978407,
      "balance_loss_mlp": 1.02211165,
      "epoch": 0.5680274153790658,
      "flos": 23550110380800.0,
      "grad_norm": 2.68317589989917,
      "language_loss": 0.77754414,
      "learning_rate": 1.6581601684683309e-06,
      "loss": 0.80160952,
      "num_input_tokens_seen": 101994815,
      "step": 4724,
      "time_per_iteration": 2.733449697494507
    },
    {
      "auxiliary_loss_clip": 0.01232431,
      "auxiliary_loss_mlp": 0.01033766,
      "balance_loss_clip": 1.05279422,
      "balance_loss_mlp": 1.02630949,
      "epoch": 0.5681476582697048,
      "flos": 22455158140800.0,
      "grad_norm": 3.9345079217982946,
      "language_loss": 0.68586653,
      "learning_rate": 1.6573926853954674e-06,
      "loss": 0.70852852,
      "num_input_tokens_seen": 102012400,
      "step": 4725,
      "time_per_iteration": 2.6822268962860107
    },
    {
      "auxiliary_loss_clip": 0.01278691,
      "auxiliary_loss_mlp": 0.010332,
      "balance_loss_clip": 1.04723775,
      "balance_loss_mlp": 1.02595246,
      "epoch": 0.5682679011603439,
      "flos": 19536913584000.0,
      "grad_norm": 2.0827819691479648,
      "language_loss": 0.83000362,
      "learning_rate": 1.6566252542958608e-06,
      "loss": 0.85312259,
      "num_input_tokens_seen": 102031900,
      "step": 4726,
      "time_per_iteration": 2.6618611812591553
    },
    {
      "auxiliary_loss_clip": 0.01320812,
      "auxiliary_loss_mlp": 0.01030475,
      "balance_loss_clip": 1.04624081,
      "balance_loss_mlp": 1.02263653,
      "epoch": 0.568388144050983,
      "flos": 28765488493440.0,
      "grad_norm": 1.8127639163180953,
      "language_loss": 0.7834909,
      "learning_rate": 1.6558578752859305e-06,
      "loss": 0.80700374,
      "num_input_tokens_seen": 102050860,
      "step": 4727,
      "time_per_iteration": 2.8196334838867188
    },
    {
      "auxiliary_loss_clip": 0.01332848,
      "auxiliary_loss_mlp": 0.01027296,
      "balance_loss_clip": 1.04595995,
      "balance_loss_mlp": 1.01962519,
      "epoch": 0.5685083869416221,
      "flos": 21209452519680.0,
      "grad_norm": 1.8878860813702019,
      "language_loss": 0.78787422,
      "learning_rate": 1.6550905484820865e-06,
      "loss": 0.81147569,
      "num_input_tokens_seen": 102069320,
      "step": 4728,
      "time_per_iteration": 2.7042295932769775
    },
    {
      "auxiliary_loss_clip": 0.0118455,
      "auxiliary_loss_mlp": 0.0102599,
      "balance_loss_clip": 1.05194616,
      "balance_loss_mlp": 1.01832163,
      "epoch": 0.5686286298322611,
      "flos": 24827021942400.0,
      "grad_norm": 2.3858696096509178,
      "language_loss": 0.78584313,
      "learning_rate": 1.6543232740007328e-06,
      "loss": 0.80794853,
      "num_input_tokens_seen": 102086435,
      "step": 4729,
      "time_per_iteration": 2.660003900527954
    },
    {
      "auxiliary_loss_clip": 0.01236853,
      "auxiliary_loss_mlp": 0.01032421,
      "balance_loss_clip": 1.05313361,
      "balance_loss_mlp": 1.02348614,
      "epoch": 0.5687488727229003,
      "flos": 26615121909120.0,
      "grad_norm": 2.594383047312774,
      "language_loss": 0.66641808,
      "learning_rate": 1.653556051958263e-06,
      "loss": 0.68911088,
      "num_input_tokens_seen": 102106115,
      "step": 4730,
      "time_per_iteration": 2.6903789043426514
    },
    {
      "auxiliary_loss_clip": 0.01416586,
      "auxiliary_loss_mlp": 0.0102725,
      "balance_loss_clip": 1.04191947,
      "balance_loss_mlp": 1.01988339,
      "epoch": 0.5688691156135394,
      "flos": 20808725414400.0,
      "grad_norm": 1.8614552345688236,
      "language_loss": 0.73947752,
      "learning_rate": 1.6527888824710642e-06,
      "loss": 0.7639159,
      "num_input_tokens_seen": 102125715,
      "step": 4731,
      "time_per_iteration": 2.760223150253296
    },
    {
      "auxiliary_loss_clip": 0.01325365,
      "auxiliary_loss_mlp": 0.01032294,
      "balance_loss_clip": 1.0439502,
      "balance_loss_mlp": 1.02396166,
      "epoch": 0.5689893585041784,
      "flos": 25880963829120.0,
      "grad_norm": 2.3402966168695256,
      "language_loss": 0.76499438,
      "learning_rate": 1.6520217656555166e-06,
      "loss": 0.788571,
      "num_input_tokens_seen": 102145005,
      "step": 4732,
      "time_per_iteration": 2.8031606674194336
    },
    {
      "auxiliary_loss_clip": 0.01275265,
      "auxiliary_loss_mlp": 0.01029398,
      "balance_loss_clip": 1.04711843,
      "balance_loss_mlp": 1.02266002,
      "epoch": 0.5691096013948175,
      "flos": 23477463123840.0,
      "grad_norm": 1.55225773506792,
      "language_loss": 0.70916206,
      "learning_rate": 1.65125470162799e-06,
      "loss": 0.73220867,
      "num_input_tokens_seen": 102165360,
      "step": 4733,
      "time_per_iteration": 2.7007946968078613
    },
    {
      "auxiliary_loss_clip": 0.01330993,
      "auxiliary_loss_mlp": 0.01024601,
      "balance_loss_clip": 1.0456574,
      "balance_loss_mlp": 1.01678729,
      "epoch": 0.5692298442854566,
      "flos": 18075600576000.0,
      "grad_norm": 2.7934266399625884,
      "language_loss": 0.69971073,
      "learning_rate": 1.6504876905048485e-06,
      "loss": 0.72326666,
      "num_input_tokens_seen": 102182320,
      "step": 4734,
      "time_per_iteration": 2.687138557434082
    },
    {
      "auxiliary_loss_clip": 0.01180043,
      "auxiliary_loss_mlp": 0.01029181,
      "balance_loss_clip": 1.05171227,
      "balance_loss_mlp": 1.02179003,
      "epoch": 0.5693500871760957,
      "flos": 23039317025280.0,
      "grad_norm": 1.8294155897288726,
      "language_loss": 0.72204638,
      "learning_rate": 1.6497207324024464e-06,
      "loss": 0.7441386,
      "num_input_tokens_seen": 102201220,
      "step": 4735,
      "time_per_iteration": 2.572396993637085
    },
    {
      "auxiliary_loss_clip": 0.01186302,
      "auxiliary_loss_mlp": 0.01025623,
      "balance_loss_clip": 1.04574955,
      "balance_loss_mlp": 1.01756406,
      "epoch": 0.5694703300667348,
      "flos": 18989670902400.0,
      "grad_norm": 2.2622934219791118,
      "language_loss": 0.82757545,
      "learning_rate": 1.6489538274371305e-06,
      "loss": 0.84969473,
      "num_input_tokens_seen": 102219825,
      "step": 4736,
      "time_per_iteration": 2.6892971992492676
    },
    {
      "auxiliary_loss_clip": 0.01227225,
      "auxiliary_loss_mlp": 0.01027096,
      "balance_loss_clip": 1.05156422,
      "balance_loss_mlp": 1.01991415,
      "epoch": 0.5695905729573739,
      "flos": 21908705558400.0,
      "grad_norm": 1.889202669486198,
      "language_loss": 0.83407092,
      "learning_rate": 1.6481869757252396e-06,
      "loss": 0.85661423,
      "num_input_tokens_seen": 102238160,
      "step": 4737,
      "time_per_iteration": 2.609886884689331
    },
    {
      "auxiliary_loss_clip": 0.01231026,
      "auxiliary_loss_mlp": 0.01026462,
      "balance_loss_clip": 1.05156934,
      "balance_loss_mlp": 1.01947367,
      "epoch": 0.569710815848013,
      "flos": 28476659232000.0,
      "grad_norm": 1.6708278704792228,
      "language_loss": 0.71827161,
      "learning_rate": 1.647420177383105e-06,
      "loss": 0.74084651,
      "num_input_tokens_seen": 102261030,
      "step": 4738,
      "time_per_iteration": 3.7080414295196533
    },
    {
      "auxiliary_loss_clip": 0.01227952,
      "auxiliary_loss_mlp": 0.01027004,
      "balance_loss_clip": 1.05338717,
      "balance_loss_mlp": 1.01979816,
      "epoch": 0.569831058738652,
      "flos": 28366162018560.0,
      "grad_norm": 2.3871244355536305,
      "language_loss": 0.72529101,
      "learning_rate": 1.646653432527049e-06,
      "loss": 0.74784058,
      "num_input_tokens_seen": 102281670,
      "step": 4739,
      "time_per_iteration": 2.687326192855835
    },
    {
      "auxiliary_loss_clip": 0.01331907,
      "auxiliary_loss_mlp": 0.01026925,
      "balance_loss_clip": 1.04849565,
      "balance_loss_mlp": 1.01967144,
      "epoch": 0.5699513016292912,
      "flos": 25849973370240.0,
      "grad_norm": 1.6086021504869235,
      "language_loss": 0.746391,
      "learning_rate": 1.645886741273387e-06,
      "loss": 0.76997936,
      "num_input_tokens_seen": 102303485,
      "step": 4740,
      "time_per_iteration": 2.7366433143615723
    },
    {
      "auxiliary_loss_clip": 0.01334321,
      "auxiliary_loss_mlp": 0.01027602,
      "balance_loss_clip": 1.05642986,
      "balance_loss_mlp": 1.01953757,
      "epoch": 0.5700715445199303,
      "flos": 18037858360320.0,
      "grad_norm": 2.885650611618166,
      "language_loss": 0.74122882,
      "learning_rate": 1.645120103738424e-06,
      "loss": 0.76484805,
      "num_input_tokens_seen": 102320995,
      "step": 4741,
      "time_per_iteration": 3.5803277492523193
    },
    {
      "auxiliary_loss_clip": 0.01223904,
      "auxiliary_loss_mlp": 0.02564896,
      "balance_loss_clip": 1.04794443,
      "balance_loss_mlp": 1.00008357,
      "epoch": 0.5701917874105693,
      "flos": 11473352392320.0,
      "grad_norm": 2.285784452093703,
      "language_loss": 0.83795005,
      "learning_rate": 1.6443535200384591e-06,
      "loss": 0.87583804,
      "num_input_tokens_seen": 102339170,
      "step": 4742,
      "time_per_iteration": 3.51320219039917
    },
    {
      "auxiliary_loss_clip": 0.0118073,
      "auxiliary_loss_mlp": 0.01025841,
      "balance_loss_clip": 1.05222857,
      "balance_loss_mlp": 1.01815212,
      "epoch": 0.5703120303012085,
      "flos": 21761759018880.0,
      "grad_norm": 1.646453015490757,
      "language_loss": 0.70799458,
      "learning_rate": 1.6435869902897827e-06,
      "loss": 0.73006028,
      "num_input_tokens_seen": 102357750,
      "step": 4743,
      "time_per_iteration": 2.598745346069336
    },
    {
      "auxiliary_loss_clip": 0.01225295,
      "auxiliary_loss_mlp": 0.01002685,
      "balance_loss_clip": 1.00845027,
      "balance_loss_mlp": 1.00164819,
      "epoch": 0.5704322731918475,
      "flos": 56746258513920.0,
      "grad_norm": 0.7946960506215589,
      "language_loss": 0.61964142,
      "learning_rate": 1.6428205146086764e-06,
      "loss": 0.64192122,
      "num_input_tokens_seen": 102419730,
      "step": 4744,
      "time_per_iteration": 3.291846752166748
    },
    {
      "auxiliary_loss_clip": 0.01286415,
      "auxiliary_loss_mlp": 0.01029276,
      "balance_loss_clip": 1.04816854,
      "balance_loss_mlp": 1.02185535,
      "epoch": 0.5705525160824866,
      "flos": 20741141975040.0,
      "grad_norm": 1.6863916626238111,
      "language_loss": 0.71478951,
      "learning_rate": 1.6420540931114142e-06,
      "loss": 0.73794651,
      "num_input_tokens_seen": 102440320,
      "step": 4745,
      "time_per_iteration": 2.6983258724212646
    },
    {
      "auxiliary_loss_clip": 0.0128054,
      "auxiliary_loss_mlp": 0.01031758,
      "balance_loss_clip": 1.04906225,
      "balance_loss_mlp": 1.02464151,
      "epoch": 0.5706727589731257,
      "flos": 18771262254720.0,
      "grad_norm": 1.7118497089299085,
      "language_loss": 0.79157442,
      "learning_rate": 1.6412877259142616e-06,
      "loss": 0.81469744,
      "num_input_tokens_seen": 102460240,
      "step": 4746,
      "time_per_iteration": 3.568301200866699
    },
    {
      "auxiliary_loss_clip": 0.01280078,
      "auxiliary_loss_mlp": 0.01027889,
      "balance_loss_clip": 1.04833984,
      "balance_loss_mlp": 1.02041423,
      "epoch": 0.5707930018637648,
      "flos": 27634733372160.0,
      "grad_norm": 2.0877140401518823,
      "language_loss": 0.74110162,
      "learning_rate": 1.6405214131334757e-06,
      "loss": 0.76418126,
      "num_input_tokens_seen": 102478765,
      "step": 4747,
      "time_per_iteration": 2.72139573097229
    },
    {
      "auxiliary_loss_clip": 0.01371706,
      "auxiliary_loss_mlp": 0.01027555,
      "balance_loss_clip": 1.04703116,
      "balance_loss_mlp": 1.02016723,
      "epoch": 0.5709132447544039,
      "flos": 27597673514880.0,
      "grad_norm": 2.228295198985117,
      "language_loss": 0.79798913,
      "learning_rate": 1.6397551548853052e-06,
      "loss": 0.82198179,
      "num_input_tokens_seen": 102496930,
      "step": 4748,
      "time_per_iteration": 2.7810299396514893
    },
    {
      "auxiliary_loss_clip": 0.0127982,
      "auxiliary_loss_mlp": 0.01024947,
      "balance_loss_clip": 1.04910517,
      "balance_loss_mlp": 1.01756191,
      "epoch": 0.571033487645043,
      "flos": 21686095019520.0,
      "grad_norm": 2.023012243093012,
      "language_loss": 0.70833284,
      "learning_rate": 1.6389889512859917e-06,
      "loss": 0.73138052,
      "num_input_tokens_seen": 102516590,
      "step": 4749,
      "time_per_iteration": 2.7019872665405273
    },
    {
      "auxiliary_loss_clip": 0.01171612,
      "auxiliary_loss_mlp": 0.01000871,
      "balance_loss_clip": 1.00864136,
      "balance_loss_mlp": 0.99988121,
      "epoch": 0.5711537305356821,
      "flos": 70181445980160.0,
      "grad_norm": 0.8152350481735203,
      "language_loss": 0.60373354,
      "learning_rate": 1.638222802451767e-06,
      "loss": 0.62545836,
      "num_input_tokens_seen": 102578070,
      "step": 4750,
      "time_per_iteration": 3.254171848297119
    },
    {
      "auxiliary_loss_clip": 0.01226207,
      "auxiliary_loss_mlp": 0.01024779,
      "balance_loss_clip": 1.05091751,
      "balance_loss_mlp": 1.0176568,
      "epoch": 0.5712739734263211,
      "flos": 24717494396160.0,
      "grad_norm": 1.7825903793577667,
      "language_loss": 0.75307274,
      "learning_rate": 1.6374567084988561e-06,
      "loss": 0.77558255,
      "num_input_tokens_seen": 102599255,
      "step": 4751,
      "time_per_iteration": 2.7250876426696777
    },
    {
      "auxiliary_loss_clip": 0.01184452,
      "auxiliary_loss_mlp": 0.01029798,
      "balance_loss_clip": 1.05096173,
      "balance_loss_mlp": 1.02220738,
      "epoch": 0.5713942163169603,
      "flos": 26578169792640.0,
      "grad_norm": 1.8946219068546517,
      "language_loss": 0.76580131,
      "learning_rate": 1.6366906695434738e-06,
      "loss": 0.78794384,
      "num_input_tokens_seen": 102621775,
      "step": 4752,
      "time_per_iteration": 2.683281898498535
    },
    {
      "auxiliary_loss_clip": 0.01234149,
      "auxiliary_loss_mlp": 0.01026632,
      "balance_loss_clip": 1.05283856,
      "balance_loss_mlp": 1.0193727,
      "epoch": 0.5715144592075994,
      "flos": 21142443697920.0,
      "grad_norm": 2.3435150546467614,
      "language_loss": 0.86230248,
      "learning_rate": 1.6359246857018275e-06,
      "loss": 0.88491035,
      "num_input_tokens_seen": 102639305,
      "step": 4753,
      "time_per_iteration": 2.629426956176758
    },
    {
      "auxiliary_loss_clip": 0.01279442,
      "auxiliary_loss_mlp": 0.01030749,
      "balance_loss_clip": 1.04321933,
      "balance_loss_mlp": 1.02302456,
      "epoch": 0.5716347020982384,
      "flos": 23330265189120.0,
      "grad_norm": 1.8868508085526876,
      "language_loss": 0.78260285,
      "learning_rate": 1.6351587570901178e-06,
      "loss": 0.80570471,
      "num_input_tokens_seen": 102659430,
      "step": 4754,
      "time_per_iteration": 2.7108097076416016
    },
    {
      "auxiliary_loss_clip": 0.01336294,
      "auxiliary_loss_mlp": 0.01027186,
      "balance_loss_clip": 1.05178928,
      "balance_loss_mlp": 1.01980114,
      "epoch": 0.5717549449888776,
      "flos": 17009555806080.0,
      "grad_norm": 2.4075153750023595,
      "language_loss": 0.75876999,
      "learning_rate": 1.634392883824534e-06,
      "loss": 0.78240484,
      "num_input_tokens_seen": 102671430,
      "step": 4755,
      "time_per_iteration": 2.6844069957733154
    },
    {
      "auxiliary_loss_clip": 0.0128181,
      "auxiliary_loss_mlp": 0.01033232,
      "balance_loss_clip": 1.04571664,
      "balance_loss_mlp": 1.02553749,
      "epoch": 0.5718751878795166,
      "flos": 35518130922240.0,
      "grad_norm": 1.8128056282474951,
      "language_loss": 0.67995918,
      "learning_rate": 1.6336270660212595e-06,
      "loss": 0.70310962,
      "num_input_tokens_seen": 102693025,
      "step": 4756,
      "time_per_iteration": 2.8281702995300293
    },
    {
      "auxiliary_loss_clip": 0.01288166,
      "auxiliary_loss_mlp": 0.01028963,
      "balance_loss_clip": 1.05744934,
      "balance_loss_mlp": 1.0213809,
      "epoch": 0.5719954307701557,
      "flos": 38613989255040.0,
      "grad_norm": 2.196331051810071,
      "language_loss": 0.6605503,
      "learning_rate": 1.6328613037964676e-06,
      "loss": 0.68372154,
      "num_input_tokens_seen": 102716090,
      "step": 4757,
      "time_per_iteration": 2.8131961822509766
    },
    {
      "auxiliary_loss_clip": 0.01229302,
      "auxiliary_loss_mlp": 0.01026797,
      "balance_loss_clip": 1.04924059,
      "balance_loss_mlp": 1.01914954,
      "epoch": 0.5721156736607949,
      "flos": 20631111638400.0,
      "grad_norm": 4.9294899827133785,
      "language_loss": 0.68539453,
      "learning_rate": 1.6320955972663241e-06,
      "loss": 0.70795548,
      "num_input_tokens_seen": 102735685,
      "step": 4758,
      "time_per_iteration": 2.6700899600982666
    },
    {
      "auxiliary_loss_clip": 0.0122822,
      "auxiliary_loss_mlp": 0.01024415,
      "balance_loss_clip": 1.04845619,
      "balance_loss_mlp": 1.01664805,
      "epoch": 0.5722359165514339,
      "flos": 37415076076800.0,
      "grad_norm": 1.7586470769756641,
      "language_loss": 0.65874839,
      "learning_rate": 1.6313299465469857e-06,
      "loss": 0.68127477,
      "num_input_tokens_seen": 102758415,
      "step": 4759,
      "time_per_iteration": 2.72322678565979
    },
    {
      "auxiliary_loss_clip": 0.01229917,
      "auxiliary_loss_mlp": 0.01026873,
      "balance_loss_clip": 1.050367,
      "balance_loss_mlp": 1.01898122,
      "epoch": 0.572356159442073,
      "flos": 21972877205760.0,
      "grad_norm": 2.7249803208593937,
      "language_loss": 0.7940011,
      "learning_rate": 1.6305643517546014e-06,
      "loss": 0.81656897,
      "num_input_tokens_seen": 102773795,
      "step": 4760,
      "time_per_iteration": 2.6486854553222656
    },
    {
      "auxiliary_loss_clip": 0.01182381,
      "auxiliary_loss_mlp": 0.01029881,
      "balance_loss_clip": 1.0530386,
      "balance_loss_mlp": 1.02297878,
      "epoch": 0.5724764023327121,
      "flos": 19135540033920.0,
      "grad_norm": 2.714352537631004,
      "language_loss": 0.84753388,
      "learning_rate": 1.629798813005311e-06,
      "loss": 0.86965644,
      "num_input_tokens_seen": 102793515,
      "step": 4761,
      "time_per_iteration": 2.5895488262176514
    },
    {
      "auxiliary_loss_clip": 0.01377627,
      "auxiliary_loss_mlp": 0.01024256,
      "balance_loss_clip": 1.05005383,
      "balance_loss_mlp": 1.01721621,
      "epoch": 0.5725966452233512,
      "flos": 22819759142400.0,
      "grad_norm": 1.9783247664762593,
      "language_loss": 0.71254176,
      "learning_rate": 1.6290333304152473e-06,
      "loss": 0.73656058,
      "num_input_tokens_seen": 102813390,
      "step": 4762,
      "time_per_iteration": 2.7515203952789307
    },
    {
      "auxiliary_loss_clip": 0.01284858,
      "auxiliary_loss_mlp": 0.01026797,
      "balance_loss_clip": 1.05471921,
      "balance_loss_mlp": 1.01919389,
      "epoch": 0.5727168881139902,
      "flos": 41496610498560.0,
      "grad_norm": 2.0936765108429323,
      "language_loss": 0.57054651,
      "learning_rate": 1.6282679041005314e-06,
      "loss": 0.59366304,
      "num_input_tokens_seen": 102838980,
      "step": 4763,
      "time_per_iteration": 2.8122787475585938
    },
    {
      "auxiliary_loss_clip": 0.01271858,
      "auxiliary_loss_mlp": 0.01025876,
      "balance_loss_clip": 1.04616606,
      "balance_loss_mlp": 1.0183332,
      "epoch": 0.5728371310046293,
      "flos": 14647675985280.0,
      "grad_norm": 2.405139475555805,
      "language_loss": 0.87432778,
      "learning_rate": 1.6275025341772789e-06,
      "loss": 0.89730519,
      "num_input_tokens_seen": 102855285,
      "step": 4764,
      "time_per_iteration": 3.5205283164978027
    },
    {
      "auxiliary_loss_clip": 0.01281936,
      "auxiliary_loss_mlp": 0.01030425,
      "balance_loss_clip": 1.04799461,
      "balance_loss_mlp": 1.02242064,
      "epoch": 0.5729573738952685,
      "flos": 21506613736320.0,
      "grad_norm": 2.3118542527215062,
      "language_loss": 0.8187874,
      "learning_rate": 1.626737220761596e-06,
      "loss": 0.84191096,
      "num_input_tokens_seen": 102872750,
      "step": 4765,
      "time_per_iteration": 2.713064193725586
    },
    {
      "auxiliary_loss_clip": 0.01231227,
      "auxiliary_loss_mlp": 0.01031991,
      "balance_loss_clip": 1.0535028,
      "balance_loss_mlp": 1.02451015,
      "epoch": 0.5730776167859075,
      "flos": 23621680229760.0,
      "grad_norm": 5.0461902993065575,
      "language_loss": 0.7863825,
      "learning_rate": 1.62597196396958e-06,
      "loss": 0.80901468,
      "num_input_tokens_seen": 102890920,
      "step": 4766,
      "time_per_iteration": 3.5387163162231445
    },
    {
      "auxiliary_loss_clip": 0.01230889,
      "auxiliary_loss_mlp": 0.01032208,
      "balance_loss_clip": 1.05162191,
      "balance_loss_mlp": 1.02459645,
      "epoch": 0.5731978596765466,
      "flos": 25739224761600.0,
      "grad_norm": 1.6976408354357884,
      "language_loss": 0.85474682,
      "learning_rate": 1.6252067639173197e-06,
      "loss": 0.87737775,
      "num_input_tokens_seen": 102912830,
      "step": 4767,
      "time_per_iteration": 2.7085955142974854
    },
    {
      "auxiliary_loss_clip": 0.01230759,
      "auxiliary_loss_mlp": 0.0102477,
      "balance_loss_clip": 1.04969323,
      "balance_loss_mlp": 1.0170424,
      "epoch": 0.5733181025671857,
      "flos": 26359509749760.0,
      "grad_norm": 2.027305000684339,
      "language_loss": 0.69777262,
      "learning_rate": 1.6244416207208956e-06,
      "loss": 0.72032791,
      "num_input_tokens_seen": 102933765,
      "step": 4768,
      "time_per_iteration": 3.640482187271118
    },
    {
      "auxiliary_loss_clip": 0.01328235,
      "auxiliary_loss_mlp": 0.01028491,
      "balance_loss_clip": 1.04790068,
      "balance_loss_mlp": 1.02120125,
      "epoch": 0.5734383454578248,
      "flos": 29423874833280.0,
      "grad_norm": 1.9391388821562516,
      "language_loss": 0.73919892,
      "learning_rate": 1.6236765344963787e-06,
      "loss": 0.76276618,
      "num_input_tokens_seen": 102955025,
      "step": 4769,
      "time_per_iteration": 2.749765396118164
    },
    {
      "auxiliary_loss_clip": 0.01278855,
      "auxiliary_loss_mlp": 0.01028013,
      "balance_loss_clip": 1.05117011,
      "balance_loss_mlp": 1.02021098,
      "epoch": 0.5735585883484638,
      "flos": 34969954487040.0,
      "grad_norm": 33.005400086409196,
      "language_loss": 0.69000524,
      "learning_rate": 1.6229115053598322e-06,
      "loss": 0.71307391,
      "num_input_tokens_seen": 102976780,
      "step": 4770,
      "time_per_iteration": 2.802699089050293
    },
    {
      "auxiliary_loss_clip": 0.01232643,
      "auxiliary_loss_mlp": 0.01028676,
      "balance_loss_clip": 1.05331159,
      "balance_loss_mlp": 1.02165139,
      "epoch": 0.573678831239103,
      "flos": 18770759464320.0,
      "grad_norm": 1.9341655720662432,
      "language_loss": 0.72266603,
      "learning_rate": 1.6221465334273108e-06,
      "loss": 0.74527919,
      "num_input_tokens_seen": 102995990,
      "step": 4771,
      "time_per_iteration": 3.4763355255126953
    },
    {
      "auxiliary_loss_clip": 0.01331984,
      "auxiliary_loss_mlp": 0.01027262,
      "balance_loss_clip": 1.04738533,
      "balance_loss_mlp": 1.01959062,
      "epoch": 0.5737990741297421,
      "flos": 25702883176320.0,
      "grad_norm": 1.9693663351122328,
      "language_loss": 0.61632776,
      "learning_rate": 1.6213816188148593e-06,
      "loss": 0.63992023,
      "num_input_tokens_seen": 103014695,
      "step": 4772,
      "time_per_iteration": 2.757981777191162
    },
    {
      "auxiliary_loss_clip": 0.01284327,
      "auxiliary_loss_mlp": 0.01026777,
      "balance_loss_clip": 1.05461812,
      "balance_loss_mlp": 1.01948786,
      "epoch": 0.5739193170203811,
      "flos": 27269234530560.0,
      "grad_norm": 1.8582079800691271,
      "language_loss": 0.77346468,
      "learning_rate": 1.6206167616385162e-06,
      "loss": 0.79657567,
      "num_input_tokens_seen": 103035760,
      "step": 4773,
      "time_per_iteration": 2.7102389335632324
    },
    {
      "auxiliary_loss_clip": 0.01284967,
      "auxiliary_loss_mlp": 0.01031974,
      "balance_loss_clip": 1.05087304,
      "balance_loss_mlp": 1.02394843,
      "epoch": 0.5740395599110203,
      "flos": 12239721993600.0,
      "grad_norm": 2.620752325614196,
      "language_loss": 0.73480672,
      "learning_rate": 1.6198519620143078e-06,
      "loss": 0.75797611,
      "num_input_tokens_seen": 103052915,
      "step": 4774,
      "time_per_iteration": 2.6490001678466797
    },
    {
      "auxiliary_loss_clip": 0.01336293,
      "auxiliary_loss_mlp": 0.01025111,
      "balance_loss_clip": 1.05148721,
      "balance_loss_mlp": 1.0177201,
      "epoch": 0.5741598028016593,
      "flos": 25921399564800.0,
      "grad_norm": 1.7585444962970336,
      "language_loss": 0.78129947,
      "learning_rate": 1.6190872200582546e-06,
      "loss": 0.80491352,
      "num_input_tokens_seen": 103074655,
      "step": 4775,
      "time_per_iteration": 2.7527284622192383
    },
    {
      "auxiliary_loss_clip": 0.01278203,
      "auxiliary_loss_mlp": 0.02568472,
      "balance_loss_clip": 1.0477773,
      "balance_loss_mlp": 1.00005782,
      "epoch": 0.5742800456922984,
      "flos": 19244133826560.0,
      "grad_norm": 2.441121299611465,
      "language_loss": 0.78540587,
      "learning_rate": 1.6183225358863676e-06,
      "loss": 0.82387257,
      "num_input_tokens_seen": 103091550,
      "step": 4776,
      "time_per_iteration": 2.6765708923339844
    },
    {
      "auxiliary_loss_clip": 0.01275132,
      "auxiliary_loss_mlp": 0.01030507,
      "balance_loss_clip": 1.04663718,
      "balance_loss_mlp": 1.0230031,
      "epoch": 0.5744002885829376,
      "flos": 30920487932160.0,
      "grad_norm": 2.206865837308673,
      "language_loss": 0.71679688,
      "learning_rate": 1.617557909614648e-06,
      "loss": 0.73985326,
      "num_input_tokens_seen": 103110985,
      "step": 4777,
      "time_per_iteration": 2.8889877796173096
    },
    {
      "auxiliary_loss_clip": 0.01324471,
      "auxiliary_loss_mlp": 0.01032354,
      "balance_loss_clip": 1.04623103,
      "balance_loss_mlp": 1.02477789,
      "epoch": 0.5745205314735766,
      "flos": 23840017050240.0,
      "grad_norm": 3.5343510766312516,
      "language_loss": 0.8627497,
      "learning_rate": 1.6167933413590899e-06,
      "loss": 0.88631797,
      "num_input_tokens_seen": 103129890,
      "step": 4778,
      "time_per_iteration": 2.671372652053833
    },
    {
      "auxiliary_loss_clip": 0.01229526,
      "auxiliary_loss_mlp": 0.01026896,
      "balance_loss_clip": 1.04926944,
      "balance_loss_mlp": 1.01982737,
      "epoch": 0.5746407743642157,
      "flos": 12311902373760.0,
      "grad_norm": 2.344647590696276,
      "language_loss": 0.91036248,
      "learning_rate": 1.6160288312356773e-06,
      "loss": 0.93292665,
      "num_input_tokens_seen": 103147020,
      "step": 4779,
      "time_per_iteration": 2.685467004776001
    },
    {
      "auxiliary_loss_clip": 0.01236754,
      "auxiliary_loss_mlp": 0.01027177,
      "balance_loss_clip": 1.05105448,
      "balance_loss_mlp": 1.01973271,
      "epoch": 0.5747610172548548,
      "flos": 24133658734080.0,
      "grad_norm": 2.0485407968829588,
      "language_loss": 0.81603503,
      "learning_rate": 1.6152643793603857e-06,
      "loss": 0.83867443,
      "num_input_tokens_seen": 103167370,
      "step": 4780,
      "time_per_iteration": 2.6075875759124756
    },
    {
      "auxiliary_loss_clip": 0.01183151,
      "auxiliary_loss_mlp": 0.01031297,
      "balance_loss_clip": 1.05378914,
      "balance_loss_mlp": 1.02336311,
      "epoch": 0.5748812601454939,
      "flos": 25408451393280.0,
      "grad_norm": 1.6955943475555175,
      "language_loss": 0.87719882,
      "learning_rate": 1.6144999858491815e-06,
      "loss": 0.89934325,
      "num_input_tokens_seen": 103186000,
      "step": 4781,
      "time_per_iteration": 2.6581413745880127
    },
    {
      "auxiliary_loss_clip": 0.01280963,
      "auxiliary_loss_mlp": 0.01029804,
      "balance_loss_clip": 1.04727006,
      "balance_loss_mlp": 1.02171516,
      "epoch": 0.575001503036133,
      "flos": 30624942827520.0,
      "grad_norm": 1.8596238483620298,
      "language_loss": 0.85902679,
      "learning_rate": 1.6137356508180232e-06,
      "loss": 0.8821345,
      "num_input_tokens_seen": 103207710,
      "step": 4782,
      "time_per_iteration": 2.717061758041382
    },
    {
      "auxiliary_loss_clip": 0.01181511,
      "auxiliary_loss_mlp": 0.02569651,
      "balance_loss_clip": 1.05044818,
      "balance_loss_mlp": 1.00008798,
      "epoch": 0.5751217459267721,
      "flos": 21726566668800.0,
      "grad_norm": 1.7015866024400699,
      "language_loss": 0.81449699,
      "learning_rate": 1.6129713743828593e-06,
      "loss": 0.85200858,
      "num_input_tokens_seen": 103226720,
      "step": 4783,
      "time_per_iteration": 2.6363980770111084
    },
    {
      "auxiliary_loss_clip": 0.0128056,
      "auxiliary_loss_mlp": 0.01029488,
      "balance_loss_clip": 1.04612374,
      "balance_loss_mlp": 1.02203405,
      "epoch": 0.5752419888174112,
      "flos": 21651620941440.0,
      "grad_norm": 1.4084662880832812,
      "language_loss": 0.75483149,
      "learning_rate": 1.6122071566596306e-06,
      "loss": 0.77793199,
      "num_input_tokens_seen": 103246995,
      "step": 4784,
      "time_per_iteration": 2.6607863903045654
    },
    {
      "auxiliary_loss_clip": 0.01231764,
      "auxiliary_loss_mlp": 0.01032237,
      "balance_loss_clip": 1.0511601,
      "balance_loss_mlp": 1.02486384,
      "epoch": 0.5753622317080502,
      "flos": 17775997234560.0,
      "grad_norm": 2.58845500733219,
      "language_loss": 0.83317429,
      "learning_rate": 1.6114429977642674e-06,
      "loss": 0.85581434,
      "num_input_tokens_seen": 103261500,
      "step": 4785,
      "time_per_iteration": 2.611516237258911
    },
    {
      "auxiliary_loss_clip": 0.01233879,
      "auxiliary_loss_mlp": 0.01028584,
      "balance_loss_clip": 1.05386901,
      "balance_loss_mlp": 1.02132988,
      "epoch": 0.5754824745986894,
      "flos": 19789616741760.0,
      "grad_norm": 1.8373139357813981,
      "language_loss": 0.73555815,
      "learning_rate": 1.6106788978126926e-06,
      "loss": 0.75818276,
      "num_input_tokens_seen": 103280475,
      "step": 4786,
      "time_per_iteration": 2.598598003387451
    },
    {
      "auxiliary_loss_clip": 0.01372212,
      "auxiliary_loss_mlp": 0.01025848,
      "balance_loss_clip": 1.04151642,
      "balance_loss_mlp": 1.0182482,
      "epoch": 0.5756027174893285,
      "flos": 30985665160320.0,
      "grad_norm": 2.2664482877017083,
      "language_loss": 0.78696305,
      "learning_rate": 1.6099148569208196e-06,
      "loss": 0.81094366,
      "num_input_tokens_seen": 103297695,
      "step": 4787,
      "time_per_iteration": 2.797889232635498
    },
    {
      "auxiliary_loss_clip": 0.01288813,
      "auxiliary_loss_mlp": 0.01030765,
      "balance_loss_clip": 1.05602598,
      "balance_loss_mlp": 1.02183616,
      "epoch": 0.5757229603799675,
      "flos": 28546864364160.0,
      "grad_norm": 1.7499104360936648,
      "language_loss": 0.63176894,
      "learning_rate": 1.6091508752045523e-06,
      "loss": 0.65496469,
      "num_input_tokens_seen": 103318575,
      "step": 4788,
      "time_per_iteration": 2.701235055923462
    },
    {
      "auxiliary_loss_clip": 0.01320579,
      "auxiliary_loss_mlp": 0.01026554,
      "balance_loss_clip": 1.04305041,
      "balance_loss_mlp": 1.01965809,
      "epoch": 0.5758432032706067,
      "flos": 22999024944000.0,
      "grad_norm": 2.2593761522529636,
      "language_loss": 0.86351216,
      "learning_rate": 1.608386952779787e-06,
      "loss": 0.88698351,
      "num_input_tokens_seen": 103337945,
      "step": 4789,
      "time_per_iteration": 2.718374490737915
    },
    {
      "auxiliary_loss_clip": 0.01284928,
      "auxiliary_loss_mlp": 0.01023089,
      "balance_loss_clip": 1.04852533,
      "balance_loss_mlp": 1.01582932,
      "epoch": 0.5759634461612457,
      "flos": 25739727552000.0,
      "grad_norm": 1.620829515571466,
      "language_loss": 0.74826694,
      "learning_rate": 1.6076230897624098e-06,
      "loss": 0.77134705,
      "num_input_tokens_seen": 103360150,
      "step": 4790,
      "time_per_iteration": 2.7147178649902344
    },
    {
      "auxiliary_loss_clip": 0.01230552,
      "auxiliary_loss_mlp": 0.01024564,
      "balance_loss_clip": 1.04741836,
      "balance_loss_mlp": 1.0167861,
      "epoch": 0.5760836890518848,
      "flos": 30591761639040.0,
      "grad_norm": 2.4910817551699167,
      "language_loss": 0.77762294,
      "learning_rate": 1.6068592862682974e-06,
      "loss": 0.80017406,
      "num_input_tokens_seen": 103378305,
      "step": 4791,
      "time_per_iteration": 3.648292064666748
    },
    {
      "auxiliary_loss_clip": 0.01282511,
      "auxiliary_loss_mlp": 0.0102936,
      "balance_loss_clip": 1.05112052,
      "balance_loss_mlp": 1.02185297,
      "epoch": 0.576203931942524,
      "flos": 36538963447680.0,
      "grad_norm": 1.9900896808974597,
      "language_loss": 0.73815036,
      "learning_rate": 1.6060955424133187e-06,
      "loss": 0.76126909,
      "num_input_tokens_seen": 103399230,
      "step": 4792,
      "time_per_iteration": 2.812748432159424
    },
    {
      "auxiliary_loss_clip": 0.01229674,
      "auxiliary_loss_mlp": 0.01031952,
      "balance_loss_clip": 1.05092263,
      "balance_loss_mlp": 1.02445376,
      "epoch": 0.576324174833163,
      "flos": 25516937445120.0,
      "grad_norm": 1.8217243133592864,
      "language_loss": 0.89712846,
      "learning_rate": 1.6053318583133332e-06,
      "loss": 0.91974473,
      "num_input_tokens_seen": 103420100,
      "step": 4793,
      "time_per_iteration": 3.5296506881713867
    },
    {
      "auxiliary_loss_clip": 0.01231167,
      "auxiliary_loss_mlp": 0.01031628,
      "balance_loss_clip": 1.050843,
      "balance_loss_mlp": 1.02371573,
      "epoch": 0.5764444177238021,
      "flos": 25119262995840.0,
      "grad_norm": 2.4490043230058647,
      "language_loss": 0.75203204,
      "learning_rate": 1.6045682340841907e-06,
      "loss": 0.77466005,
      "num_input_tokens_seen": 103439025,
      "step": 4794,
      "time_per_iteration": 3.5852842330932617
    },
    {
      "auxiliary_loss_clip": 0.01227798,
      "auxiliary_loss_mlp": 0.02508821,
      "balance_loss_clip": 1.01057696,
      "balance_loss_mlp": 1.00023866,
      "epoch": 0.5765646606144411,
      "flos": 62212687758720.0,
      "grad_norm": 0.7500367498041917,
      "language_loss": 0.57912683,
      "learning_rate": 1.6038046698417336e-06,
      "loss": 0.61649299,
      "num_input_tokens_seen": 103499920,
      "step": 4795,
      "time_per_iteration": 3.226593255996704
    },
    {
      "auxiliary_loss_clip": 0.01228305,
      "auxiliary_loss_mlp": 0.01031319,
      "balance_loss_clip": 1.04791689,
      "balance_loss_mlp": 1.02351701,
      "epoch": 0.5766849035050803,
      "flos": 25118760205440.0,
      "grad_norm": 2.435746811852171,
      "language_loss": 0.6910429,
      "learning_rate": 1.6030411657017919e-06,
      "loss": 0.71363902,
      "num_input_tokens_seen": 103519575,
      "step": 4796,
      "time_per_iteration": 2.6472907066345215
    },
    {
      "auxiliary_loss_clip": 0.0122648,
      "auxiliary_loss_mlp": 0.01035648,
      "balance_loss_clip": 1.04936624,
      "balance_loss_mlp": 1.02796197,
      "epoch": 0.5768051463957193,
      "flos": 15991093578240.0,
      "grad_norm": 1.799870099401235,
      "language_loss": 0.84533966,
      "learning_rate": 1.6022777217801903e-06,
      "loss": 0.86796099,
      "num_input_tokens_seen": 103536530,
      "step": 4797,
      "time_per_iteration": 2.6663472652435303
    },
    {
      "auxiliary_loss_clip": 0.01330156,
      "auxiliary_loss_mlp": 0.0102886,
      "balance_loss_clip": 1.05097127,
      "balance_loss_mlp": 1.02124274,
      "epoch": 0.5769253892863584,
      "flos": 22163635359360.0,
      "grad_norm": 1.9129731547245359,
      "language_loss": 0.74081618,
      "learning_rate": 1.601514338192742e-06,
      "loss": 0.76440632,
      "num_input_tokens_seen": 103556460,
      "step": 4798,
      "time_per_iteration": 3.6281416416168213
    },
    {
      "auxiliary_loss_clip": 0.01175211,
      "auxiliary_loss_mlp": 0.01027946,
      "balance_loss_clip": 1.04895139,
      "balance_loss_mlp": 1.02116561,
      "epoch": 0.5770456321769976,
      "flos": 22856388036480.0,
      "grad_norm": 2.363700419654705,
      "language_loss": 0.71050644,
      "learning_rate": 1.6007510150552514e-06,
      "loss": 0.73253804,
      "num_input_tokens_seen": 103574520,
      "step": 4799,
      "time_per_iteration": 2.6197187900543213
    },
    {
      "auxiliary_loss_clip": 0.01233523,
      "auxiliary_loss_mlp": 0.01031001,
      "balance_loss_clip": 1.04839468,
      "balance_loss_mlp": 1.02299631,
      "epoch": 0.5771658750676366,
      "flos": 46353672489600.0,
      "grad_norm": 1.6437784289167907,
      "language_loss": 0.62070101,
      "learning_rate": 1.599987752483515e-06,
      "loss": 0.64334619,
      "num_input_tokens_seen": 103598965,
      "step": 4800,
      "time_per_iteration": 2.8816757202148438
    },
    {
      "auxiliary_loss_clip": 0.01326414,
      "auxiliary_loss_mlp": 0.01026065,
      "balance_loss_clip": 1.04547834,
      "balance_loss_mlp": 1.01860797,
      "epoch": 0.5772861179582757,
      "flos": 22159972172160.0,
      "grad_norm": 1.5896672190292311,
      "language_loss": 0.68435818,
      "learning_rate": 1.5992245505933184e-06,
      "loss": 0.707883,
      "num_input_tokens_seen": 103618665,
      "step": 4801,
      "time_per_iteration": 2.686903715133667
    },
    {
      "auxiliary_loss_clip": 0.01184726,
      "auxiliary_loss_mlp": 0.01025504,
      "balance_loss_clip": 1.05364716,
      "balance_loss_mlp": 1.01855123,
      "epoch": 0.5774063608489148,
      "flos": 31248926916480.0,
      "grad_norm": 2.3316194907503975,
      "language_loss": 0.71480864,
      "learning_rate": 1.5984614095004388e-06,
      "loss": 0.736911,
      "num_input_tokens_seen": 103639800,
      "step": 4802,
      "time_per_iteration": 2.6644363403320312
    },
    {
      "auxiliary_loss_clip": 0.01231609,
      "auxiliary_loss_mlp": 0.01028491,
      "balance_loss_clip": 1.05129528,
      "balance_loss_mlp": 1.02118301,
      "epoch": 0.5775266037395539,
      "flos": 22527123039360.0,
      "grad_norm": 2.5689693837002783,
      "language_loss": 0.81068051,
      "learning_rate": 1.5976983293206438e-06,
      "loss": 0.83328152,
      "num_input_tokens_seen": 103655605,
      "step": 4803,
      "time_per_iteration": 2.633852481842041
    },
    {
      "auxiliary_loss_clip": 0.01278456,
      "auxiliary_loss_mlp": 0.01031252,
      "balance_loss_clip": 1.04483461,
      "balance_loss_mlp": 1.02309251,
      "epoch": 0.577646846630193,
      "flos": 21068790860160.0,
      "grad_norm": 2.182558673707391,
      "language_loss": 0.71141517,
      "learning_rate": 1.5969353101696928e-06,
      "loss": 0.73451221,
      "num_input_tokens_seen": 103674045,
      "step": 4804,
      "time_per_iteration": 2.6870031356811523
    },
    {
      "auxiliary_loss_clip": 0.0123228,
      "auxiliary_loss_mlp": 0.01028569,
      "balance_loss_clip": 1.05033112,
      "balance_loss_mlp": 1.0213654,
      "epoch": 0.5777670895208321,
      "flos": 29714284293120.0,
      "grad_norm": 1.6791778892028784,
      "language_loss": 0.80036741,
      "learning_rate": 1.5961723521633341e-06,
      "loss": 0.82297593,
      "num_input_tokens_seen": 103695285,
      "step": 4805,
      "time_per_iteration": 2.7238833904266357
    },
    {
      "auxiliary_loss_clip": 0.01179374,
      "auxiliary_loss_mlp": 0.01026359,
      "balance_loss_clip": 1.04644418,
      "balance_loss_mlp": 1.01917696,
      "epoch": 0.5778873324114712,
      "flos": 19500428344320.0,
      "grad_norm": 1.9900594277492383,
      "language_loss": 0.90959299,
      "learning_rate": 1.5954094554173097e-06,
      "loss": 0.93165028,
      "num_input_tokens_seen": 103713275,
      "step": 4806,
      "time_per_iteration": 2.6554222106933594
    },
    {
      "auxiliary_loss_clip": 0.01285622,
      "auxiliary_loss_mlp": 0.01025242,
      "balance_loss_clip": 1.05098796,
      "balance_loss_mlp": 1.01832175,
      "epoch": 0.5780075753021102,
      "flos": 14136846716160.0,
      "grad_norm": 2.321269763078272,
      "language_loss": 0.79485208,
      "learning_rate": 1.5946466200473482e-06,
      "loss": 0.81796074,
      "num_input_tokens_seen": 103731185,
      "step": 4807,
      "time_per_iteration": 2.7147319316864014
    },
    {
      "auxiliary_loss_clip": 0.01284718,
      "auxiliary_loss_mlp": 0.01025807,
      "balance_loss_clip": 1.04787457,
      "balance_loss_mlp": 1.01863074,
      "epoch": 0.5781278181927494,
      "flos": 15262178883840.0,
      "grad_norm": 1.731544401088325,
      "language_loss": 0.83327675,
      "learning_rate": 1.5938838461691723e-06,
      "loss": 0.85638201,
      "num_input_tokens_seen": 103748095,
      "step": 4808,
      "time_per_iteration": 2.6082839965820312
    },
    {
      "auxiliary_loss_clip": 0.0118352,
      "auxiliary_loss_mlp": 0.01028444,
      "balance_loss_clip": 1.05403674,
      "balance_loss_mlp": 1.02117765,
      "epoch": 0.5782480610833884,
      "flos": 16726831856640.0,
      "grad_norm": 2.234158267709471,
      "language_loss": 0.83296025,
      "learning_rate": 1.593121133898494e-06,
      "loss": 0.85507989,
      "num_input_tokens_seen": 103765300,
      "step": 4809,
      "time_per_iteration": 2.6428661346435547
    },
    {
      "auxiliary_loss_clip": 0.01238558,
      "auxiliary_loss_mlp": 0.0103027,
      "balance_loss_clip": 1.05293691,
      "balance_loss_mlp": 1.02301645,
      "epoch": 0.5783683039740275,
      "flos": 25482140144640.0,
      "grad_norm": 2.5637402119041974,
      "language_loss": 0.78982174,
      "learning_rate": 1.592358483351016e-06,
      "loss": 0.81251001,
      "num_input_tokens_seen": 103785475,
      "step": 4810,
      "time_per_iteration": 2.6304194927215576
    },
    {
      "auxiliary_loss_clip": 0.01226018,
      "auxiliary_loss_mlp": 0.01027218,
      "balance_loss_clip": 1.04992926,
      "balance_loss_mlp": 1.02000546,
      "epoch": 0.5784885468646667,
      "flos": 18405835240320.0,
      "grad_norm": 2.118972807050266,
      "language_loss": 0.72088492,
      "learning_rate": 1.5915958946424326e-06,
      "loss": 0.74341726,
      "num_input_tokens_seen": 103804160,
      "step": 4811,
      "time_per_iteration": 2.63069748878479
    },
    {
      "auxiliary_loss_clip": 0.01339621,
      "auxiliary_loss_mlp": 0.02570954,
      "balance_loss_clip": 1.05106974,
      "balance_loss_mlp": 1.00013638,
      "epoch": 0.5786087897553057,
      "flos": 46100717936640.0,
      "grad_norm": 5.3087604541277065,
      "language_loss": 0.74325836,
      "learning_rate": 1.5908333678884271e-06,
      "loss": 0.78236401,
      "num_input_tokens_seen": 103830580,
      "step": 4812,
      "time_per_iteration": 2.881743907928467
    },
    {
      "auxiliary_loss_clip": 0.01230544,
      "auxiliary_loss_mlp": 0.01027829,
      "balance_loss_clip": 1.05251563,
      "balance_loss_mlp": 1.02077138,
      "epoch": 0.5787290326459448,
      "flos": 12385950261120.0,
      "grad_norm": 1.9179577014403295,
      "language_loss": 0.73726285,
      "learning_rate": 1.5900709032046743e-06,
      "loss": 0.75984657,
      "num_input_tokens_seen": 103848655,
      "step": 4813,
      "time_per_iteration": 2.646627187728882
    },
    {
      "auxiliary_loss_clip": 0.01278729,
      "auxiliary_loss_mlp": 0.0102982,
      "balance_loss_clip": 1.05207324,
      "balance_loss_mlp": 1.02217317,
      "epoch": 0.5788492755365839,
      "flos": 23290332243840.0,
      "grad_norm": 1.9720035265371234,
      "language_loss": 0.78438139,
      "learning_rate": 1.5893085007068391e-06,
      "loss": 0.80746686,
      "num_input_tokens_seen": 103866215,
      "step": 4814,
      "time_per_iteration": 2.640532970428467
    },
    {
      "auxiliary_loss_clip": 0.01272979,
      "auxiliary_loss_mlp": 0.01030868,
      "balance_loss_clip": 1.04479694,
      "balance_loss_mlp": 1.02205825,
      "epoch": 0.578969518427223,
      "flos": 24061047390720.0,
      "grad_norm": 2.241207615653504,
      "language_loss": 0.71115792,
      "learning_rate": 1.5885461605105786e-06,
      "loss": 0.73419636,
      "num_input_tokens_seen": 103887815,
      "step": 4815,
      "time_per_iteration": 2.727259635925293
    },
    {
      "auxiliary_loss_clip": 0.01289972,
      "auxiliary_loss_mlp": 0.0102773,
      "balance_loss_clip": 1.05245268,
      "balance_loss_mlp": 1.01993394,
      "epoch": 0.579089761317862,
      "flos": 21871825269120.0,
      "grad_norm": 2.272715159166717,
      "language_loss": 0.77203906,
      "learning_rate": 1.5877838827315375e-06,
      "loss": 0.79521608,
      "num_input_tokens_seen": 103906360,
      "step": 4816,
      "time_per_iteration": 2.685189962387085
    },
    {
      "auxiliary_loss_clip": 0.01182644,
      "auxiliary_loss_mlp": 0.01033407,
      "balance_loss_clip": 1.05453122,
      "balance_loss_mlp": 1.02605486,
      "epoch": 0.5792100042085012,
      "flos": 22929681738240.0,
      "grad_norm": 2.795079008196703,
      "language_loss": 0.70458865,
      "learning_rate": 1.587021667485355e-06,
      "loss": 0.72674918,
      "num_input_tokens_seen": 103925730,
      "step": 4817,
      "time_per_iteration": 3.5299205780029297
    },
    {
      "auxiliary_loss_clip": 0.01283405,
      "auxiliary_loss_mlp": 0.01025328,
      "balance_loss_clip": 1.04765105,
      "balance_loss_mlp": 1.0184021,
      "epoch": 0.5793302470991403,
      "flos": 21470056669440.0,
      "grad_norm": 1.75175457919788,
      "language_loss": 0.78471959,
      "learning_rate": 1.5862595148876559e-06,
      "loss": 0.80780685,
      "num_input_tokens_seen": 103945835,
      "step": 4818,
      "time_per_iteration": 2.6879210472106934
    },
    {
      "auxiliary_loss_clip": 0.01387243,
      "auxiliary_loss_mlp": 0.01033536,
      "balance_loss_clip": 1.04950726,
      "balance_loss_mlp": 1.02523923,
      "epoch": 0.5794504899897793,
      "flos": 12711013367040.0,
      "grad_norm": 2.33178823390954,
      "language_loss": 0.76065099,
      "learning_rate": 1.58549742505406e-06,
      "loss": 0.78485882,
      "num_input_tokens_seen": 103960580,
      "step": 4819,
      "time_per_iteration": 3.6230781078338623
    },
    {
      "auxiliary_loss_clip": 0.01180911,
      "auxiliary_loss_mlp": 0.01028008,
      "balance_loss_clip": 1.0504508,
      "balance_loss_mlp": 1.02004433,
      "epoch": 0.5795707328804185,
      "flos": 14867054300160.0,
      "grad_norm": 2.5995514640334965,
      "language_loss": 0.75689006,
      "learning_rate": 1.5847353981001747e-06,
      "loss": 0.77897924,
      "num_input_tokens_seen": 103977760,
      "step": 4820,
      "time_per_iteration": 3.4900455474853516
    },
    {
      "auxiliary_loss_clip": 0.01278362,
      "auxiliary_loss_mlp": 0.01029009,
      "balance_loss_clip": 1.04513121,
      "balance_loss_mlp": 1.02093256,
      "epoch": 0.5796909757710575,
      "flos": 36430046432640.0,
      "grad_norm": 1.744666156371003,
      "language_loss": 0.69842374,
      "learning_rate": 1.5839734341415993e-06,
      "loss": 0.72149742,
      "num_input_tokens_seen": 103999960,
      "step": 4821,
      "time_per_iteration": 2.874349355697632
    },
    {
      "auxiliary_loss_clip": 0.01227997,
      "auxiliary_loss_mlp": 0.01027065,
      "balance_loss_clip": 1.05479157,
      "balance_loss_mlp": 1.02055883,
      "epoch": 0.5798112186616966,
      "flos": 23039891642880.0,
      "grad_norm": 1.5709638015616176,
      "language_loss": 0.76552582,
      "learning_rate": 1.5832115332939238e-06,
      "loss": 0.78807646,
      "num_input_tokens_seen": 104018400,
      "step": 4822,
      "time_per_iteration": 2.662214756011963
    },
    {
      "auxiliary_loss_clip": 0.01232956,
      "auxiliary_loss_mlp": 0.01028738,
      "balance_loss_clip": 1.05211067,
      "balance_loss_mlp": 1.02162111,
      "epoch": 0.5799314615523358,
      "flos": 16652604401280.0,
      "grad_norm": 1.5766886873598256,
      "language_loss": 0.74722499,
      "learning_rate": 1.5824496956727272e-06,
      "loss": 0.76984191,
      "num_input_tokens_seen": 104035605,
      "step": 4823,
      "time_per_iteration": 2.6852262020111084
    },
    {
      "auxiliary_loss_clip": 0.01278623,
      "auxiliary_loss_mlp": 0.01024486,
      "balance_loss_clip": 1.04698515,
      "balance_loss_mlp": 1.01735163,
      "epoch": 0.5800517044429748,
      "flos": 20485673470080.0,
      "grad_norm": 2.472379708660372,
      "language_loss": 0.73089969,
      "learning_rate": 1.5816879213935797e-06,
      "loss": 0.75393081,
      "num_input_tokens_seen": 104054415,
      "step": 4824,
      "time_per_iteration": 3.501483201980591
    },
    {
      "auxiliary_loss_clip": 0.0122719,
      "auxiliary_loss_mlp": 0.01026735,
      "balance_loss_clip": 1.05097592,
      "balance_loss_mlp": 1.01983857,
      "epoch": 0.5801719473336139,
      "flos": 31538258968320.0,
      "grad_norm": 1.806251251230941,
      "language_loss": 0.79887593,
      "learning_rate": 1.5809262105720416e-06,
      "loss": 0.82141519,
      "num_input_tokens_seen": 104075455,
      "step": 4825,
      "time_per_iteration": 2.693979263305664
    },
    {
      "auxiliary_loss_clip": 0.01179161,
      "auxiliary_loss_mlp": 0.01028089,
      "balance_loss_clip": 1.05166435,
      "balance_loss_mlp": 1.02097821,
      "epoch": 0.580292190224253,
      "flos": 20375966355840.0,
      "grad_norm": 1.869206833053959,
      "language_loss": 0.7947346,
      "learning_rate": 1.5801645633236644e-06,
      "loss": 0.81680715,
      "num_input_tokens_seen": 104096440,
      "step": 4826,
      "time_per_iteration": 2.630711078643799
    },
    {
      "auxiliary_loss_clip": 0.01273623,
      "auxiliary_loss_mlp": 0.01028873,
      "balance_loss_clip": 1.04527676,
      "balance_loss_mlp": 1.02136636,
      "epoch": 0.5804124331148921,
      "flos": 26615373304320.0,
      "grad_norm": 1.90727517965331,
      "language_loss": 0.77544498,
      "learning_rate": 1.579402979763989e-06,
      "loss": 0.79846996,
      "num_input_tokens_seen": 104116775,
      "step": 4827,
      "time_per_iteration": 2.6901824474334717
    },
    {
      "auxiliary_loss_clip": 0.01342219,
      "auxiliary_loss_mlp": 0.01024883,
      "balance_loss_clip": 1.04652309,
      "balance_loss_mlp": 1.01785302,
      "epoch": 0.5805326760055312,
      "flos": 13478496289920.0,
      "grad_norm": 2.174281632677387,
      "language_loss": 0.81211573,
      "learning_rate": 1.578641460008548e-06,
      "loss": 0.8357867,
      "num_input_tokens_seen": 104134510,
      "step": 4828,
      "time_per_iteration": 2.7782506942749023
    },
    {
      "auxiliary_loss_clip": 0.01225803,
      "auxiliary_loss_mlp": 0.01026086,
      "balance_loss_clip": 1.04972959,
      "balance_loss_mlp": 1.01895452,
      "epoch": 0.5806529188961702,
      "flos": 12091374823680.0,
      "grad_norm": 2.2979740790139536,
      "language_loss": 0.68355614,
      "learning_rate": 1.5778800041728613e-06,
      "loss": 0.70607501,
      "num_input_tokens_seen": 104150800,
      "step": 4829,
      "time_per_iteration": 2.592829942703247
    },
    {
      "auxiliary_loss_clip": 0.01225674,
      "auxiliary_loss_mlp": 0.01023594,
      "balance_loss_clip": 1.05074883,
      "balance_loss_mlp": 1.0166831,
      "epoch": 0.5807731617868094,
      "flos": 26214107495040.0,
      "grad_norm": 1.5467400022033504,
      "language_loss": 0.66102952,
      "learning_rate": 1.577118612372443e-06,
      "loss": 0.68352222,
      "num_input_tokens_seen": 104172640,
      "step": 4830,
      "time_per_iteration": 2.8792619705200195
    },
    {
      "auxiliary_loss_clip": 0.01276108,
      "auxiliary_loss_mlp": 0.02570292,
      "balance_loss_clip": 1.04701471,
      "balance_loss_mlp": 1.00004578,
      "epoch": 0.5808934046774484,
      "flos": 37962139190400.0,
      "grad_norm": 2.4570303343346596,
      "language_loss": 0.70617473,
      "learning_rate": 1.5763572847227943e-06,
      "loss": 0.74463868,
      "num_input_tokens_seen": 104193525,
      "step": 4831,
      "time_per_iteration": 2.797685146331787
    },
    {
      "auxiliary_loss_clip": 0.01227385,
      "auxiliary_loss_mlp": 0.01025088,
      "balance_loss_clip": 1.04743028,
      "balance_loss_mlp": 1.01861811,
      "epoch": 0.5810136475680875,
      "flos": 20485853038080.0,
      "grad_norm": 2.3535103702669082,
      "language_loss": 0.81662625,
      "learning_rate": 1.5755960213394091e-06,
      "loss": 0.83915102,
      "num_input_tokens_seen": 104210625,
      "step": 4832,
      "time_per_iteration": 2.703148603439331
    },
    {
      "auxiliary_loss_clip": 0.01233476,
      "auxiliary_loss_mlp": 0.01025771,
      "balance_loss_clip": 1.04698122,
      "balance_loss_mlp": 1.01896095,
      "epoch": 0.5811338904587267,
      "flos": 17530153574400.0,
      "grad_norm": 1.8644100318575374,
      "language_loss": 0.78451765,
      "learning_rate": 1.5748348223377703e-06,
      "loss": 0.80711007,
      "num_input_tokens_seen": 104228180,
      "step": 4833,
      "time_per_iteration": 2.807342529296875
    },
    {
      "auxiliary_loss_clip": 0.012774,
      "auxiliary_loss_mlp": 0.01023925,
      "balance_loss_clip": 1.0494554,
      "balance_loss_mlp": 1.0169754,
      "epoch": 0.5812541333493657,
      "flos": 19458017360640.0,
      "grad_norm": 1.5944308543982018,
      "language_loss": 0.78098243,
      "learning_rate": 1.5740736878333507e-06,
      "loss": 0.80399573,
      "num_input_tokens_seen": 104246020,
      "step": 4834,
      "time_per_iteration": 2.6747958660125732
    },
    {
      "auxiliary_loss_clip": 0.01283665,
      "auxiliary_loss_mlp": 0.01027039,
      "balance_loss_clip": 1.04823315,
      "balance_loss_mlp": 1.0192194,
      "epoch": 0.5813743762400048,
      "flos": 20594949621120.0,
      "grad_norm": 4.201151641071148,
      "language_loss": 0.78006279,
      "learning_rate": 1.5733126179416143e-06,
      "loss": 0.80316985,
      "num_input_tokens_seen": 104260505,
      "step": 4835,
      "time_per_iteration": 2.6477737426757812
    },
    {
      "auxiliary_loss_clip": 0.01226509,
      "auxiliary_loss_mlp": 0.0102201,
      "balance_loss_clip": 1.04924583,
      "balance_loss_mlp": 1.01481545,
      "epoch": 0.5814946191306439,
      "flos": 33178227246720.0,
      "grad_norm": 2.215275937590883,
      "language_loss": 0.72159445,
      "learning_rate": 1.5725516127780137e-06,
      "loss": 0.74407959,
      "num_input_tokens_seen": 104282640,
      "step": 4836,
      "time_per_iteration": 2.802097797393799
    },
    {
      "auxiliary_loss_clip": 0.01234239,
      "auxiliary_loss_mlp": 0.01024288,
      "balance_loss_clip": 1.04861546,
      "balance_loss_mlp": 1.01698601,
      "epoch": 0.581614862021283,
      "flos": 16143283503360.0,
      "grad_norm": 3.713376823607042,
      "language_loss": 0.88857073,
      "learning_rate": 1.5717906724579943e-06,
      "loss": 0.91115594,
      "num_input_tokens_seen": 104299700,
      "step": 4837,
      "time_per_iteration": 2.6694746017456055
    },
    {
      "auxiliary_loss_clip": 0.01337403,
      "auxiliary_loss_mlp": 0.01026032,
      "balance_loss_clip": 1.0460906,
      "balance_loss_mlp": 1.01874816,
      "epoch": 0.581735104911922,
      "flos": 33802642298880.0,
      "grad_norm": 2.1032924858356212,
      "language_loss": 0.68200749,
      "learning_rate": 1.571029797096989e-06,
      "loss": 0.70564181,
      "num_input_tokens_seen": 104320805,
      "step": 4838,
      "time_per_iteration": 2.7962183952331543
    },
    {
      "auxiliary_loss_clip": 0.01177896,
      "auxiliary_loss_mlp": 0.01026035,
      "balance_loss_clip": 1.04995227,
      "balance_loss_mlp": 1.01908529,
      "epoch": 0.5818553478025612,
      "flos": 23331163029120.0,
      "grad_norm": 1.871598908674626,
      "language_loss": 0.79230797,
      "learning_rate": 1.570268986810423e-06,
      "loss": 0.81434727,
      "num_input_tokens_seen": 104340700,
      "step": 4839,
      "time_per_iteration": 2.6555752754211426
    },
    {
      "auxiliary_loss_clip": 0.01275842,
      "auxiliary_loss_mlp": 0.01025112,
      "balance_loss_clip": 1.04710174,
      "balance_loss_mlp": 1.01799846,
      "epoch": 0.5819755906932003,
      "flos": 20996143603200.0,
      "grad_norm": 4.941044375805735,
      "language_loss": 0.74943435,
      "learning_rate": 1.5695082417137096e-06,
      "loss": 0.77244389,
      "num_input_tokens_seen": 104358575,
      "step": 4840,
      "time_per_iteration": 2.612604856491089
    },
    {
      "auxiliary_loss_clip": 0.01275816,
      "auxiliary_loss_mlp": 0.01027993,
      "balance_loss_clip": 1.04491973,
      "balance_loss_mlp": 1.02139497,
      "epoch": 0.5820958335838393,
      "flos": 21431668008960.0,
      "grad_norm": 1.7521577694618597,
      "language_loss": 0.75805008,
      "learning_rate": 1.5687475619222539e-06,
      "loss": 0.78108817,
      "num_input_tokens_seen": 104378530,
      "step": 4841,
      "time_per_iteration": 2.7020676136016846
    },
    {
      "auxiliary_loss_clip": 0.01278406,
      "auxiliary_loss_mlp": 0.01025577,
      "balance_loss_clip": 1.04645181,
      "balance_loss_mlp": 1.01849937,
      "epoch": 0.5822160764744785,
      "flos": 17967473660160.0,
      "grad_norm": 2.185038307732515,
      "language_loss": 0.73468983,
      "learning_rate": 1.5679869475514496e-06,
      "loss": 0.75772965,
      "num_input_tokens_seen": 104395465,
      "step": 4842,
      "time_per_iteration": 2.6540772914886475
    },
    {
      "auxiliary_loss_clip": 0.01225694,
      "auxiliary_loss_mlp": 0.01030649,
      "balance_loss_clip": 1.04824972,
      "balance_loss_mlp": 1.02313638,
      "epoch": 0.5823363193651175,
      "flos": 23033858158080.0,
      "grad_norm": 2.014111977216435,
      "language_loss": 0.81173813,
      "learning_rate": 1.567226398716682e-06,
      "loss": 0.83430159,
      "num_input_tokens_seen": 104415380,
      "step": 4843,
      "time_per_iteration": 3.5804665088653564
    },
    {
      "auxiliary_loss_clip": 0.01288374,
      "auxiliary_loss_mlp": 0.01035204,
      "balance_loss_clip": 1.05057728,
      "balance_loss_mlp": 1.02728224,
      "epoch": 0.5824565622557566,
      "flos": 32891840110080.0,
      "grad_norm": 1.729620737474344,
      "language_loss": 0.61962694,
      "learning_rate": 1.566465915533326e-06,
      "loss": 0.64286274,
      "num_input_tokens_seen": 104437410,
      "step": 4844,
      "time_per_iteration": 3.764631748199463
    },
    {
      "auxiliary_loss_clip": 0.01225497,
      "auxiliary_loss_mlp": 0.0102711,
      "balance_loss_clip": 1.04804242,
      "balance_loss_mlp": 1.01986182,
      "epoch": 0.5825768051463958,
      "flos": 22229674513920.0,
      "grad_norm": 2.6272636903491002,
      "language_loss": 0.8823477,
      "learning_rate": 1.5657054981167458e-06,
      "loss": 0.90487379,
      "num_input_tokens_seen": 104456305,
      "step": 4845,
      "time_per_iteration": 3.5296711921691895
    },
    {
      "auxiliary_loss_clip": 0.01225208,
      "auxiliary_loss_mlp": 0.01023056,
      "balance_loss_clip": 1.04730153,
      "balance_loss_mlp": 1.0163238,
      "epoch": 0.5826970480370348,
      "flos": 28001561016960.0,
      "grad_norm": 1.8206936142666568,
      "language_loss": 0.67994767,
      "learning_rate": 1.5649451465822965e-06,
      "loss": 0.70243037,
      "num_input_tokens_seen": 104477695,
      "step": 4846,
      "time_per_iteration": 2.704608201980591
    },
    {
      "auxiliary_loss_clip": 0.01376499,
      "auxiliary_loss_mlp": 0.01028414,
      "balance_loss_clip": 1.0482645,
      "balance_loss_mlp": 1.02147293,
      "epoch": 0.5828172909276739,
      "flos": 17858053854720.0,
      "grad_norm": 1.73099298143603,
      "language_loss": 0.83861411,
      "learning_rate": 1.5641848610453218e-06,
      "loss": 0.86266327,
      "num_input_tokens_seen": 104496355,
      "step": 4847,
      "time_per_iteration": 2.705509662628174
    },
    {
      "auxiliary_loss_clip": 0.01226127,
      "auxiliary_loss_mlp": 0.0102534,
      "balance_loss_clip": 1.04932523,
      "balance_loss_mlp": 1.01787758,
      "epoch": 0.582937533818313,
      "flos": 19865244827520.0,
      "grad_norm": 2.4619939873528494,
      "language_loss": 0.86238885,
      "learning_rate": 1.563424641621158e-06,
      "loss": 0.88490349,
      "num_input_tokens_seen": 104515535,
      "step": 4848,
      "time_per_iteration": 2.622224807739258
    },
    {
      "auxiliary_loss_clip": 0.01282426,
      "auxiliary_loss_mlp": 0.01031623,
      "balance_loss_clip": 1.0486939,
      "balance_loss_mlp": 1.02404141,
      "epoch": 0.5830577767089521,
      "flos": 26870734068480.0,
      "grad_norm": 2.339824930721859,
      "language_loss": 0.69978344,
      "learning_rate": 1.5626644884251282e-06,
      "loss": 0.72292393,
      "num_input_tokens_seen": 104535055,
      "step": 4849,
      "time_per_iteration": 2.7268850803375244
    },
    {
      "auxiliary_loss_clip": 0.01180555,
      "auxiliary_loss_mlp": 0.01022902,
      "balance_loss_clip": 1.05125833,
      "balance_loss_mlp": 1.01630068,
      "epoch": 0.5831780195995911,
      "flos": 25298205575040.0,
      "grad_norm": 1.8180830366339165,
      "language_loss": 0.88092351,
      "learning_rate": 1.5619044015725488e-06,
      "loss": 0.9029581,
      "num_input_tokens_seen": 104554745,
      "step": 4850,
      "time_per_iteration": 3.5213418006896973
    },
    {
      "auxiliary_loss_clip": 0.01188686,
      "auxiliary_loss_mlp": 0.0102968,
      "balance_loss_clip": 1.05549383,
      "balance_loss_mlp": 1.02192533,
      "epoch": 0.5832982624902303,
      "flos": 14756988049920.0,
      "grad_norm": 2.286041133119409,
      "language_loss": 0.86902404,
      "learning_rate": 1.5611443811787224e-06,
      "loss": 0.8912077,
      "num_input_tokens_seen": 104568870,
      "step": 4851,
      "time_per_iteration": 2.629037618637085
    },
    {
      "auxiliary_loss_clip": 0.01223223,
      "auxiliary_loss_mlp": 0.01027564,
      "balance_loss_clip": 1.04888868,
      "balance_loss_mlp": 1.02030969,
      "epoch": 0.5834185053808694,
      "flos": 20444555376000.0,
      "grad_norm": 2.262701474149554,
      "language_loss": 0.69566929,
      "learning_rate": 1.560384427358945e-06,
      "loss": 0.71817714,
      "num_input_tokens_seen": 104588415,
      "step": 4852,
      "time_per_iteration": 2.6456990242004395
    },
    {
      "auxiliary_loss_clip": 0.01274999,
      "auxiliary_loss_mlp": 0.01027871,
      "balance_loss_clip": 1.04356802,
      "balance_loss_mlp": 1.02019954,
      "epoch": 0.5835387482715084,
      "flos": 27200394115200.0,
      "grad_norm": 3.4025726155310583,
      "language_loss": 0.73109829,
      "learning_rate": 1.5596245402284998e-06,
      "loss": 0.75412703,
      "num_input_tokens_seen": 104611940,
      "step": 4853,
      "time_per_iteration": 2.7701127529144287
    },
    {
      "auxiliary_loss_clip": 0.01233587,
      "auxiliary_loss_mlp": 0.0102899,
      "balance_loss_clip": 1.05390954,
      "balance_loss_mlp": 1.02195024,
      "epoch": 0.5836589911621476,
      "flos": 16654615562880.0,
      "grad_norm": 1.8162073619363863,
      "language_loss": 0.82020545,
      "learning_rate": 1.5588647199026619e-06,
      "loss": 0.84283125,
      "num_input_tokens_seen": 104629675,
      "step": 4854,
      "time_per_iteration": 2.5997965335845947
    },
    {
      "auxiliary_loss_clip": 0.01186812,
      "auxiliary_loss_mlp": 0.0102854,
      "balance_loss_clip": 1.05503798,
      "balance_loss_mlp": 1.02069044,
      "epoch": 0.5837792340527866,
      "flos": 20446817932800.0,
      "grad_norm": 2.136248848528647,
      "language_loss": 0.87464166,
      "learning_rate": 1.5581049664966956e-06,
      "loss": 0.89679515,
      "num_input_tokens_seen": 104647435,
      "step": 4855,
      "time_per_iteration": 2.6021649837493896
    },
    {
      "auxiliary_loss_clip": 0.01341042,
      "auxiliary_loss_mlp": 0.01001956,
      "balance_loss_clip": 1.01051557,
      "balance_loss_mlp": 1.0008831,
      "epoch": 0.5838994769434257,
      "flos": 65995480765440.0,
      "grad_norm": 0.9887608753668745,
      "language_loss": 0.65036178,
      "learning_rate": 1.5573452801258545e-06,
      "loss": 0.67379177,
      "num_input_tokens_seen": 104694605,
      "step": 4856,
      "time_per_iteration": 3.121966600418091
    },
    {
      "auxiliary_loss_clip": 0.01239117,
      "auxiliary_loss_mlp": 0.01027216,
      "balance_loss_clip": 1.05273092,
      "balance_loss_mlp": 1.01985431,
      "epoch": 0.5840197198340649,
      "flos": 21470523546240.0,
      "grad_norm": 2.2228892841275285,
      "language_loss": 0.63716245,
      "learning_rate": 1.5565856609053824e-06,
      "loss": 0.6598258,
      "num_input_tokens_seen": 104713400,
      "step": 4857,
      "time_per_iteration": 2.6598527431488037
    },
    {
      "auxiliary_loss_clip": 0.01182467,
      "auxiliary_loss_mlp": 0.0103396,
      "balance_loss_clip": 1.05282044,
      "balance_loss_mlp": 1.02590704,
      "epoch": 0.5841399627247039,
      "flos": 19135144984320.0,
      "grad_norm": 2.194533631779001,
      "language_loss": 0.80119431,
      "learning_rate": 1.5558261089505127e-06,
      "loss": 0.82335854,
      "num_input_tokens_seen": 104732130,
      "step": 4858,
      "time_per_iteration": 2.615567684173584
    },
    {
      "auxiliary_loss_clip": 0.01230197,
      "auxiliary_loss_mlp": 0.010275,
      "balance_loss_clip": 1.05140138,
      "balance_loss_mlp": 1.02010274,
      "epoch": 0.584260205615343,
      "flos": 26425692558720.0,
      "grad_norm": 2.0890239022740587,
      "language_loss": 0.80079508,
      "learning_rate": 1.5550666243764697e-06,
      "loss": 0.82337201,
      "num_input_tokens_seen": 104750290,
      "step": 4859,
      "time_per_iteration": 2.6441969871520996
    },
    {
      "auxiliary_loss_clip": 0.01229584,
      "auxiliary_loss_mlp": 0.01027088,
      "balance_loss_clip": 1.05061173,
      "balance_loss_mlp": 1.01978958,
      "epoch": 0.584380448505982,
      "flos": 13881809174400.0,
      "grad_norm": 2.35505343464797,
      "language_loss": 0.77842104,
      "learning_rate": 1.554307207298465e-06,
      "loss": 0.80098778,
      "num_input_tokens_seen": 104768550,
      "step": 4860,
      "time_per_iteration": 2.62872314453125
    },
    {
      "auxiliary_loss_clip": 0.01183554,
      "auxiliary_loss_mlp": 0.01024696,
      "balance_loss_clip": 1.05257416,
      "balance_loss_mlp": 1.0166254,
      "epoch": 0.5845006913966212,
      "flos": 21543709507200.0,
      "grad_norm": 3.6071499289872677,
      "language_loss": 0.79036343,
      "learning_rate": 1.553547857831704e-06,
      "loss": 0.81244594,
      "num_input_tokens_seen": 104785060,
      "step": 4861,
      "time_per_iteration": 2.554435968399048
    },
    {
      "auxiliary_loss_clip": 0.01067526,
      "auxiliary_loss_mlp": 0.01002128,
      "balance_loss_clip": 1.00978303,
      "balance_loss_mlp": 1.00121641,
      "epoch": 0.5846209342872603,
      "flos": 58375452712320.0,
      "grad_norm": 0.8844613864978814,
      "language_loss": 0.64139724,
      "learning_rate": 1.5527885760913771e-06,
      "loss": 0.66209376,
      "num_input_tokens_seen": 104834950,
      "step": 4862,
      "time_per_iteration": 2.985801935195923
    },
    {
      "auxiliary_loss_clip": 0.01278866,
      "auxiliary_loss_mlp": 0.0102881,
      "balance_loss_clip": 1.05106258,
      "balance_loss_mlp": 1.02202106,
      "epoch": 0.5847411771778993,
      "flos": 18588045957120.0,
      "grad_norm": 1.624249311798478,
      "language_loss": 0.7677232,
      "learning_rate": 1.552029362192668e-06,
      "loss": 0.79079998,
      "num_input_tokens_seen": 104854210,
      "step": 4863,
      "time_per_iteration": 2.644968032836914
    },
    {
      "auxiliary_loss_clip": 0.01325126,
      "auxiliary_loss_mlp": 0.01026871,
      "balance_loss_clip": 1.0466907,
      "balance_loss_mlp": 1.0194298,
      "epoch": 0.5848614200685385,
      "flos": 24240780069120.0,
      "grad_norm": 3.3786523461187525,
      "language_loss": 0.72816324,
      "learning_rate": 1.5512702162507478e-06,
      "loss": 0.75168318,
      "num_input_tokens_seen": 104874525,
      "step": 4864,
      "time_per_iteration": 2.7942891120910645
    },
    {
      "auxiliary_loss_clip": 0.01171997,
      "auxiliary_loss_mlp": 0.01001188,
      "balance_loss_clip": 1.00777459,
      "balance_loss_mlp": 1.00025773,
      "epoch": 0.5849816629591775,
      "flos": 71660245933440.0,
      "grad_norm": 1.0844009218728101,
      "language_loss": 0.55655968,
      "learning_rate": 1.5505111383807792e-06,
      "loss": 0.57829154,
      "num_input_tokens_seen": 104937195,
      "step": 4865,
      "time_per_iteration": 3.297513484954834
    },
    {
      "auxiliary_loss_clip": 0.01380598,
      "auxiliary_loss_mlp": 0.01033974,
      "balance_loss_clip": 1.04572761,
      "balance_loss_mlp": 1.02654767,
      "epoch": 0.5851019058498166,
      "flos": 23802095266560.0,
      "grad_norm": 1.9194578783197533,
      "language_loss": 0.80840737,
      "learning_rate": 1.5497521286979138e-06,
      "loss": 0.83255303,
      "num_input_tokens_seen": 104957435,
      "step": 4866,
      "time_per_iteration": 2.741168737411499
    },
    {
      "auxiliary_loss_clip": 0.01335428,
      "auxiliary_loss_mlp": 0.01025441,
      "balance_loss_clip": 1.04782403,
      "balance_loss_mlp": 1.01721573,
      "epoch": 0.5852221487404557,
      "flos": 24388516707840.0,
      "grad_norm": 2.0782930992104696,
      "language_loss": 0.74765921,
      "learning_rate": 1.5489931873172927e-06,
      "loss": 0.77126789,
      "num_input_tokens_seen": 104978755,
      "step": 4867,
      "time_per_iteration": 2.766758680343628
    },
    {
      "auxiliary_loss_clip": 0.0141486,
      "auxiliary_loss_mlp": 0.01029973,
      "balance_loss_clip": 1.038481,
      "balance_loss_mlp": 1.02252555,
      "epoch": 0.5853423916310948,
      "flos": 27271425260160.0,
      "grad_norm": 4.446565618672845,
      "language_loss": 0.79279333,
      "learning_rate": 1.5482343143540467e-06,
      "loss": 0.81724173,
      "num_input_tokens_seen": 105000020,
      "step": 4868,
      "time_per_iteration": 2.8766002655029297
    },
    {
      "auxiliary_loss_clip": 0.01231756,
      "auxiliary_loss_mlp": 0.02565804,
      "balance_loss_clip": 1.04653251,
      "balance_loss_mlp": 1.00007367,
      "epoch": 0.5854626345217339,
      "flos": 11983786611840.0,
      "grad_norm": 2.213372081468802,
      "language_loss": 0.82905316,
      "learning_rate": 1.547475509923295e-06,
      "loss": 0.86702877,
      "num_input_tokens_seen": 105017060,
      "step": 4869,
      "time_per_iteration": 3.7288763523101807
    },
    {
      "auxiliary_loss_clip": 0.01183239,
      "auxiliary_loss_mlp": 0.01000303,
      "balance_loss_clip": 1.00789404,
      "balance_loss_mlp": 0.99930811,
      "epoch": 0.585582877412373,
      "flos": 64342335173760.0,
      "grad_norm": 0.7251557890195105,
      "language_loss": 0.56085795,
      "learning_rate": 1.5467167741401495e-06,
      "loss": 0.5826934,
      "num_input_tokens_seen": 105078540,
      "step": 4870,
      "time_per_iteration": 4.171045541763306
    },
    {
      "auxiliary_loss_clip": 0.01278809,
      "auxiliary_loss_mlp": 0.01025082,
      "balance_loss_clip": 1.04507041,
      "balance_loss_mlp": 1.01781023,
      "epoch": 0.5857031203030121,
      "flos": 17011926103680.0,
      "grad_norm": 2.4884213988673,
      "language_loss": 0.71343338,
      "learning_rate": 1.5459581071197083e-06,
      "loss": 0.73647225,
      "num_input_tokens_seen": 105094200,
      "step": 4871,
      "time_per_iteration": 3.5228490829467773
    },
    {
      "auxiliary_loss_clip": 0.01232077,
      "auxiliary_loss_mlp": 0.01028523,
      "balance_loss_clip": 1.05172849,
      "balance_loss_mlp": 1.02141809,
      "epoch": 0.5858233631936511,
      "flos": 20885682303360.0,
      "grad_norm": 2.590804599974693,
      "language_loss": 0.83063281,
      "learning_rate": 1.5451995089770624e-06,
      "loss": 0.85323882,
      "num_input_tokens_seen": 105113985,
      "step": 4872,
      "time_per_iteration": 2.6350066661834717
    },
    {
      "auxiliary_loss_clip": 0.01179327,
      "auxiliary_loss_mlp": 0.0102591,
      "balance_loss_clip": 1.05222344,
      "balance_loss_mlp": 1.01925802,
      "epoch": 0.5859436060842903,
      "flos": 23191902000000.0,
      "grad_norm": 1.4767662141615823,
      "language_loss": 0.71815538,
      "learning_rate": 1.5444409798272885e-06,
      "loss": 0.74020773,
      "num_input_tokens_seen": 105138075,
      "step": 4873,
      "time_per_iteration": 2.697554349899292
    },
    {
      "auxiliary_loss_clip": 0.01328315,
      "auxiliary_loss_mlp": 0.01025342,
      "balance_loss_clip": 1.0461278,
      "balance_loss_mlp": 1.01828814,
      "epoch": 0.5860638489749294,
      "flos": 22492648961280.0,
      "grad_norm": 1.735769549963971,
      "language_loss": 0.80273783,
      "learning_rate": 1.543682519785456e-06,
      "loss": 0.82627439,
      "num_input_tokens_seen": 105156555,
      "step": 4874,
      "time_per_iteration": 2.725426435470581
    },
    {
      "auxiliary_loss_clip": 0.01277407,
      "auxiliary_loss_mlp": 0.01027198,
      "balance_loss_clip": 1.04674172,
      "balance_loss_mlp": 1.02033424,
      "epoch": 0.5861840918655684,
      "flos": 17566243764480.0,
      "grad_norm": 2.9006918176079295,
      "language_loss": 0.80534518,
      "learning_rate": 1.5429241289666219e-06,
      "loss": 0.82839125,
      "num_input_tokens_seen": 105174055,
      "step": 4875,
      "time_per_iteration": 2.663512706756592
    },
    {
      "auxiliary_loss_clip": 0.01274737,
      "auxiliary_loss_mlp": 0.01028416,
      "balance_loss_clip": 1.04751182,
      "balance_loss_mlp": 1.02128422,
      "epoch": 0.5863043347562076,
      "flos": 25556152118400.0,
      "grad_norm": 2.094731679885173,
      "language_loss": 0.69787788,
      "learning_rate": 1.5421658074858342e-06,
      "loss": 0.72090936,
      "num_input_tokens_seen": 105192160,
      "step": 4876,
      "time_per_iteration": 3.6098787784576416
    },
    {
      "auxiliary_loss_clip": 0.01275336,
      "auxiliary_loss_mlp": 0.01028756,
      "balance_loss_clip": 1.04776287,
      "balance_loss_mlp": 1.0206883,
      "epoch": 0.5864245776468466,
      "flos": 20667525050880.0,
      "grad_norm": 2.1159662685825618,
      "language_loss": 0.66213822,
      "learning_rate": 1.5414075554581298e-06,
      "loss": 0.68517923,
      "num_input_tokens_seen": 105210205,
      "step": 4877,
      "time_per_iteration": 2.7206687927246094
    },
    {
      "auxiliary_loss_clip": 0.01183381,
      "auxiliary_loss_mlp": 0.010311,
      "balance_loss_clip": 1.05252516,
      "balance_loss_mlp": 1.02341104,
      "epoch": 0.5865448205374857,
      "flos": 28913907490560.0,
      "grad_norm": 2.4450014153567,
      "language_loss": 0.780402,
      "learning_rate": 1.5406493729985348e-06,
      "loss": 0.80254686,
      "num_input_tokens_seen": 105229400,
      "step": 4878,
      "time_per_iteration": 2.6856584548950195
    },
    {
      "auxiliary_loss_clip": 0.01286675,
      "auxiliary_loss_mlp": 0.02566839,
      "balance_loss_clip": 1.04997826,
      "balance_loss_mlp": 1.0000304,
      "epoch": 0.5866650634281249,
      "flos": 25842575168640.0,
      "grad_norm": 2.049020735797245,
      "language_loss": 0.720415,
      "learning_rate": 1.5398912602220644e-06,
      "loss": 0.75895011,
      "num_input_tokens_seen": 105248675,
      "step": 4879,
      "time_per_iteration": 2.747319459915161
    },
    {
      "auxiliary_loss_clip": 0.01292535,
      "auxiliary_loss_mlp": 0.01030212,
      "balance_loss_clip": 1.04828262,
      "balance_loss_mlp": 1.02278793,
      "epoch": 0.5867853063187639,
      "flos": 17052325925760.0,
      "grad_norm": 2.779498105598538,
      "language_loss": 0.78762424,
      "learning_rate": 1.539133217243724e-06,
      "loss": 0.81085169,
      "num_input_tokens_seen": 105265695,
      "step": 4880,
      "time_per_iteration": 2.7096054553985596
    },
    {
      "auxiliary_loss_clip": 0.01334114,
      "auxiliary_loss_mlp": 0.01032632,
      "balance_loss_clip": 1.0479629,
      "balance_loss_mlp": 1.02470422,
      "epoch": 0.586905549209403,
      "flos": 24645026707200.0,
      "grad_norm": 2.561190026680902,
      "language_loss": 0.75993133,
      "learning_rate": 1.5383752441785081e-06,
      "loss": 0.78359878,
      "num_input_tokens_seen": 105284920,
      "step": 4881,
      "time_per_iteration": 2.7251193523406982
    },
    {
      "auxiliary_loss_clip": 0.01234975,
      "auxiliary_loss_mlp": 0.01032836,
      "balance_loss_clip": 1.05078316,
      "balance_loss_mlp": 1.02502489,
      "epoch": 0.5870257921000421,
      "flos": 14720538723840.0,
      "grad_norm": 2.218530458053412,
      "language_loss": 0.8600018,
      "learning_rate": 1.5376173411414003e-06,
      "loss": 0.88267994,
      "num_input_tokens_seen": 105302960,
      "step": 4882,
      "time_per_iteration": 2.619295120239258
    },
    {
      "auxiliary_loss_clip": 0.01280071,
      "auxiliary_loss_mlp": 0.01026762,
      "balance_loss_clip": 1.04471278,
      "balance_loss_mlp": 1.01933801,
      "epoch": 0.5871460349906812,
      "flos": 23914998691200.0,
      "grad_norm": 1.8752600933302197,
      "language_loss": 0.78839397,
      "learning_rate": 1.5368595082473753e-06,
      "loss": 0.81146234,
      "num_input_tokens_seen": 105321260,
      "step": 4883,
      "time_per_iteration": 2.6768808364868164
    },
    {
      "auxiliary_loss_clip": 0.01232054,
      "auxiliary_loss_mlp": 0.01024328,
      "balance_loss_clip": 1.04954875,
      "balance_loss_mlp": 1.01713407,
      "epoch": 0.5872662778813202,
      "flos": 22164174063360.0,
      "grad_norm": 1.7917421240308316,
      "language_loss": 0.77925515,
      "learning_rate": 1.5361017456113935e-06,
      "loss": 0.80181897,
      "num_input_tokens_seen": 105341610,
      "step": 4884,
      "time_per_iteration": 2.6603455543518066
    },
    {
      "auxiliary_loss_clip": 0.01231894,
      "auxiliary_loss_mlp": 0.01028189,
      "balance_loss_clip": 1.04840565,
      "balance_loss_mlp": 1.02090216,
      "epoch": 0.5873865207719594,
      "flos": 18441925430400.0,
      "grad_norm": 2.1950775166792242,
      "language_loss": 0.85587752,
      "learning_rate": 1.5353440533484085e-06,
      "loss": 0.87847829,
      "num_input_tokens_seen": 105360465,
      "step": 4885,
      "time_per_iteration": 2.6024210453033447
    },
    {
      "auxiliary_loss_clip": 0.01282539,
      "auxiliary_loss_mlp": 0.01031288,
      "balance_loss_clip": 1.04921067,
      "balance_loss_mlp": 1.02365279,
      "epoch": 0.5875067636625985,
      "flos": 54015321427200.0,
      "grad_norm": 2.0214986564758655,
      "language_loss": 0.66419852,
      "learning_rate": 1.534586431573361e-06,
      "loss": 0.6873368,
      "num_input_tokens_seen": 105385405,
      "step": 4886,
      "time_per_iteration": 2.9422740936279297
    },
    {
      "auxiliary_loss_clip": 0.01426463,
      "auxiliary_loss_mlp": 0.01027698,
      "balance_loss_clip": 1.04163194,
      "balance_loss_mlp": 1.01939523,
      "epoch": 0.5876270065532375,
      "flos": 27995707100160.0,
      "grad_norm": 2.2577997180893403,
      "language_loss": 0.79083103,
      "learning_rate": 1.5338288804011817e-06,
      "loss": 0.81537265,
      "num_input_tokens_seen": 105404905,
      "step": 4887,
      "time_per_iteration": 2.800546884536743
    },
    {
      "auxiliary_loss_clip": 0.01279589,
      "auxiliary_loss_mlp": 0.01028535,
      "balance_loss_clip": 1.04636335,
      "balance_loss_mlp": 1.02118015,
      "epoch": 0.5877472494438767,
      "flos": 21361462876800.0,
      "grad_norm": 2.2275535363432515,
      "language_loss": 0.71010906,
      "learning_rate": 1.533071399946791e-06,
      "loss": 0.7331903,
      "num_input_tokens_seen": 105423650,
      "step": 4888,
      "time_per_iteration": 2.6662819385528564
    },
    {
      "auxiliary_loss_clip": 0.01180035,
      "auxiliary_loss_mlp": 0.01025186,
      "balance_loss_clip": 1.04598379,
      "balance_loss_mlp": 1.01808667,
      "epoch": 0.5878674923345157,
      "flos": 22383013674240.0,
      "grad_norm": 1.8765733569299288,
      "language_loss": 0.57339597,
      "learning_rate": 1.5323139903250977e-06,
      "loss": 0.5954482,
      "num_input_tokens_seen": 105444255,
      "step": 4889,
      "time_per_iteration": 2.6916184425354004
    },
    {
      "auxiliary_loss_clip": 0.01283737,
      "auxiliary_loss_mlp": 0.01030906,
      "balance_loss_clip": 1.05308914,
      "balance_loss_mlp": 1.02345252,
      "epoch": 0.5879877352251548,
      "flos": 21868664872320.0,
      "grad_norm": 1.505714500069263,
      "language_loss": 0.76921403,
      "learning_rate": 1.5315566516510002e-06,
      "loss": 0.79236042,
      "num_input_tokens_seen": 105462425,
      "step": 4890,
      "time_per_iteration": 2.693025827407837
    },
    {
      "auxiliary_loss_clip": 0.01182664,
      "auxiliary_loss_mlp": 0.01028408,
      "balance_loss_clip": 1.05311668,
      "balance_loss_mlp": 1.02071846,
      "epoch": 0.5881079781157939,
      "flos": 17493811989120.0,
      "grad_norm": 2.1091547560871056,
      "language_loss": 0.67522407,
      "learning_rate": 1.5307993840393857e-06,
      "loss": 0.69733477,
      "num_input_tokens_seen": 105480505,
      "step": 4891,
      "time_per_iteration": 2.5640146732330322
    },
    {
      "auxiliary_loss_clip": 0.01177848,
      "auxiliary_loss_mlp": 0.01026347,
      "balance_loss_clip": 1.04925764,
      "balance_loss_mlp": 1.01960874,
      "epoch": 0.588228221006433,
      "flos": 22601853285120.0,
      "grad_norm": 3.2765494452427752,
      "language_loss": 0.80537146,
      "learning_rate": 1.530042187605132e-06,
      "loss": 0.82741344,
      "num_input_tokens_seen": 105499760,
      "step": 4892,
      "time_per_iteration": 2.660757541656494
    },
    {
      "auxiliary_loss_clip": 0.01232932,
      "auxiliary_loss_mlp": 0.02559297,
      "balance_loss_clip": 1.05228662,
      "balance_loss_mlp": 1.00002623,
      "epoch": 0.5883484638970721,
      "flos": 26176939896960.0,
      "grad_norm": 1.4238618828705611,
      "language_loss": 0.8424834,
      "learning_rate": 1.5292850624631044e-06,
      "loss": 0.88040566,
      "num_input_tokens_seen": 105521955,
      "step": 4893,
      "time_per_iteration": 2.709372043609619
    },
    {
      "auxiliary_loss_clip": 0.01225392,
      "auxiliary_loss_mlp": 0.0102701,
      "balance_loss_clip": 1.04937983,
      "balance_loss_mlp": 1.01923776,
      "epoch": 0.5884687067877111,
      "flos": 30443737691520.0,
      "grad_norm": 1.9065226455021858,
      "language_loss": 0.80295539,
      "learning_rate": 1.5285280087281593e-06,
      "loss": 0.82547939,
      "num_input_tokens_seen": 105542685,
      "step": 4894,
      "time_per_iteration": 2.7115821838378906
    },
    {
      "auxiliary_loss_clip": 0.01174719,
      "auxiliary_loss_mlp": 0.01001952,
      "balance_loss_clip": 1.01021886,
      "balance_loss_mlp": 1.00104654,
      "epoch": 0.5885889496783503,
      "flos": 70507550580480.0,
      "grad_norm": 0.6382227889364126,
      "language_loss": 0.56590366,
      "learning_rate": 1.5277710265151398e-06,
      "loss": 0.58767033,
      "num_input_tokens_seen": 105612165,
      "step": 4895,
      "time_per_iteration": 4.418470859527588
    },
    {
      "auxiliary_loss_clip": 0.01231861,
      "auxiliary_loss_mlp": 0.01032544,
      "balance_loss_clip": 1.05003333,
      "balance_loss_mlp": 1.02432489,
      "epoch": 0.5887091925689893,
      "flos": 19098767485440.0,
      "grad_norm": 3.6914189502320265,
      "language_loss": 0.76945555,
      "learning_rate": 1.5270141159388803e-06,
      "loss": 0.7920996,
      "num_input_tokens_seen": 105629185,
      "step": 4896,
      "time_per_iteration": 3.5653738975524902
    },
    {
      "auxiliary_loss_clip": 0.0117933,
      "auxiliary_loss_mlp": 0.01030189,
      "balance_loss_clip": 1.04902351,
      "balance_loss_mlp": 1.02277994,
      "epoch": 0.5888294354596284,
      "flos": 23294282739840.0,
      "grad_norm": 1.6588395038730217,
      "language_loss": 0.80567241,
      "learning_rate": 1.526257277114203e-06,
      "loss": 0.82776761,
      "num_input_tokens_seen": 105650260,
      "step": 4897,
      "time_per_iteration": 3.5308895111083984
    },
    {
      "auxiliary_loss_clip": 0.01276996,
      "auxiliary_loss_mlp": 0.01023888,
      "balance_loss_clip": 1.04985988,
      "balance_loss_mlp": 1.01698852,
      "epoch": 0.5889496783502676,
      "flos": 21981532383360.0,
      "grad_norm": 2.185215540655273,
      "language_loss": 0.79906261,
      "learning_rate": 1.5255005101559201e-06,
      "loss": 0.82207143,
      "num_input_tokens_seen": 105667870,
      "step": 4898,
      "time_per_iteration": 2.7029061317443848
    },
    {
      "auxiliary_loss_clip": 0.01134194,
      "auxiliary_loss_mlp": 0.01029969,
      "balance_loss_clip": 1.05113661,
      "balance_loss_mlp": 1.02272153,
      "epoch": 0.5890699212409066,
      "flos": 21685233093120.0,
      "grad_norm": 1.9460835167240103,
      "language_loss": 0.76792991,
      "learning_rate": 1.524743815178833e-06,
      "loss": 0.78957152,
      "num_input_tokens_seen": 105685830,
      "step": 4899,
      "time_per_iteration": 2.607790946960449
    },
    {
      "auxiliary_loss_clip": 0.01280209,
      "auxiliary_loss_mlp": 0.01021578,
      "balance_loss_clip": 1.04641616,
      "balance_loss_mlp": 1.01505768,
      "epoch": 0.5891901641315457,
      "flos": 19464553635840.0,
      "grad_norm": 1.6834040087677489,
      "language_loss": 0.80887157,
      "learning_rate": 1.5239871922977315e-06,
      "loss": 0.83188939,
      "num_input_tokens_seen": 105705745,
      "step": 4900,
      "time_per_iteration": 2.8031342029571533
    },
    {
      "auxiliary_loss_clip": 0.01277642,
      "auxiliary_loss_mlp": 0.01026296,
      "balance_loss_clip": 1.04499936,
      "balance_loss_mlp": 1.01932204,
      "epoch": 0.5893104070221848,
      "flos": 19609884063360.0,
      "grad_norm": 2.0273388057386414,
      "language_loss": 0.8984099,
      "learning_rate": 1.523230641627394e-06,
      "loss": 0.9214493,
      "num_input_tokens_seen": 105724730,
      "step": 4901,
      "time_per_iteration": 2.6556529998779297
    },
    {
      "auxiliary_loss_clip": 0.01379361,
      "auxiliary_loss_mlp": 0.01029659,
      "balance_loss_clip": 1.04088378,
      "balance_loss_mlp": 1.02274787,
      "epoch": 0.5894306499128239,
      "flos": 29060063930880.0,
      "grad_norm": 2.5961645722358235,
      "language_loss": 0.72723007,
      "learning_rate": 1.5224741632825888e-06,
      "loss": 0.7513203,
      "num_input_tokens_seen": 105744920,
      "step": 4902,
      "time_per_iteration": 2.961007595062256
    },
    {
      "auxiliary_loss_clip": 0.01184268,
      "auxiliary_loss_mlp": 0.01034613,
      "balance_loss_clip": 1.05353367,
      "balance_loss_mlp": 1.02636361,
      "epoch": 0.589550892803463,
      "flos": 42298890721920.0,
      "grad_norm": 1.9051272877245542,
      "language_loss": 0.69417393,
      "learning_rate": 1.521717757378074e-06,
      "loss": 0.71636283,
      "num_input_tokens_seen": 105765465,
      "step": 4903,
      "time_per_iteration": 3.630335807800293
    },
    {
      "auxiliary_loss_clip": 0.01234591,
      "auxiliary_loss_mlp": 0.01028472,
      "balance_loss_clip": 1.05143094,
      "balance_loss_mlp": 1.02076554,
      "epoch": 0.5896711356941021,
      "flos": 14137062197760.0,
      "grad_norm": 1.7770570064866504,
      "language_loss": 0.69033635,
      "learning_rate": 1.5209614240285943e-06,
      "loss": 0.71296692,
      "num_input_tokens_seen": 105783120,
      "step": 4904,
      "time_per_iteration": 2.5994129180908203
    },
    {
      "auxiliary_loss_clip": 0.0117948,
      "auxiliary_loss_mlp": 0.02566016,
      "balance_loss_clip": 1.05041111,
      "balance_loss_mlp": 1.000054,
      "epoch": 0.5897913785847412,
      "flos": 17201355454080.0,
      "grad_norm": 2.422222914744051,
      "language_loss": 0.84641343,
      "learning_rate": 1.520205163348887e-06,
      "loss": 0.8838684,
      "num_input_tokens_seen": 105801055,
      "step": 4905,
      "time_per_iteration": 2.6055867671966553
    },
    {
      "auxiliary_loss_clip": 0.01233068,
      "auxiliary_loss_mlp": 0.0099815,
      "balance_loss_clip": 1.01135731,
      "balance_loss_mlp": 0.99717247,
      "epoch": 0.5899116214753802,
      "flos": 48794164202880.0,
      "grad_norm": 0.7243282852635303,
      "language_loss": 0.56908643,
      "learning_rate": 1.519448975453674e-06,
      "loss": 0.59139872,
      "num_input_tokens_seen": 105856155,
      "step": 4906,
      "time_per_iteration": 3.127502202987671
    },
    {
      "auxiliary_loss_clip": 0.01230268,
      "auxiliary_loss_mlp": 0.02566224,
      "balance_loss_clip": 1.05317533,
      "balance_loss_mlp": 1.00002551,
      "epoch": 0.5900318643660194,
      "flos": 21103659987840.0,
      "grad_norm": 1.9462892785717094,
      "language_loss": 0.76484525,
      "learning_rate": 1.5186928604576696e-06,
      "loss": 0.80281013,
      "num_input_tokens_seen": 105873350,
      "step": 4907,
      "time_per_iteration": 2.61940336227417
    },
    {
      "auxiliary_loss_clip": 0.01280488,
      "auxiliary_loss_mlp": 0.01031598,
      "balance_loss_clip": 1.04807997,
      "balance_loss_mlp": 1.02454078,
      "epoch": 0.5901521072566585,
      "flos": 21178390233600.0,
      "grad_norm": 2.283521560852937,
      "language_loss": 0.77184325,
      "learning_rate": 1.5179368184755752e-06,
      "loss": 0.79496413,
      "num_input_tokens_seen": 105891435,
      "step": 4908,
      "time_per_iteration": 2.6896493434906006
    },
    {
      "auxiliary_loss_clip": 0.01279083,
      "auxiliary_loss_mlp": 0.01029412,
      "balance_loss_clip": 1.05018151,
      "balance_loss_mlp": 1.02263153,
      "epoch": 0.5902723501472975,
      "flos": 20225967160320.0,
      "grad_norm": 1.7979602650968298,
      "language_loss": 0.82472658,
      "learning_rate": 1.5171808496220821e-06,
      "loss": 0.84781158,
      "num_input_tokens_seen": 105910190,
      "step": 4909,
      "time_per_iteration": 2.6496469974517822
    },
    {
      "auxiliary_loss_clip": 0.01286847,
      "auxiliary_loss_mlp": 0.01024161,
      "balance_loss_clip": 1.05041599,
      "balance_loss_mlp": 1.01701403,
      "epoch": 0.5903925930379367,
      "flos": 22964407211520.0,
      "grad_norm": 1.7614502752277854,
      "language_loss": 0.8129425,
      "learning_rate": 1.5164249540118708e-06,
      "loss": 0.83605254,
      "num_input_tokens_seen": 105929315,
      "step": 4910,
      "time_per_iteration": 2.7177062034606934
    },
    {
      "auxiliary_loss_clip": 0.01430621,
      "auxiliary_loss_mlp": 0.01027805,
      "balance_loss_clip": 1.04378343,
      "balance_loss_mlp": 1.02104568,
      "epoch": 0.5905128359285757,
      "flos": 23367720096000.0,
      "grad_norm": 1.5648577063070823,
      "language_loss": 0.83342493,
      "learning_rate": 1.5156691317596093e-06,
      "loss": 0.85800922,
      "num_input_tokens_seen": 105950740,
      "step": 4911,
      "time_per_iteration": 2.8109965324401855
    },
    {
      "auxiliary_loss_clip": 0.01236536,
      "auxiliary_loss_mlp": 0.02562734,
      "balance_loss_clip": 1.05217075,
      "balance_loss_mlp": 1.0000453,
      "epoch": 0.5906330788192148,
      "flos": 28032335994240.0,
      "grad_norm": 2.896457067939656,
      "language_loss": 0.6688832,
      "learning_rate": 1.5149133829799556e-06,
      "loss": 0.70687592,
      "num_input_tokens_seen": 105968735,
      "step": 4912,
      "time_per_iteration": 2.750356912612915
    },
    {
      "auxiliary_loss_clip": 0.01189123,
      "auxiliary_loss_mlp": 0.01029235,
      "balance_loss_clip": 1.04974449,
      "balance_loss_mlp": 1.02230024,
      "epoch": 0.590753321709854,
      "flos": 18477943793280.0,
      "grad_norm": 1.9356435555514857,
      "language_loss": 0.81350088,
      "learning_rate": 1.5141577077875556e-06,
      "loss": 0.83568448,
      "num_input_tokens_seen": 105986060,
      "step": 4913,
      "time_per_iteration": 2.650779962539673
    },
    {
      "auxiliary_loss_clip": 0.01233878,
      "auxiliary_loss_mlp": 0.01024777,
      "balance_loss_clip": 1.0508728,
      "balance_loss_mlp": 1.01699555,
      "epoch": 0.590873564600493,
      "flos": 16873706568960.0,
      "grad_norm": 2.1738160727642573,
      "language_loss": 0.7233454,
      "learning_rate": 1.5134021062970451e-06,
      "loss": 0.74593198,
      "num_input_tokens_seen": 106004440,
      "step": 4914,
      "time_per_iteration": 2.6439640522003174
    },
    {
      "auxiliary_loss_clip": 0.01326422,
      "auxiliary_loss_mlp": 0.01028952,
      "balance_loss_clip": 1.04918301,
      "balance_loss_mlp": 1.02182615,
      "epoch": 0.5909938074911321,
      "flos": 13516166678400.0,
      "grad_norm": 2.0589331540156492,
      "language_loss": 0.81199527,
      "learning_rate": 1.5126465786230483e-06,
      "loss": 0.83554906,
      "num_input_tokens_seen": 106021215,
      "step": 4915,
      "time_per_iteration": 2.6942994594573975
    },
    {
      "auxiliary_loss_clip": 0.01179533,
      "auxiliary_loss_mlp": 0.01022063,
      "balance_loss_clip": 1.0504967,
      "balance_loss_mlp": 1.01452327,
      "epoch": 0.5911140503817712,
      "flos": 26024067613440.0,
      "grad_norm": 2.481435285146199,
      "language_loss": 0.81960499,
      "learning_rate": 1.5118911248801787e-06,
      "loss": 0.84162098,
      "num_input_tokens_seen": 106039225,
      "step": 4916,
      "time_per_iteration": 2.6322696208953857
    },
    {
      "auxiliary_loss_clip": 0.01222207,
      "auxiliary_loss_mlp": 0.01029187,
      "balance_loss_clip": 1.04722643,
      "balance_loss_mlp": 1.02186716,
      "epoch": 0.5912342932724103,
      "flos": 23258731253760.0,
      "grad_norm": 2.5739457616673587,
      "language_loss": 0.79651517,
      "learning_rate": 1.5111357451830364e-06,
      "loss": 0.81902909,
      "num_input_tokens_seen": 106057920,
      "step": 4917,
      "time_per_iteration": 2.593173027038574
    },
    {
      "auxiliary_loss_clip": 0.01229775,
      "auxiliary_loss_mlp": 0.01023019,
      "balance_loss_clip": 1.04991901,
      "balance_loss_mlp": 1.01624191,
      "epoch": 0.5913545361630493,
      "flos": 19573039687680.0,
      "grad_norm": 3.5723430638978972,
      "language_loss": 0.71113139,
      "learning_rate": 1.5103804396462131e-06,
      "loss": 0.73365933,
      "num_input_tokens_seen": 106077855,
      "step": 4918,
      "time_per_iteration": 2.6730809211730957
    },
    {
      "auxiliary_loss_clip": 0.01234179,
      "auxiliary_loss_mlp": 0.0102702,
      "balance_loss_clip": 1.04944599,
      "balance_loss_mlp": 1.01907432,
      "epoch": 0.5914747790536885,
      "flos": 26213532877440.0,
      "grad_norm": 2.1222232227162876,
      "language_loss": 0.80418116,
      "learning_rate": 1.5096252083842877e-06,
      "loss": 0.82679313,
      "num_input_tokens_seen": 106097065,
      "step": 4919,
      "time_per_iteration": 2.6532180309295654
    },
    {
      "auxiliary_loss_clip": 0.01227002,
      "auxiliary_loss_mlp": 0.01027542,
      "balance_loss_clip": 1.04527259,
      "balance_loss_mlp": 1.02004385,
      "epoch": 0.5915950219443276,
      "flos": 27417545786880.0,
      "grad_norm": 1.7839762179434189,
      "language_loss": 0.85452539,
      "learning_rate": 1.5088700515118285e-06,
      "loss": 0.87707078,
      "num_input_tokens_seen": 106116385,
      "step": 4920,
      "time_per_iteration": 2.6998281478881836
    },
    {
      "auxiliary_loss_clip": 0.0133033,
      "auxiliary_loss_mlp": 0.01026579,
      "balance_loss_clip": 1.04916203,
      "balance_loss_mlp": 1.01966751,
      "epoch": 0.5917152648349666,
      "flos": 21907879545600.0,
      "grad_norm": 1.6475585561793864,
      "language_loss": 0.66462827,
      "learning_rate": 1.508114969143392e-06,
      "loss": 0.68819737,
      "num_input_tokens_seen": 106136370,
      "step": 4921,
      "time_per_iteration": 3.693866014480591
    },
    {
      "auxiliary_loss_clip": 0.01280198,
      "auxiliary_loss_mlp": 0.01024085,
      "balance_loss_clip": 1.04542899,
      "balance_loss_mlp": 1.01710796,
      "epoch": 0.5918355077256057,
      "flos": 28109185142400.0,
      "grad_norm": 1.3789306119883358,
      "language_loss": 0.77831978,
      "learning_rate": 1.5073599613935238e-06,
      "loss": 0.80136263,
      "num_input_tokens_seen": 106158490,
      "step": 4922,
      "time_per_iteration": 3.6270229816436768
    },
    {
      "auxiliary_loss_clip": 0.01280638,
      "auxiliary_loss_mlp": 0.01022469,
      "balance_loss_clip": 1.04827595,
      "balance_loss_mlp": 1.01512909,
      "epoch": 0.5919557506162448,
      "flos": 28183807647360.0,
      "grad_norm": 2.216935361769543,
      "language_loss": 0.57385898,
      "learning_rate": 1.5066050283767574e-06,
      "loss": 0.59689003,
      "num_input_tokens_seen": 106179170,
      "step": 4923,
      "time_per_iteration": 2.720912456512451
    },
    {
      "auxiliary_loss_clip": 0.0117888,
      "auxiliary_loss_mlp": 0.01022965,
      "balance_loss_clip": 1.04827142,
      "balance_loss_mlp": 1.01576138,
      "epoch": 0.5920759935068839,
      "flos": 12094355652480.0,
      "grad_norm": 1.9925239205227272,
      "language_loss": 0.83073866,
      "learning_rate": 1.505850170207616e-06,
      "loss": 0.85275716,
      "num_input_tokens_seen": 106196035,
      "step": 4924,
      "time_per_iteration": 3.55501127243042
    },
    {
      "auxiliary_loss_clip": 0.01275836,
      "auxiliary_loss_mlp": 0.0102524,
      "balance_loss_clip": 1.04557347,
      "balance_loss_mlp": 1.01783442,
      "epoch": 0.592196236397523,
      "flos": 29424772673280.0,
      "grad_norm": 2.1492739769620712,
      "language_loss": 0.78149772,
      "learning_rate": 1.505095387000611e-06,
      "loss": 0.80450857,
      "num_input_tokens_seen": 106218335,
      "step": 4925,
      "time_per_iteration": 2.7500159740448
    },
    {
      "auxiliary_loss_clip": 0.01273896,
      "auxiliary_loss_mlp": 0.01028995,
      "balance_loss_clip": 1.04765463,
      "balance_loss_mlp": 1.02217662,
      "epoch": 0.5923164792881621,
      "flos": 24384709866240.0,
      "grad_norm": 2.1627234024104367,
      "language_loss": 0.74355823,
      "learning_rate": 1.504340678870242e-06,
      "loss": 0.76658714,
      "num_input_tokens_seen": 106236550,
      "step": 4926,
      "time_per_iteration": 2.7132694721221924
    },
    {
      "auxiliary_loss_clip": 0.01228207,
      "auxiliary_loss_mlp": 0.01028087,
      "balance_loss_clip": 1.04966593,
      "balance_loss_mlp": 1.02050519,
      "epoch": 0.5924367221788012,
      "flos": 24024238928640.0,
      "grad_norm": 2.0879309541886797,
      "language_loss": 0.89954334,
      "learning_rate": 1.5035860459309989e-06,
      "loss": 0.92210627,
      "num_input_tokens_seen": 106254265,
      "step": 4927,
      "time_per_iteration": 2.686352014541626
    },
    {
      "auxiliary_loss_clip": 0.01277956,
      "auxiliary_loss_mlp": 0.01024063,
      "balance_loss_clip": 1.04862285,
      "balance_loss_mlp": 1.01664853,
      "epoch": 0.5925569650694402,
      "flos": 26870590414080.0,
      "grad_norm": 1.8361767061074357,
      "language_loss": 0.63782811,
      "learning_rate": 1.5028314882973568e-06,
      "loss": 0.66084832,
      "num_input_tokens_seen": 106274670,
      "step": 4928,
      "time_per_iteration": 2.709462881088257
    },
    {
      "auxiliary_loss_clip": 0.0128094,
      "auxiliary_loss_mlp": 0.01027338,
      "balance_loss_clip": 1.0498302,
      "balance_loss_mlp": 1.02029312,
      "epoch": 0.5926772079600794,
      "flos": 22302788647680.0,
      "grad_norm": 1.8701327108025374,
      "language_loss": 0.84340405,
      "learning_rate": 1.502077006083783e-06,
      "loss": 0.86648685,
      "num_input_tokens_seen": 106293330,
      "step": 4929,
      "time_per_iteration": 3.572216749191284
    },
    {
      "auxiliary_loss_clip": 0.01134519,
      "auxiliary_loss_mlp": 0.0256223,
      "balance_loss_clip": 1.05126119,
      "balance_loss_mlp": 1.00005686,
      "epoch": 0.5927974508507184,
      "flos": 19865244827520.0,
      "grad_norm": 2.116306467028422,
      "language_loss": 0.7642563,
      "learning_rate": 1.5013225994047315e-06,
      "loss": 0.80122375,
      "num_input_tokens_seen": 106310960,
      "step": 4930,
      "time_per_iteration": 2.629594564437866
    },
    {
      "auxiliary_loss_clip": 0.01232615,
      "auxiliary_loss_mlp": 0.02564226,
      "balance_loss_clip": 1.05161572,
      "balance_loss_mlp": 1.00005913,
      "epoch": 0.5929176937413575,
      "flos": 15776743167360.0,
      "grad_norm": 1.8486755150075518,
      "language_loss": 0.80764556,
      "learning_rate": 1.5005682683746452e-06,
      "loss": 0.84561396,
      "num_input_tokens_seen": 106329475,
      "step": 4931,
      "time_per_iteration": 2.6687803268432617
    },
    {
      "auxiliary_loss_clip": 0.01232639,
      "auxiliary_loss_mlp": 0.01027521,
      "balance_loss_clip": 1.05399752,
      "balance_loss_mlp": 1.02032053,
      "epoch": 0.5930379366319967,
      "flos": 17601472028160.0,
      "grad_norm": 2.4326342815862176,
      "language_loss": 0.72541404,
      "learning_rate": 1.4998140131079553e-06,
      "loss": 0.74801564,
      "num_input_tokens_seen": 106345565,
      "step": 4932,
      "time_per_iteration": 2.6272292137145996
    },
    {
      "auxiliary_loss_clip": 0.01422108,
      "auxiliary_loss_mlp": 0.02561526,
      "balance_loss_clip": 1.04321837,
      "balance_loss_mlp": 1.00011218,
      "epoch": 0.5931581795226357,
      "flos": 17704283731200.0,
      "grad_norm": 3.7415046094473148,
      "language_loss": 0.73167503,
      "learning_rate": 1.4990598337190821e-06,
      "loss": 0.77151144,
      "num_input_tokens_seen": 106361920,
      "step": 4933,
      "time_per_iteration": 2.8123438358306885
    },
    {
      "auxiliary_loss_clip": 0.01181622,
      "auxiliary_loss_mlp": 0.02564264,
      "balance_loss_clip": 1.05157948,
      "balance_loss_mlp": 1.00007033,
      "epoch": 0.5932784224132748,
      "flos": 24280102483200.0,
      "grad_norm": 1.698412608235749,
      "language_loss": 0.68097746,
      "learning_rate": 1.4983057303224338e-06,
      "loss": 0.71843624,
      "num_input_tokens_seen": 106381735,
      "step": 4934,
      "time_per_iteration": 2.6316516399383545
    },
    {
      "auxiliary_loss_clip": 0.01376666,
      "auxiliary_loss_mlp": 0.01035017,
      "balance_loss_clip": 1.04480839,
      "balance_loss_mlp": 1.02782857,
      "epoch": 0.5933986653039139,
      "flos": 22926700909440.0,
      "grad_norm": 2.1688876545382545,
      "language_loss": 0.87645298,
      "learning_rate": 1.4975517030324072e-06,
      "loss": 0.90056986,
      "num_input_tokens_seen": 106399745,
      "step": 4935,
      "time_per_iteration": 2.796238660812378
    },
    {
      "auxiliary_loss_clip": 0.01068339,
      "auxiliary_loss_mlp": 0.02507365,
      "balance_loss_clip": 1.01116765,
      "balance_loss_mlp": 1.0001229,
      "epoch": 0.593518908194553,
      "flos": 71121730256640.0,
      "grad_norm": 0.7812739531134544,
      "language_loss": 0.6176458,
      "learning_rate": 1.4967977519633882e-06,
      "loss": 0.65340275,
      "num_input_tokens_seen": 106457205,
      "step": 4936,
      "time_per_iteration": 3.28973388671875
    },
    {
      "auxiliary_loss_clip": 0.01321913,
      "auxiliary_loss_mlp": 0.01024754,
      "balance_loss_clip": 1.0449307,
      "balance_loss_mlp": 1.01725578,
      "epoch": 0.593639151085192,
      "flos": 20448649526400.0,
      "grad_norm": 2.2282156462265257,
      "language_loss": 0.78465092,
      "learning_rate": 1.4960438772297494e-06,
      "loss": 0.80811757,
      "num_input_tokens_seen": 106474250,
      "step": 4937,
      "time_per_iteration": 2.7128188610076904
    },
    {
      "auxiliary_loss_clip": 0.01280302,
      "auxiliary_loss_mlp": 0.01024261,
      "balance_loss_clip": 1.04538846,
      "balance_loss_mlp": 1.01692653,
      "epoch": 0.5937593939758312,
      "flos": 30883428074880.0,
      "grad_norm": 2.431686285218414,
      "language_loss": 0.73852468,
      "learning_rate": 1.495290078945855e-06,
      "loss": 0.76157033,
      "num_input_tokens_seen": 106494015,
      "step": 4938,
      "time_per_iteration": 2.7139196395874023
    },
    {
      "auxiliary_loss_clip": 0.01179208,
      "auxiliary_loss_mlp": 0.01027898,
      "balance_loss_clip": 1.05173886,
      "balance_loss_mlp": 1.02082944,
      "epoch": 0.5938796368664703,
      "flos": 36898069668480.0,
      "grad_norm": 2.295080198097253,
      "language_loss": 0.74331701,
      "learning_rate": 1.4945363572260529e-06,
      "loss": 0.76538807,
      "num_input_tokens_seen": 106515010,
      "step": 4939,
      "time_per_iteration": 2.6969687938690186
    },
    {
      "auxiliary_loss_clip": 0.01225013,
      "auxiliary_loss_mlp": 0.01028041,
      "balance_loss_clip": 1.04714251,
      "balance_loss_mlp": 1.02104318,
      "epoch": 0.5939998797571093,
      "flos": 23842926051840.0,
      "grad_norm": 2.314074953676868,
      "language_loss": 0.68264091,
      "learning_rate": 1.4937827121846845e-06,
      "loss": 0.70517147,
      "num_input_tokens_seen": 106535265,
      "step": 4940,
      "time_per_iteration": 2.669736862182617
    },
    {
      "auxiliary_loss_clip": 0.01328956,
      "auxiliary_loss_mlp": 0.01025839,
      "balance_loss_clip": 1.0499835,
      "balance_loss_mlp": 1.01930904,
      "epoch": 0.5941201226477485,
      "flos": 25191407462400.0,
      "grad_norm": 1.5387300989986363,
      "language_loss": 0.73549068,
      "learning_rate": 1.4930291439360755e-06,
      "loss": 0.75903863,
      "num_input_tokens_seen": 106557830,
      "step": 4941,
      "time_per_iteration": 2.697032928466797
    },
    {
      "auxiliary_loss_clip": 0.01229711,
      "auxiliary_loss_mlp": 0.01034052,
      "balance_loss_clip": 1.05002201,
      "balance_loss_mlp": 1.02716768,
      "epoch": 0.5942403655383875,
      "flos": 22418996123520.0,
      "grad_norm": 2.396466780630469,
      "language_loss": 0.79355788,
      "learning_rate": 1.4922756525945427e-06,
      "loss": 0.81619549,
      "num_input_tokens_seen": 106577140,
      "step": 4942,
      "time_per_iteration": 2.717913866043091
    },
    {
      "auxiliary_loss_clip": 0.01022486,
      "auxiliary_loss_mlp": 0.01006837,
      "balance_loss_clip": 1.01134241,
      "balance_loss_mlp": 1.0058955,
      "epoch": 0.5943606084290266,
      "flos": 67629310796160.0,
      "grad_norm": 0.7717893035362148,
      "language_loss": 0.59478658,
      "learning_rate": 1.4915222382743894e-06,
      "loss": 0.61507982,
      "num_input_tokens_seen": 106635975,
      "step": 4943,
      "time_per_iteration": 3.200469732284546
    },
    {
      "auxiliary_loss_clip": 0.01229933,
      "auxiliary_loss_mlp": 0.01024695,
      "balance_loss_clip": 1.05160725,
      "balance_loss_mlp": 1.01725674,
      "epoch": 0.5944808513196658,
      "flos": 18223157646720.0,
      "grad_norm": 2.1842079324228143,
      "language_loss": 0.72307658,
      "learning_rate": 1.4907689010899085e-06,
      "loss": 0.74562287,
      "num_input_tokens_seen": 106653555,
      "step": 4944,
      "time_per_iteration": 2.614574909210205
    },
    {
      "auxiliary_loss_clip": 0.01276664,
      "auxiliary_loss_mlp": 0.01029601,
      "balance_loss_clip": 1.04832339,
      "balance_loss_mlp": 1.02284801,
      "epoch": 0.5946010942103048,
      "flos": 24790824011520.0,
      "grad_norm": 1.969972762925356,
      "language_loss": 0.62712991,
      "learning_rate": 1.4900156411553804e-06,
      "loss": 0.65019262,
      "num_input_tokens_seen": 106673385,
      "step": 4945,
      "time_per_iteration": 2.711216926574707
    },
    {
      "auxiliary_loss_clip": 0.0127951,
      "auxiliary_loss_mlp": 0.01030182,
      "balance_loss_clip": 1.04968905,
      "balance_loss_mlp": 1.02244568,
      "epoch": 0.5947213371009439,
      "flos": 15231619388160.0,
      "grad_norm": 1.9708919071621118,
      "language_loss": 0.85604429,
      "learning_rate": 1.4892624585850739e-06,
      "loss": 0.87914127,
      "num_input_tokens_seen": 106691740,
      "step": 4946,
      "time_per_iteration": 3.6050212383270264
    },
    {
      "auxiliary_loss_clip": 0.01180981,
      "auxiliary_loss_mlp": 0.01031775,
      "balance_loss_clip": 1.05148673,
      "balance_loss_mlp": 1.02473593,
      "epoch": 0.594841579991583,
      "flos": 25848069949440.0,
      "grad_norm": 2.6010535308205065,
      "language_loss": 0.79724896,
      "learning_rate": 1.4885093534932465e-06,
      "loss": 0.81937647,
      "num_input_tokens_seen": 106709705,
      "step": 4947,
      "time_per_iteration": 2.6808226108551025
    },
    {
      "auxiliary_loss_clip": 0.01275019,
      "auxiliary_loss_mlp": 0.01027664,
      "balance_loss_clip": 1.04973412,
      "balance_loss_mlp": 1.02011502,
      "epoch": 0.5949618228822221,
      "flos": 23981109672960.0,
      "grad_norm": 2.1904105338403363,
      "language_loss": 0.7145583,
      "learning_rate": 1.4877563259941433e-06,
      "loss": 0.73758513,
      "num_input_tokens_seen": 106727560,
      "step": 4948,
      "time_per_iteration": 2.599902391433716
    },
    {
      "auxiliary_loss_clip": 0.01236247,
      "auxiliary_loss_mlp": 0.01024621,
      "balance_loss_clip": 1.05095506,
      "balance_loss_mlp": 1.01718187,
      "epoch": 0.5950820657728612,
      "flos": 40547491476480.0,
      "grad_norm": 1.987947884797166,
      "language_loss": 0.67675436,
      "learning_rate": 1.4870033762019988e-06,
      "loss": 0.69936305,
      "num_input_tokens_seen": 106747725,
      "step": 4949,
      "time_per_iteration": 4.535677433013916
    },
    {
      "auxiliary_loss_clip": 0.01280518,
      "auxiliary_loss_mlp": 0.01029172,
      "balance_loss_clip": 1.0479027,
      "balance_loss_mlp": 1.02206421,
      "epoch": 0.5952023086635003,
      "flos": 23184467884800.0,
      "grad_norm": 2.4506688107665737,
      "language_loss": 0.73587537,
      "learning_rate": 1.4862505042310334e-06,
      "loss": 0.75897229,
      "num_input_tokens_seen": 106767010,
      "step": 4950,
      "time_per_iteration": 2.679192304611206
    },
    {
      "auxiliary_loss_clip": 0.01271943,
      "auxiliary_loss_mlp": 0.01026736,
      "balance_loss_clip": 1.04805958,
      "balance_loss_mlp": 1.01958382,
      "epoch": 0.5953225515541394,
      "flos": 33653289548160.0,
      "grad_norm": 2.5377553200190257,
      "language_loss": 0.69745159,
      "learning_rate": 1.4854977101954587e-06,
      "loss": 0.72043836,
      "num_input_tokens_seen": 106789230,
      "step": 4951,
      "time_per_iteration": 2.7574377059936523
    },
    {
      "auxiliary_loss_clip": 0.01229963,
      "auxiliary_loss_mlp": 0.01024066,
      "balance_loss_clip": 1.04714119,
      "balance_loss_mlp": 1.01676154,
      "epoch": 0.5954427944447784,
      "flos": 24459619680000.0,
      "grad_norm": 1.968153578359552,
      "language_loss": 0.86315805,
      "learning_rate": 1.4847449942094716e-06,
      "loss": 0.88569832,
      "num_input_tokens_seen": 106808110,
      "step": 4952,
      "time_per_iteration": 2.681190252304077
    },
    {
      "auxiliary_loss_clip": 0.01272038,
      "auxiliary_loss_mlp": 0.01026136,
      "balance_loss_clip": 1.04723287,
      "balance_loss_mlp": 1.01868224,
      "epoch": 0.5955630373354175,
      "flos": 18551848026240.0,
      "grad_norm": 2.192264759373798,
      "language_loss": 0.85907388,
      "learning_rate": 1.4839923563872598e-06,
      "loss": 0.88205558,
      "num_input_tokens_seen": 106826650,
      "step": 4953,
      "time_per_iteration": 2.62654185295105
    },
    {
      "auxiliary_loss_clip": 0.01335185,
      "auxiliary_loss_mlp": 0.01028863,
      "balance_loss_clip": 1.05126452,
      "balance_loss_mlp": 1.02190685,
      "epoch": 0.5956832802260567,
      "flos": 19791699730560.0,
      "grad_norm": 1.763263001429737,
      "language_loss": 0.75947183,
      "learning_rate": 1.483239796842997e-06,
      "loss": 0.78311229,
      "num_input_tokens_seen": 106844680,
      "step": 4954,
      "time_per_iteration": 3.682220220565796
    },
    {
      "auxiliary_loss_clip": 0.01226676,
      "auxiliary_loss_mlp": 0.01026003,
      "balance_loss_clip": 1.04913998,
      "balance_loss_mlp": 1.01922297,
      "epoch": 0.5958035231166957,
      "flos": 19750868945280.0,
      "grad_norm": 1.841145731353735,
      "language_loss": 0.83943522,
      "learning_rate": 1.4824873156908462e-06,
      "loss": 0.86196202,
      "num_input_tokens_seen": 106862605,
      "step": 4955,
      "time_per_iteration": 2.71268892288208
    },
    {
      "auxiliary_loss_clip": 0.0123167,
      "auxiliary_loss_mlp": 0.02570765,
      "balance_loss_clip": 1.05208731,
      "balance_loss_mlp": 1.00001335,
      "epoch": 0.5959237660073348,
      "flos": 21652806090240.0,
      "grad_norm": 1.573787749032818,
      "language_loss": 0.75697714,
      "learning_rate": 1.4817349130449584e-06,
      "loss": 0.79500151,
      "num_input_tokens_seen": 106882325,
      "step": 4956,
      "time_per_iteration": 2.673875570297241
    },
    {
      "auxiliary_loss_clip": 0.01227735,
      "auxiliary_loss_mlp": 0.01023402,
      "balance_loss_clip": 1.05040646,
      "balance_loss_mlp": 1.01621628,
      "epoch": 0.5960440088979739,
      "flos": 21171207513600.0,
      "grad_norm": 1.8655585309736469,
      "language_loss": 0.83121669,
      "learning_rate": 1.4809825890194717e-06,
      "loss": 0.85372806,
      "num_input_tokens_seen": 106900995,
      "step": 4957,
      "time_per_iteration": 2.6471030712127686
    },
    {
      "auxiliary_loss_clip": 0.01276872,
      "auxiliary_loss_mlp": 0.01029128,
      "balance_loss_clip": 1.0446744,
      "balance_loss_mlp": 1.02205312,
      "epoch": 0.596164251788613,
      "flos": 14757526753920.0,
      "grad_norm": 1.8709159935305104,
      "language_loss": 0.77381819,
      "learning_rate": 1.4802303437285139e-06,
      "loss": 0.79687822,
      "num_input_tokens_seen": 106918265,
      "step": 4958,
      "time_per_iteration": 2.648054599761963
    },
    {
      "auxiliary_loss_clip": 0.01277616,
      "auxiliary_loss_mlp": 0.01026358,
      "balance_loss_clip": 1.04781818,
      "balance_loss_mlp": 1.01897252,
      "epoch": 0.596284494679252,
      "flos": 20485924865280.0,
      "grad_norm": 2.28495750844855,
      "language_loss": 0.80980414,
      "learning_rate": 1.4794781772861994e-06,
      "loss": 0.8328439,
      "num_input_tokens_seen": 106934760,
      "step": 4959,
      "time_per_iteration": 2.6609606742858887
    },
    {
      "auxiliary_loss_clip": 0.01278991,
      "auxiliary_loss_mlp": 0.02563107,
      "balance_loss_clip": 1.04830647,
      "balance_loss_mlp": 1.00001359,
      "epoch": 0.5964047375698912,
      "flos": 31212262108800.0,
      "grad_norm": 2.085054705174388,
      "language_loss": 0.67044973,
      "learning_rate": 1.4787260898066324e-06,
      "loss": 0.70887071,
      "num_input_tokens_seen": 106954760,
      "step": 4960,
      "time_per_iteration": 2.788867712020874
    },
    {
      "auxiliary_loss_clip": 0.01179637,
      "auxiliary_loss_mlp": 0.01024997,
      "balance_loss_clip": 1.05265188,
      "balance_loss_mlp": 1.0181334,
      "epoch": 0.5965249804605303,
      "flos": 27483620855040.0,
      "grad_norm": 2.4366707052510197,
      "language_loss": 0.85894567,
      "learning_rate": 1.4779740814039023e-06,
      "loss": 0.880992,
      "num_input_tokens_seen": 106974845,
      "step": 4961,
      "time_per_iteration": 2.6918458938598633
    },
    {
      "auxiliary_loss_clip": 0.01179255,
      "auxiliary_loss_mlp": 0.01025572,
      "balance_loss_clip": 1.05098319,
      "balance_loss_mlp": 1.01864624,
      "epoch": 0.5966452233511693,
      "flos": 30773936442240.0,
      "grad_norm": 3.53947034211198,
      "language_loss": 0.68515122,
      "learning_rate": 1.4772221521920894e-06,
      "loss": 0.70719957,
      "num_input_tokens_seen": 106994870,
      "step": 4962,
      "time_per_iteration": 2.6488308906555176
    },
    {
      "auxiliary_loss_clip": 0.01285044,
      "auxiliary_loss_mlp": 0.01029597,
      "balance_loss_clip": 1.05251014,
      "balance_loss_mlp": 1.0223285,
      "epoch": 0.5967654662418085,
      "flos": 25481170477440.0,
      "grad_norm": 2.422183017124287,
      "language_loss": 0.74271607,
      "learning_rate": 1.4764703022852598e-06,
      "loss": 0.76586246,
      "num_input_tokens_seen": 107015390,
      "step": 4963,
      "time_per_iteration": 2.695699453353882
    },
    {
      "auxiliary_loss_clip": 0.01415664,
      "auxiliary_loss_mlp": 0.01026249,
      "balance_loss_clip": 1.04093313,
      "balance_loss_mlp": 1.01929927,
      "epoch": 0.5968857091324475,
      "flos": 19099126621440.0,
      "grad_norm": 1.862581002981766,
      "language_loss": 0.77134001,
      "learning_rate": 1.4757185317974696e-06,
      "loss": 0.7957592,
      "num_input_tokens_seen": 107033775,
      "step": 4964,
      "time_per_iteration": 2.748850107192993
    },
    {
      "auxiliary_loss_clip": 0.01229159,
      "auxiliary_loss_mlp": 0.01027246,
      "balance_loss_clip": 1.04876649,
      "balance_loss_mlp": 1.01934862,
      "epoch": 0.5970059520230866,
      "flos": 23692711374720.0,
      "grad_norm": 2.4984421177537945,
      "language_loss": 0.70798975,
      "learning_rate": 1.474966840842761e-06,
      "loss": 0.73055375,
      "num_input_tokens_seen": 107053355,
      "step": 4965,
      "time_per_iteration": 2.6595513820648193
    },
    {
      "auxiliary_loss_clip": 0.01231776,
      "auxiliary_loss_mlp": 0.01024328,
      "balance_loss_clip": 1.04913902,
      "balance_loss_mlp": 1.01746452,
      "epoch": 0.5971261949137258,
      "flos": 23185545292800.0,
      "grad_norm": 1.8099492978715312,
      "language_loss": 0.87059081,
      "learning_rate": 1.4742152295351655e-06,
      "loss": 0.89315182,
      "num_input_tokens_seen": 107072510,
      "step": 4966,
      "time_per_iteration": 2.647264242172241
    },
    {
      "auxiliary_loss_clip": 0.0123178,
      "auxiliary_loss_mlp": 0.02568442,
      "balance_loss_clip": 1.05082095,
      "balance_loss_mlp": 1.00004745,
      "epoch": 0.5972464378043648,
      "flos": 20557710195840.0,
      "grad_norm": 2.5619880514731483,
      "language_loss": 0.64141893,
      "learning_rate": 1.4734636979887016e-06,
      "loss": 0.67942119,
      "num_input_tokens_seen": 107089970,
      "step": 4967,
      "time_per_iteration": 2.6911959648132324
    },
    {
      "auxiliary_loss_clip": 0.01332021,
      "auxiliary_loss_mlp": 0.01026602,
      "balance_loss_clip": 1.0457437,
      "balance_loss_mlp": 1.01883531,
      "epoch": 0.5973666806950039,
      "flos": 29387030457600.0,
      "grad_norm": 2.4279775343243575,
      "language_loss": 0.90534782,
      "learning_rate": 1.4727122463173755e-06,
      "loss": 0.9289341,
      "num_input_tokens_seen": 107108500,
      "step": 4968,
      "time_per_iteration": 2.7546751499176025
    },
    {
      "auxiliary_loss_clip": 0.01276981,
      "auxiliary_loss_mlp": 0.01027149,
      "balance_loss_clip": 1.05022955,
      "balance_loss_mlp": 1.02009773,
      "epoch": 0.597486923585643,
      "flos": 22273522041600.0,
      "grad_norm": 1.7783423812992145,
      "language_loss": 0.6413914,
      "learning_rate": 1.471960874635183e-06,
      "loss": 0.6644327,
      "num_input_tokens_seen": 107128060,
      "step": 4969,
      "time_per_iteration": 2.626188039779663
    },
    {
      "auxiliary_loss_clip": 0.01272129,
      "auxiliary_loss_mlp": 0.01029794,
      "balance_loss_clip": 1.04549348,
      "balance_loss_mlp": 1.02202201,
      "epoch": 0.5976071664762821,
      "flos": 13772461196160.0,
      "grad_norm": 2.316053087415415,
      "language_loss": 0.70836508,
      "learning_rate": 1.4712095830561055e-06,
      "loss": 0.7313844,
      "num_input_tokens_seen": 107146550,
      "step": 4970,
      "time_per_iteration": 2.651453733444214
    },
    {
      "auxiliary_loss_clip": 0.01279566,
      "auxiliary_loss_mlp": 0.01026394,
      "balance_loss_clip": 1.04671311,
      "balance_loss_mlp": 1.01960528,
      "epoch": 0.5977274093669211,
      "flos": 19098623831040.0,
      "grad_norm": 5.382848437052313,
      "language_loss": 0.8113845,
      "learning_rate": 1.4704583716941147e-06,
      "loss": 0.83444411,
      "num_input_tokens_seen": 107165415,
      "step": 4971,
      "time_per_iteration": 2.666487693786621
    },
    {
      "auxiliary_loss_clip": 0.01226824,
      "auxiliary_loss_mlp": 0.01029582,
      "balance_loss_clip": 1.05078816,
      "balance_loss_mlp": 1.02264738,
      "epoch": 0.5978476522575603,
      "flos": 20376002269440.0,
      "grad_norm": 2.0551511584108977,
      "language_loss": 0.72301042,
      "learning_rate": 1.4697072406631672e-06,
      "loss": 0.74557459,
      "num_input_tokens_seen": 107185320,
      "step": 4972,
      "time_per_iteration": 2.7093098163604736
    },
    {
      "auxiliary_loss_clip": 0.01381805,
      "auxiliary_loss_mlp": 0.01028587,
      "balance_loss_clip": 1.04878592,
      "balance_loss_mlp": 1.02161908,
      "epoch": 0.5979678951481994,
      "flos": 29023147728000.0,
      "grad_norm": 1.8861516180703375,
      "language_loss": 0.73140049,
      "learning_rate": 1.4689561900772097e-06,
      "loss": 0.75550437,
      "num_input_tokens_seen": 107205380,
      "step": 4973,
      "time_per_iteration": 3.739610195159912
    },
    {
      "auxiliary_loss_clip": 0.01274923,
      "auxiliary_loss_mlp": 0.01027048,
      "balance_loss_clip": 1.04625249,
      "balance_loss_mlp": 1.01986313,
      "epoch": 0.5980881380388384,
      "flos": 17967689141760.0,
      "grad_norm": 2.621648589853824,
      "language_loss": 0.72831249,
      "learning_rate": 1.4682052200501758e-06,
      "loss": 0.75133222,
      "num_input_tokens_seen": 107222585,
      "step": 4974,
      "time_per_iteration": 3.6090829372406006
    },
    {
      "auxiliary_loss_clip": 0.01180588,
      "auxiliary_loss_mlp": 0.01024625,
      "balance_loss_clip": 1.05198085,
      "balance_loss_mlp": 1.01771092,
      "epoch": 0.5982083809294776,
      "flos": 22962827013120.0,
      "grad_norm": 1.8607055152965746,
      "language_loss": 0.80234241,
      "learning_rate": 1.4674543306959876e-06,
      "loss": 0.82439452,
      "num_input_tokens_seen": 107242055,
      "step": 4975,
      "time_per_iteration": 3.5440964698791504
    },
    {
      "auxiliary_loss_clip": 0.01285404,
      "auxiliary_loss_mlp": 0.01034873,
      "balance_loss_clip": 1.04918861,
      "balance_loss_mlp": 1.02723789,
      "epoch": 0.5983286238201166,
      "flos": 20991941712000.0,
      "grad_norm": 2.554889394101337,
      "language_loss": 0.84562808,
      "learning_rate": 1.4667035221285535e-06,
      "loss": 0.8688308,
      "num_input_tokens_seen": 107259695,
      "step": 4976,
      "time_per_iteration": 2.6623101234436035
    },
    {
      "auxiliary_loss_clip": 0.01226586,
      "auxiliary_loss_mlp": 0.01021299,
      "balance_loss_clip": 1.05080211,
      "balance_loss_mlp": 1.01470125,
      "epoch": 0.5984488667107557,
      "flos": 28183448511360.0,
      "grad_norm": 2.5621422620279986,
      "language_loss": 0.74745607,
      "learning_rate": 1.4659527944617715e-06,
      "loss": 0.76993489,
      "num_input_tokens_seen": 107279640,
      "step": 4977,
      "time_per_iteration": 2.7912042140960693
    },
    {
      "auxiliary_loss_clip": 0.01422003,
      "auxiliary_loss_mlp": 0.01027565,
      "balance_loss_clip": 1.0410049,
      "balance_loss_mlp": 1.02078474,
      "epoch": 0.5985691096013949,
      "flos": 16471794314880.0,
      "grad_norm": 1.870878767918522,
      "language_loss": 0.76324403,
      "learning_rate": 1.465202147809526e-06,
      "loss": 0.78773975,
      "num_input_tokens_seen": 107298135,
      "step": 4978,
      "time_per_iteration": 2.7675669193267822
    },
    {
      "auxiliary_loss_clip": 0.01179147,
      "auxiliary_loss_mlp": 0.01023689,
      "balance_loss_clip": 1.05019879,
      "balance_loss_mlp": 1.01705217,
      "epoch": 0.5986893524920339,
      "flos": 26719046933760.0,
      "grad_norm": 2.061824626798423,
      "language_loss": 0.76322949,
      "learning_rate": 1.4644515822856888e-06,
      "loss": 0.78525788,
      "num_input_tokens_seen": 107316570,
      "step": 4979,
      "time_per_iteration": 2.670954465866089
    },
    {
      "auxiliary_loss_clip": 0.01226338,
      "auxiliary_loss_mlp": 0.01006214,
      "balance_loss_clip": 1.01022124,
      "balance_loss_mlp": 1.00528419,
      "epoch": 0.598809595382673,
      "flos": 61608061100160.0,
      "grad_norm": 0.7565022048817791,
      "language_loss": 0.5650444,
      "learning_rate": 1.4637010980041215e-06,
      "loss": 0.58736992,
      "num_input_tokens_seen": 107378680,
      "step": 4980,
      "time_per_iteration": 3.280036449432373
    },
    {
      "auxiliary_loss_clip": 0.01178606,
      "auxiliary_loss_mlp": 0.01029672,
      "balance_loss_clip": 1.04910588,
      "balance_loss_mlp": 1.02198017,
      "epoch": 0.5989298382733121,
      "flos": 11801719549440.0,
      "grad_norm": 2.170206332495691,
      "language_loss": 0.8988874,
      "learning_rate": 1.4629506950786707e-06,
      "loss": 0.9209702,
      "num_input_tokens_seen": 107394860,
      "step": 4981,
      "time_per_iteration": 3.511972188949585
    },
    {
      "auxiliary_loss_clip": 0.01067739,
      "auxiliary_loss_mlp": 0.01003679,
      "balance_loss_clip": 1.01057708,
      "balance_loss_mlp": 1.002702,
      "epoch": 0.5990500811639512,
      "flos": 60025800021120.0,
      "grad_norm": 0.8116728744637332,
      "language_loss": 0.56057322,
      "learning_rate": 1.4622003736231733e-06,
      "loss": 0.58128738,
      "num_input_tokens_seen": 107453850,
      "step": 4982,
      "time_per_iteration": 3.2264480590820312
    },
    {
      "auxiliary_loss_clip": 0.0123592,
      "auxiliary_loss_mlp": 0.01029301,
      "balance_loss_clip": 1.05350912,
      "balance_loss_mlp": 1.02164733,
      "epoch": 0.5991703240545903,
      "flos": 18222726683520.0,
      "grad_norm": 3.2796118829485557,
      "language_loss": 0.80513275,
      "learning_rate": 1.461450133751451e-06,
      "loss": 0.8277849,
      "num_input_tokens_seen": 107471920,
      "step": 4983,
      "time_per_iteration": 2.674395799636841
    },
    {
      "auxiliary_loss_clip": 0.0113435,
      "auxiliary_loss_mlp": 0.01024964,
      "balance_loss_clip": 1.05054021,
      "balance_loss_mlp": 1.01798987,
      "epoch": 0.5992905669452293,
      "flos": 27709894581120.0,
      "grad_norm": 1.778179844774408,
      "language_loss": 0.7619313,
      "learning_rate": 1.4606999755773153e-06,
      "loss": 0.78352439,
      "num_input_tokens_seen": 107493125,
      "step": 4984,
      "time_per_iteration": 2.6875221729278564
    },
    {
      "auxiliary_loss_clip": 0.01179337,
      "auxiliary_loss_mlp": 0.01027663,
      "balance_loss_clip": 1.05208433,
      "balance_loss_mlp": 1.02084744,
      "epoch": 0.5994108098358685,
      "flos": 20449008662400.0,
      "grad_norm": 1.6551710610030634,
      "language_loss": 0.82652843,
      "learning_rate": 1.4599498992145643e-06,
      "loss": 0.84859848,
      "num_input_tokens_seen": 107513150,
      "step": 4985,
      "time_per_iteration": 2.6555368900299072
    },
    {
      "auxiliary_loss_clip": 0.01284935,
      "auxiliary_loss_mlp": 0.02565454,
      "balance_loss_clip": 1.04961622,
      "balance_loss_mlp": 1.00005198,
      "epoch": 0.5995310527265075,
      "flos": 22269966595200.0,
      "grad_norm": 1.8889249690480252,
      "language_loss": 0.7107867,
      "learning_rate": 1.4591999047769846e-06,
      "loss": 0.74929059,
      "num_input_tokens_seen": 107532005,
      "step": 4986,
      "time_per_iteration": 2.6539177894592285
    },
    {
      "auxiliary_loss_clip": 0.01426845,
      "auxiliary_loss_mlp": 0.01034477,
      "balance_loss_clip": 1.04193783,
      "balance_loss_mlp": 1.02708638,
      "epoch": 0.5996512956171466,
      "flos": 18916951818240.0,
      "grad_norm": 5.608182133896914,
      "language_loss": 0.75120193,
      "learning_rate": 1.4584499923783486e-06,
      "loss": 0.77581513,
      "num_input_tokens_seen": 107550585,
      "step": 4987,
      "time_per_iteration": 2.7686898708343506
    },
    {
      "auxiliary_loss_clip": 0.01278591,
      "auxiliary_loss_mlp": 0.0102605,
      "balance_loss_clip": 1.04977345,
      "balance_loss_mlp": 1.0191896,
      "epoch": 0.5997715385077858,
      "flos": 15370916330880.0,
      "grad_norm": 2.630237564523838,
      "language_loss": 0.76292837,
      "learning_rate": 1.457700162132419e-06,
      "loss": 0.78597474,
      "num_input_tokens_seen": 107567575,
      "step": 4988,
      "time_per_iteration": 2.6642487049102783
    },
    {
      "auxiliary_loss_clip": 0.01374981,
      "auxiliary_loss_mlp": 0.01026522,
      "balance_loss_clip": 1.04688025,
      "balance_loss_mlp": 1.01962829,
      "epoch": 0.5998917813984248,
      "flos": 25264844818560.0,
      "grad_norm": 2.3493794544032003,
      "language_loss": 0.72507954,
      "learning_rate": 1.4569504141529433e-06,
      "loss": 0.74909461,
      "num_input_tokens_seen": 107585410,
      "step": 4989,
      "time_per_iteration": 2.7434444427490234
    },
    {
      "auxiliary_loss_clip": 0.01235842,
      "auxiliary_loss_mlp": 0.01028632,
      "balance_loss_clip": 1.05495656,
      "balance_loss_mlp": 1.02115726,
      "epoch": 0.6000120242890639,
      "flos": 22054502862720.0,
      "grad_norm": 2.655564464153192,
      "language_loss": 0.72349858,
      "learning_rate": 1.456200748553658e-06,
      "loss": 0.74614328,
      "num_input_tokens_seen": 107603405,
      "step": 4990,
      "time_per_iteration": 2.6351711750030518
    },
    {
      "auxiliary_loss_clip": 0.01182083,
      "auxiliary_loss_mlp": 0.01024129,
      "balance_loss_clip": 1.05373394,
      "balance_loss_mlp": 1.01677966,
      "epoch": 0.600132267179703,
      "flos": 29863421562240.0,
      "grad_norm": 1.6185906012753408,
      "language_loss": 0.78932512,
      "learning_rate": 1.455451165448287e-06,
      "loss": 0.81138724,
      "num_input_tokens_seen": 107626060,
      "step": 4991,
      "time_per_iteration": 2.704078197479248
    },
    {
      "auxiliary_loss_clip": 0.01277605,
      "auxiliary_loss_mlp": 0.01026438,
      "balance_loss_clip": 1.04905009,
      "balance_loss_mlp": 1.0190109,
      "epoch": 0.6002525100703421,
      "flos": 25045358762880.0,
      "grad_norm": 2.633404017167911,
      "language_loss": 0.73187929,
      "learning_rate": 1.4547016649505407e-06,
      "loss": 0.75491977,
      "num_input_tokens_seen": 107644070,
      "step": 4992,
      "time_per_iteration": 2.6411120891571045
    },
    {
      "auxiliary_loss_clip": 0.01328455,
      "auxiliary_loss_mlp": 0.01025853,
      "balance_loss_clip": 1.04413712,
      "balance_loss_mlp": 1.01879025,
      "epoch": 0.6003727529609811,
      "flos": 20849592113280.0,
      "grad_norm": 16.03565795513229,
      "language_loss": 0.84734726,
      "learning_rate": 1.4539522471741193e-06,
      "loss": 0.87089032,
      "num_input_tokens_seen": 107661495,
      "step": 4993,
      "time_per_iteration": 2.7058205604553223
    },
    {
      "auxiliary_loss_clip": 0.01234317,
      "auxiliary_loss_mlp": 0.01031659,
      "balance_loss_clip": 1.05124021,
      "balance_loss_mlp": 1.02398181,
      "epoch": 0.6004929958516203,
      "flos": 15594604277760.0,
      "grad_norm": 2.897092949821549,
      "language_loss": 0.70965207,
      "learning_rate": 1.4532029122327067e-06,
      "loss": 0.73231179,
      "num_input_tokens_seen": 107678280,
      "step": 4994,
      "time_per_iteration": 2.690169334411621
    },
    {
      "auxiliary_loss_clip": 0.01325091,
      "auxiliary_loss_mlp": 0.01024731,
      "balance_loss_clip": 1.05053258,
      "balance_loss_mlp": 1.01809049,
      "epoch": 0.6006132387422594,
      "flos": 21763267390080.0,
      "grad_norm": 2.2076842159742087,
      "language_loss": 0.75611639,
      "learning_rate": 1.4524536602399783e-06,
      "loss": 0.77961463,
      "num_input_tokens_seen": 107697370,
      "step": 4995,
      "time_per_iteration": 2.7360687255859375
    },
    {
      "auxiliary_loss_clip": 0.0127236,
      "auxiliary_loss_mlp": 0.01037277,
      "balance_loss_clip": 1.04996502,
      "balance_loss_mlp": 1.02959991,
      "epoch": 0.6007334816328984,
      "flos": 22858542852480.0,
      "grad_norm": 2.020546614684073,
      "language_loss": 0.77177954,
      "learning_rate": 1.4517044913095938e-06,
      "loss": 0.79487592,
      "num_input_tokens_seen": 107717790,
      "step": 4996,
      "time_per_iteration": 2.6986284255981445
    },
    {
      "auxiliary_loss_clip": 0.01228264,
      "auxiliary_loss_mlp": 0.0102379,
      "balance_loss_clip": 1.05115879,
      "balance_loss_mlp": 1.01677799,
      "epoch": 0.6008537245235376,
      "flos": 28324577047680.0,
      "grad_norm": 1.797895157939401,
      "language_loss": 0.81607342,
      "learning_rate": 1.4509554055552022e-06,
      "loss": 0.83859396,
      "num_input_tokens_seen": 107738020,
      "step": 4997,
      "time_per_iteration": 2.7828688621520996
    },
    {
      "auxiliary_loss_clip": 0.01277227,
      "auxiliary_loss_mlp": 0.01030338,
      "balance_loss_clip": 1.04780316,
      "balance_loss_mlp": 1.02290511,
      "epoch": 0.6009739674141766,
      "flos": 20886113266560.0,
      "grad_norm": 2.3437118836755393,
      "language_loss": 0.83947289,
      "learning_rate": 1.450206403090439e-06,
      "loss": 0.86254847,
      "num_input_tokens_seen": 107756215,
      "step": 4998,
      "time_per_iteration": 3.755180597305298
    },
    {
      "auxiliary_loss_clip": 0.01222979,
      "auxiliary_loss_mlp": 0.01026444,
      "balance_loss_clip": 1.04960907,
      "balance_loss_mlp": 1.02001858,
      "epoch": 0.6010942103048157,
      "flos": 20481004702080.0,
      "grad_norm": 2.0609127894036656,
      "language_loss": 0.86758608,
      "learning_rate": 1.4494574840289274e-06,
      "loss": 0.89008033,
      "num_input_tokens_seen": 107773330,
      "step": 4999,
      "time_per_iteration": 2.658665895462036
    },
    {
      "auxiliary_loss_clip": 0.0123595,
      "auxiliary_loss_mlp": 0.01028193,
      "balance_loss_clip": 1.05032873,
      "balance_loss_mlp": 1.02045012,
      "epoch": 0.6012144531954549,
      "flos": 23805973935360.0,
      "grad_norm": 2.428905565698822,
      "language_loss": 0.73938626,
      "learning_rate": 1.4487086484842782e-06,
      "loss": 0.76202768,
      "num_input_tokens_seen": 107791975,
      "step": 5000,
      "time_per_iteration": 3.602520227432251
    },
    {
      "auxiliary_loss_clip": 0.01179325,
      "auxiliary_loss_mlp": 0.01025892,
      "balance_loss_clip": 1.05158162,
      "balance_loss_mlp": 1.01897764,
      "epoch": 0.6013346960860939,
      "flos": 18988378012800.0,
      "grad_norm": 2.008251606523724,
      "language_loss": 0.60384429,
      "learning_rate": 1.4479598965700878e-06,
      "loss": 0.62589645,
      "num_input_tokens_seen": 107809240,
      "step": 5001,
      "time_per_iteration": 3.495940923690796
    },
    {
      "auxiliary_loss_clip": 0.01325189,
      "auxiliary_loss_mlp": 0.0102827,
      "balance_loss_clip": 1.04440928,
      "balance_loss_mlp": 1.02105188,
      "epoch": 0.601454938976733,
      "flos": 24025316336640.0,
      "grad_norm": 2.2058892719794896,
      "language_loss": 0.68816835,
      "learning_rate": 1.4472112283999427e-06,
      "loss": 0.71170294,
      "num_input_tokens_seen": 107827895,
      "step": 5002,
      "time_per_iteration": 2.711676597595215
    },
    {
      "auxiliary_loss_clip": 0.01230128,
      "auxiliary_loss_mlp": 0.01024781,
      "balance_loss_clip": 1.05254197,
      "balance_loss_mlp": 1.01760435,
      "epoch": 0.6015751818673721,
      "flos": 26427129102720.0,
      "grad_norm": 1.9882913523248578,
      "language_loss": 0.69265223,
      "learning_rate": 1.4464626440874143e-06,
      "loss": 0.71520138,
      "num_input_tokens_seen": 107847010,
      "step": 5003,
      "time_per_iteration": 2.6581599712371826
    },
    {
      "auxiliary_loss_clip": 0.01287127,
      "auxiliary_loss_mlp": 0.0102872,
      "balance_loss_clip": 1.04364741,
      "balance_loss_mlp": 1.02126944,
      "epoch": 0.6016954247580112,
      "flos": 13115260005120.0,
      "grad_norm": 2.4728459276616945,
      "language_loss": 0.74232656,
      "learning_rate": 1.4457141437460636e-06,
      "loss": 0.76548505,
      "num_input_tokens_seen": 107864235,
      "step": 5004,
      "time_per_iteration": 2.7309627532958984
    },
    {
      "auxiliary_loss_clip": 0.01278933,
      "auxiliary_loss_mlp": 0.01026381,
      "balance_loss_clip": 1.04918361,
      "balance_loss_mlp": 1.01852548,
      "epoch": 0.6018156676486502,
      "flos": 23768447201280.0,
      "grad_norm": 1.689996030921193,
      "language_loss": 0.73184574,
      "learning_rate": 1.444965727489436e-06,
      "loss": 0.75489891,
      "num_input_tokens_seen": 107883680,
      "step": 5005,
      "time_per_iteration": 2.715041399002075
    },
    {
      "auxiliary_loss_clip": 0.01328689,
      "auxiliary_loss_mlp": 0.01027083,
      "balance_loss_clip": 1.04526901,
      "balance_loss_mlp": 1.01970649,
      "epoch": 0.6019359105392894,
      "flos": 26469360518400.0,
      "grad_norm": 1.7003024254803598,
      "language_loss": 0.63348317,
      "learning_rate": 1.444217395431066e-06,
      "loss": 0.65704089,
      "num_input_tokens_seen": 107906220,
      "step": 5006,
      "time_per_iteration": 3.7348692417144775
    },
    {
      "auxiliary_loss_clip": 0.01228912,
      "auxiliary_loss_mlp": 0.01003527,
      "balance_loss_clip": 1.01576447,
      "balance_loss_mlp": 1.00258529,
      "epoch": 0.6020561534299285,
      "flos": 69190849728000.0,
      "grad_norm": 0.7972476205367459,
      "language_loss": 0.55762768,
      "learning_rate": 1.4434691476844755e-06,
      "loss": 0.57995212,
      "num_input_tokens_seen": 107967195,
      "step": 5007,
      "time_per_iteration": 3.3232083320617676
    },
    {
      "auxiliary_loss_clip": 0.01276316,
      "auxiliary_loss_mlp": 0.01032397,
      "balance_loss_clip": 1.05035603,
      "balance_loss_mlp": 1.02501237,
      "epoch": 0.6021763963205675,
      "flos": 21835304115840.0,
      "grad_norm": 2.218038815036032,
      "language_loss": 0.66917479,
      "learning_rate": 1.4427209843631729e-06,
      "loss": 0.69226193,
      "num_input_tokens_seen": 107984245,
      "step": 5008,
      "time_per_iteration": 2.8046412467956543
    },
    {
      "auxiliary_loss_clip": 0.01178219,
      "auxiliary_loss_mlp": 0.02566774,
      "balance_loss_clip": 1.0514133,
      "balance_loss_mlp": 1.00004959,
      "epoch": 0.6022966392112067,
      "flos": 26578636669440.0,
      "grad_norm": 1.904332361372777,
      "language_loss": 0.81346107,
      "learning_rate": 1.4419729055806534e-06,
      "loss": 0.85091102,
      "num_input_tokens_seen": 108003680,
      "step": 5009,
      "time_per_iteration": 2.660416841506958
    },
    {
      "auxiliary_loss_clip": 0.01272898,
      "auxiliary_loss_mlp": 0.02564853,
      "balance_loss_clip": 1.04989588,
      "balance_loss_mlp": 0.99999154,
      "epoch": 0.6024168821018457,
      "flos": 20703722981760.0,
      "grad_norm": 1.7338182783881266,
      "language_loss": 0.82429445,
      "learning_rate": 1.441224911450401e-06,
      "loss": 0.86267197,
      "num_input_tokens_seen": 108019635,
      "step": 5010,
      "time_per_iteration": 2.7176027297973633
    },
    {
      "auxiliary_loss_clip": 0.01232457,
      "auxiliary_loss_mlp": 0.01027394,
      "balance_loss_clip": 1.04979181,
      "balance_loss_mlp": 1.01986313,
      "epoch": 0.6025371249924848,
      "flos": 24680973242880.0,
      "grad_norm": 1.9081200557476563,
      "language_loss": 0.82389176,
      "learning_rate": 1.4404770020858851e-06,
      "loss": 0.84649026,
      "num_input_tokens_seen": 108039120,
      "step": 5011,
      "time_per_iteration": 2.7317960262298584
    },
    {
      "auxiliary_loss_clip": 0.01223152,
      "auxiliary_loss_mlp": 0.01023632,
      "balance_loss_clip": 1.04894447,
      "balance_loss_mlp": 1.01681352,
      "epoch": 0.602657367883124,
      "flos": 25955801815680.0,
      "grad_norm": 2.080285334622596,
      "language_loss": 0.86318272,
      "learning_rate": 1.439729177600563e-06,
      "loss": 0.88565052,
      "num_input_tokens_seen": 108059615,
      "step": 5012,
      "time_per_iteration": 2.796569585800171
    },
    {
      "auxiliary_loss_clip": 0.01228813,
      "auxiliary_loss_mlp": 0.01029264,
      "balance_loss_clip": 1.05380964,
      "balance_loss_mlp": 1.02217376,
      "epoch": 0.602777610773763,
      "flos": 16690633925760.0,
      "grad_norm": 2.2254281548957224,
      "language_loss": 0.73425752,
      "learning_rate": 1.4389814381078793e-06,
      "loss": 0.75683832,
      "num_input_tokens_seen": 108078855,
      "step": 5013,
      "time_per_iteration": 2.686742067337036
    },
    {
      "auxiliary_loss_clip": 0.01577815,
      "auxiliary_loss_mlp": 0.01027471,
      "balance_loss_clip": 1.04248786,
      "balance_loss_mlp": 1.02048874,
      "epoch": 0.6028978536644021,
      "flos": 13334243270400.0,
      "grad_norm": 2.1875956001995553,
      "language_loss": 0.80085951,
      "learning_rate": 1.438233783721265e-06,
      "loss": 0.8269124,
      "num_input_tokens_seen": 108095020,
      "step": 5014,
      "time_per_iteration": 3.048121452331543
    },
    {
      "auxiliary_loss_clip": 0.01278946,
      "auxiliary_loss_mlp": 0.0102523,
      "balance_loss_clip": 1.05517852,
      "balance_loss_mlp": 1.0184145,
      "epoch": 0.6030180965550412,
      "flos": 19644825018240.0,
      "grad_norm": 2.102401068104626,
      "language_loss": 0.78084451,
      "learning_rate": 1.43748621455414e-06,
      "loss": 0.80388629,
      "num_input_tokens_seen": 108111455,
      "step": 5015,
      "time_per_iteration": 3.2835686206817627
    },
    {
      "auxiliary_loss_clip": 0.01274135,
      "auxiliary_loss_mlp": 0.01028942,
      "balance_loss_clip": 1.04849255,
      "balance_loss_mlp": 1.02089548,
      "epoch": 0.6031383394456803,
      "flos": 14458390289280.0,
      "grad_norm": 2.5356648149773457,
      "language_loss": 0.80820858,
      "learning_rate": 1.4367387307199082e-06,
      "loss": 0.83123934,
      "num_input_tokens_seen": 108128305,
      "step": 5016,
      "time_per_iteration": 2.686060667037964
    },
    {
      "auxiliary_loss_clip": 0.01225617,
      "auxiliary_loss_mlp": 0.01025777,
      "balance_loss_clip": 1.047194,
      "balance_loss_mlp": 1.01810026,
      "epoch": 0.6032585823363193,
      "flos": 13917791623680.0,
      "grad_norm": 1.7355313547302338,
      "language_loss": 0.82548952,
      "learning_rate": 1.4359913323319632e-06,
      "loss": 0.84800351,
      "num_input_tokens_seen": 108145475,
      "step": 5017,
      "time_per_iteration": 2.7059874534606934
    },
    {
      "auxiliary_loss_clip": 0.01420945,
      "auxiliary_loss_mlp": 0.01031628,
      "balance_loss_clip": 1.0419662,
      "balance_loss_mlp": 1.02430582,
      "epoch": 0.6033788252269584,
      "flos": 24353252530560.0,
      "grad_norm": 2.010110050340027,
      "language_loss": 0.77951711,
      "learning_rate": 1.4352440195036847e-06,
      "loss": 0.80404282,
      "num_input_tokens_seen": 108165650,
      "step": 5018,
      "time_per_iteration": 2.886763095855713
    },
    {
      "auxiliary_loss_clip": 0.01380064,
      "auxiliary_loss_mlp": 0.01031275,
      "balance_loss_clip": 1.0405798,
      "balance_loss_mlp": 1.02423286,
      "epoch": 0.6034990681175976,
      "flos": 25521247077120.0,
      "grad_norm": 2.1098677521393507,
      "language_loss": 0.80119169,
      "learning_rate": 1.4344967923484395e-06,
      "loss": 0.8253051,
      "num_input_tokens_seen": 108187620,
      "step": 5019,
      "time_per_iteration": 2.916806221008301
    },
    {
      "auxiliary_loss_clip": 0.01226752,
      "auxiliary_loss_mlp": 0.01029331,
      "balance_loss_clip": 1.05001616,
      "balance_loss_mlp": 1.0218749,
      "epoch": 0.6036193110082366,
      "flos": 25958387594880.0,
      "grad_norm": 2.1711853162579815,
      "language_loss": 0.72255075,
      "learning_rate": 1.433749650979581e-06,
      "loss": 0.74511158,
      "num_input_tokens_seen": 108207605,
      "step": 5020,
      "time_per_iteration": 2.7875478267669678
    },
    {
      "auxiliary_loss_clip": 0.01233862,
      "auxiliary_loss_mlp": 0.01021005,
      "balance_loss_clip": 1.04860198,
      "balance_loss_mlp": 1.01411819,
      "epoch": 0.6037395538988757,
      "flos": 25593427457280.0,
      "grad_norm": 1.8945913533465975,
      "language_loss": 0.68201149,
      "learning_rate": 1.433002595510451e-06,
      "loss": 0.70456016,
      "num_input_tokens_seen": 108226385,
      "step": 5021,
      "time_per_iteration": 2.8953819274902344
    },
    {
      "auxiliary_loss_clip": 0.0127711,
      "auxiliary_loss_mlp": 0.0256761,
      "balance_loss_clip": 1.04688931,
      "balance_loss_mlp": 0.99998057,
      "epoch": 0.6038597967895148,
      "flos": 17816253402240.0,
      "grad_norm": 1.871680781186828,
      "language_loss": 0.71912295,
      "learning_rate": 1.4322556260543757e-06,
      "loss": 0.75757015,
      "num_input_tokens_seen": 108242960,
      "step": 5022,
      "time_per_iteration": 2.783935785293579
    },
    {
      "auxiliary_loss_clip": 0.01226116,
      "auxiliary_loss_mlp": 0.01002395,
      "balance_loss_clip": 1.00887036,
      "balance_loss_mlp": 1.00148857,
      "epoch": 0.6039800396801539,
      "flos": 65169213235200.0,
      "grad_norm": 0.8983325597508501,
      "language_loss": 0.62655783,
      "learning_rate": 1.4315087427246703e-06,
      "loss": 0.64884293,
      "num_input_tokens_seen": 108296785,
      "step": 5023,
      "time_per_iteration": 3.2141313552856445
    },
    {
      "auxiliary_loss_clip": 0.01065338,
      "auxiliary_loss_mlp": 0.01004061,
      "balance_loss_clip": 1.00782537,
      "balance_loss_mlp": 1.00310695,
      "epoch": 0.604100282570793,
      "flos": 67386409073280.0,
      "grad_norm": 0.8689754599828391,
      "language_loss": 0.58425117,
      "learning_rate": 1.4307619456346372e-06,
      "loss": 0.60494512,
      "num_input_tokens_seen": 108341090,
      "step": 5024,
      "time_per_iteration": 4.5787739753723145
    },
    {
      "auxiliary_loss_clip": 0.01232447,
      "auxiliary_loss_mlp": 0.01025663,
      "balance_loss_clip": 1.04818475,
      "balance_loss_mlp": 1.01825452,
      "epoch": 0.6042205254614321,
      "flos": 35297495631360.0,
      "grad_norm": 2.9019303056002057,
      "language_loss": 0.74097514,
      "learning_rate": 1.430015234897564e-06,
      "loss": 0.76355624,
      "num_input_tokens_seen": 108364370,
      "step": 5025,
      "time_per_iteration": 2.8195230960845947
    },
    {
      "auxiliary_loss_clip": 0.01179874,
      "auxiliary_loss_mlp": 0.02568029,
      "balance_loss_clip": 1.05033517,
      "balance_loss_mlp": 0.99997336,
      "epoch": 0.6043407683520712,
      "flos": 45658262206080.0,
      "grad_norm": 1.6076980803066245,
      "language_loss": 0.66399097,
      "learning_rate": 1.4292686106267274e-06,
      "loss": 0.70146996,
      "num_input_tokens_seen": 108387220,
      "step": 5026,
      "time_per_iteration": 3.7468481063842773
    },
    {
      "auxiliary_loss_clip": 0.01236322,
      "auxiliary_loss_mlp": 0.01032866,
      "balance_loss_clip": 1.05159044,
      "balance_loss_mlp": 1.02570176,
      "epoch": 0.6044610112427102,
      "flos": 16180020138240.0,
      "grad_norm": 1.7525948249560988,
      "language_loss": 0.7750802,
      "learning_rate": 1.4285220729353876e-06,
      "loss": 0.79777205,
      "num_input_tokens_seen": 108405760,
      "step": 5027,
      "time_per_iteration": 3.929567813873291
    },
    {
      "auxiliary_loss_clip": 0.01279253,
      "auxiliary_loss_mlp": 0.01030591,
      "balance_loss_clip": 1.04539084,
      "balance_loss_mlp": 1.02280688,
      "epoch": 0.6045812541333494,
      "flos": 13804062186240.0,
      "grad_norm": 3.966907182364601,
      "language_loss": 0.77950674,
      "learning_rate": 1.4277756219367957e-06,
      "loss": 0.80260515,
      "num_input_tokens_seen": 108422785,
      "step": 5028,
      "time_per_iteration": 2.7531237602233887
    },
    {
      "auxiliary_loss_clip": 0.01339315,
      "auxiliary_loss_mlp": 0.01028041,
      "balance_loss_clip": 1.05094194,
      "balance_loss_mlp": 1.01992321,
      "epoch": 0.6047014970239885,
      "flos": 19975059682560.0,
      "grad_norm": 3.0033499232898087,
      "language_loss": 0.7974084,
      "learning_rate": 1.4270292577441864e-06,
      "loss": 0.82108188,
      "num_input_tokens_seen": 108442290,
      "step": 5029,
      "time_per_iteration": 2.7097952365875244
    },
    {
      "auxiliary_loss_clip": 0.01233704,
      "auxiliary_loss_mlp": 0.01026336,
      "balance_loss_clip": 1.04827952,
      "balance_loss_mlp": 1.01881957,
      "epoch": 0.6048217399146275,
      "flos": 25337097025920.0,
      "grad_norm": 1.9545509554948204,
      "language_loss": 0.71638322,
      "learning_rate": 1.4262829804707836e-06,
      "loss": 0.73898357,
      "num_input_tokens_seen": 108464280,
      "step": 5030,
      "time_per_iteration": 2.7280526161193848
    },
    {
      "auxiliary_loss_clip": 0.01233521,
      "auxiliary_loss_mlp": 0.01034262,
      "balance_loss_clip": 1.04931927,
      "balance_loss_mlp": 1.02647817,
      "epoch": 0.6049419828052667,
      "flos": 26030819370240.0,
      "grad_norm": 1.4066600806637084,
      "language_loss": 0.70025283,
      "learning_rate": 1.4255367902297958e-06,
      "loss": 0.72293067,
      "num_input_tokens_seen": 108485610,
      "step": 5031,
      "time_per_iteration": 2.7799429893493652
    },
    {
      "auxiliary_loss_clip": 0.011801,
      "auxiliary_loss_mlp": 0.01029391,
      "balance_loss_clip": 1.05210042,
      "balance_loss_mlp": 1.0225575,
      "epoch": 0.6050622256959057,
      "flos": 14648106948480.0,
      "grad_norm": 2.6613396842999713,
      "language_loss": 0.78924572,
      "learning_rate": 1.4247906871344215e-06,
      "loss": 0.81134063,
      "num_input_tokens_seen": 108501005,
      "step": 5032,
      "time_per_iteration": 3.550607919692993
    },
    {
      "auxiliary_loss_clip": 0.01272555,
      "auxiliary_loss_mlp": 0.01025608,
      "balance_loss_clip": 1.04404557,
      "balance_loss_mlp": 1.0186516,
      "epoch": 0.6051824685865448,
      "flos": 23331450337920.0,
      "grad_norm": 2.5644566588622633,
      "language_loss": 0.74938524,
      "learning_rate": 1.4240446712978415e-06,
      "loss": 0.77236682,
      "num_input_tokens_seen": 108519990,
      "step": 5033,
      "time_per_iteration": 2.817582607269287
    },
    {
      "auxiliary_loss_clip": 0.01234036,
      "auxiliary_loss_mlp": 0.01029941,
      "balance_loss_clip": 1.04955792,
      "balance_loss_mlp": 1.02169716,
      "epoch": 0.605302711477184,
      "flos": 27563307177600.0,
      "grad_norm": 1.9070273822012955,
      "language_loss": 0.74593818,
      "learning_rate": 1.423298742833227e-06,
      "loss": 0.76857793,
      "num_input_tokens_seen": 108538650,
      "step": 5034,
      "time_per_iteration": 2.7709474563598633
    },
    {
      "auxiliary_loss_clip": 0.01333371,
      "auxiliary_loss_mlp": 0.01024834,
      "balance_loss_clip": 1.04626477,
      "balance_loss_mlp": 1.01726973,
      "epoch": 0.605422954367823,
      "flos": 15154698412800.0,
      "grad_norm": 2.3823539755416467,
      "language_loss": 0.71914679,
      "learning_rate": 1.4225529018537352e-06,
      "loss": 0.74272883,
      "num_input_tokens_seen": 108554155,
      "step": 5035,
      "time_per_iteration": 2.739236354827881
    },
    {
      "auxiliary_loss_clip": 0.01181977,
      "auxiliary_loss_mlp": 0.01034175,
      "balance_loss_clip": 1.05219126,
      "balance_loss_mlp": 1.02677214,
      "epoch": 0.6055431972584621,
      "flos": 27673912131840.0,
      "grad_norm": 3.603574733174264,
      "language_loss": 0.77924377,
      "learning_rate": 1.4218071484725082e-06,
      "loss": 0.80140531,
      "num_input_tokens_seen": 108576275,
      "step": 5036,
      "time_per_iteration": 2.693207025527954
    },
    {
      "auxiliary_loss_clip": 0.01280212,
      "auxiliary_loss_mlp": 0.01031364,
      "balance_loss_clip": 1.05347753,
      "balance_loss_mlp": 1.02349019,
      "epoch": 0.6056634401491012,
      "flos": 19387489006080.0,
      "grad_norm": 2.086728143538502,
      "language_loss": 0.76417339,
      "learning_rate": 1.4210614828026786e-06,
      "loss": 0.78728914,
      "num_input_tokens_seen": 108594125,
      "step": 5037,
      "time_per_iteration": 2.709721326828003
    },
    {
      "auxiliary_loss_clip": 0.011792,
      "auxiliary_loss_mlp": 0.01031934,
      "balance_loss_clip": 1.04998922,
      "balance_loss_mlp": 1.02493072,
      "epoch": 0.6057836830397403,
      "flos": 24789459294720.0,
      "grad_norm": 2.031425669881099,
      "language_loss": 0.74455422,
      "learning_rate": 1.4203159049573605e-06,
      "loss": 0.76666558,
      "num_input_tokens_seen": 108615360,
      "step": 5038,
      "time_per_iteration": 2.8645055294036865
    },
    {
      "auxiliary_loss_clip": 0.01187378,
      "auxiliary_loss_mlp": 0.01026192,
      "balance_loss_clip": 1.04781139,
      "balance_loss_mlp": 1.01887512,
      "epoch": 0.6059039259303793,
      "flos": 20558248899840.0,
      "grad_norm": 2.2937301483063397,
      "language_loss": 0.87160778,
      "learning_rate": 1.4195704150496593e-06,
      "loss": 0.89374352,
      "num_input_tokens_seen": 108633075,
      "step": 5039,
      "time_per_iteration": 2.7440223693847656
    },
    {
      "auxiliary_loss_clip": 0.01281362,
      "auxiliary_loss_mlp": 0.01028728,
      "balance_loss_clip": 1.05087733,
      "balance_loss_mlp": 1.02100921,
      "epoch": 0.6060241688210185,
      "flos": 21069724613760.0,
      "grad_norm": 1.7075092883101335,
      "language_loss": 0.73981327,
      "learning_rate": 1.4188250131926639e-06,
      "loss": 0.76291418,
      "num_input_tokens_seen": 108651875,
      "step": 5040,
      "time_per_iteration": 2.6357831954956055
    },
    {
      "auxiliary_loss_clip": 0.01278966,
      "auxiliary_loss_mlp": 0.01025436,
      "balance_loss_clip": 1.04639375,
      "balance_loss_mlp": 1.01812506,
      "epoch": 0.6061444117116576,
      "flos": 16361081619840.0,
      "grad_norm": 2.3033657665835627,
      "language_loss": 0.80372596,
      "learning_rate": 1.4180796994994525e-06,
      "loss": 0.82677001,
      "num_input_tokens_seen": 108669290,
      "step": 5041,
      "time_per_iteration": 2.6870367527008057
    },
    {
      "auxiliary_loss_clip": 0.01276345,
      "auxiliary_loss_mlp": 0.01028131,
      "balance_loss_clip": 1.04726529,
      "balance_loss_mlp": 1.0204004,
      "epoch": 0.6062646546022966,
      "flos": 21507296094720.0,
      "grad_norm": 1.9399255968810547,
      "language_loss": 0.71836913,
      "learning_rate": 1.4173344740830877e-06,
      "loss": 0.74141389,
      "num_input_tokens_seen": 108688420,
      "step": 5042,
      "time_per_iteration": 2.7284653186798096
    },
    {
      "auxiliary_loss_clip": 0.01283088,
      "auxiliary_loss_mlp": 0.01026125,
      "balance_loss_clip": 1.05286574,
      "balance_loss_mlp": 1.01877522,
      "epoch": 0.6063848974929358,
      "flos": 38983151283840.0,
      "grad_norm": 1.6117692071480423,
      "language_loss": 0.70862311,
      "learning_rate": 1.4165893370566206e-06,
      "loss": 0.7317152,
      "num_input_tokens_seen": 108712175,
      "step": 5043,
      "time_per_iteration": 2.7989611625671387
    },
    {
      "auxiliary_loss_clip": 0.01229442,
      "auxiliary_loss_mlp": 0.01027001,
      "balance_loss_clip": 1.04831886,
      "balance_loss_mlp": 1.01930881,
      "epoch": 0.6065051403835748,
      "flos": 19646584784640.0,
      "grad_norm": 1.8178210417728258,
      "language_loss": 0.77845156,
      "learning_rate": 1.4158442885330865e-06,
      "loss": 0.80101597,
      "num_input_tokens_seen": 108730745,
      "step": 5044,
      "time_per_iteration": 2.6994550228118896
    },
    {
      "auxiliary_loss_clip": 0.01226311,
      "auxiliary_loss_mlp": 0.01026883,
      "balance_loss_clip": 1.04766738,
      "balance_loss_mlp": 1.01953411,
      "epoch": 0.6066253832742139,
      "flos": 23513086437120.0,
      "grad_norm": 1.9846539028871306,
      "language_loss": 0.78971744,
      "learning_rate": 1.4150993286255094e-06,
      "loss": 0.81224942,
      "num_input_tokens_seen": 108749995,
      "step": 5045,
      "time_per_iteration": 2.6694443225860596
    },
    {
      "auxiliary_loss_clip": 0.01178727,
      "auxiliary_loss_mlp": 0.01031582,
      "balance_loss_clip": 1.05026615,
      "balance_loss_mlp": 1.02461398,
      "epoch": 0.6067456261648531,
      "flos": 19133708440320.0,
      "grad_norm": 2.1178600239081207,
      "language_loss": 0.79970717,
      "learning_rate": 1.4143544574468993e-06,
      "loss": 0.82181025,
      "num_input_tokens_seen": 108768355,
      "step": 5046,
      "time_per_iteration": 2.6635169982910156
    },
    {
      "auxiliary_loss_clip": 0.01225376,
      "auxiliary_loss_mlp": 0.01026207,
      "balance_loss_clip": 1.04951727,
      "balance_loss_mlp": 1.01891458,
      "epoch": 0.6068658690554921,
      "flos": 20520614424960.0,
      "grad_norm": 1.705358468667441,
      "language_loss": 0.82405257,
      "learning_rate": 1.4136096751102523e-06,
      "loss": 0.84656841,
      "num_input_tokens_seen": 108786685,
      "step": 5047,
      "time_per_iteration": 2.773298740386963
    },
    {
      "auxiliary_loss_clip": 0.01281522,
      "auxiliary_loss_mlp": 0.01031037,
      "balance_loss_clip": 1.05042124,
      "balance_loss_mlp": 1.02362788,
      "epoch": 0.6069861119461312,
      "flos": 27374560185600.0,
      "grad_norm": 2.382053825167616,
      "language_loss": 0.8273164,
      "learning_rate": 1.4128649817285516e-06,
      "loss": 0.85044199,
      "num_input_tokens_seen": 108804820,
      "step": 5048,
      "time_per_iteration": 2.830150842666626
    },
    {
      "auxiliary_loss_clip": 0.01278899,
      "auxiliary_loss_mlp": 0.01031956,
      "balance_loss_clip": 1.04491341,
      "balance_loss_mlp": 1.02444839,
      "epoch": 0.6071063548367702,
      "flos": 25626500904960.0,
      "grad_norm": 1.8738239737712616,
      "language_loss": 0.63194084,
      "learning_rate": 1.412120377414766e-06,
      "loss": 0.65504938,
      "num_input_tokens_seen": 108825010,
      "step": 5049,
      "time_per_iteration": 2.8006887435913086
    },
    {
      "auxiliary_loss_clip": 0.01178871,
      "auxiliary_loss_mlp": 0.01035334,
      "balance_loss_clip": 1.05214274,
      "balance_loss_mlp": 1.02740908,
      "epoch": 0.6072265977274094,
      "flos": 24460517520000.0,
      "grad_norm": 1.6620979789864923,
      "language_loss": 0.71454352,
      "learning_rate": 1.4113758622818522e-06,
      "loss": 0.73668551,
      "num_input_tokens_seen": 108845075,
      "step": 5050,
      "time_per_iteration": 4.141252040863037
    },
    {
      "auxiliary_loss_clip": 0.01286883,
      "auxiliary_loss_mlp": 0.02564543,
      "balance_loss_clip": 1.05218256,
      "balance_loss_mlp": 0.99996841,
      "epoch": 0.6073468406180484,
      "flos": 18149253413760.0,
      "grad_norm": 2.0492725985942313,
      "language_loss": 0.83405542,
      "learning_rate": 1.410631436442751e-06,
      "loss": 0.87256968,
      "num_input_tokens_seen": 108863870,
      "step": 5051,
      "time_per_iteration": 3.732515573501587
    },
    {
      "auxiliary_loss_clip": 0.01233554,
      "auxiliary_loss_mlp": 0.01032185,
      "balance_loss_clip": 1.04850221,
      "balance_loss_mlp": 1.02479386,
      "epoch": 0.6074670835086875,
      "flos": 20697617669760.0,
      "grad_norm": 1.9869314742199573,
      "language_loss": 0.86571693,
      "learning_rate": 1.4098871000103936e-06,
      "loss": 0.88837427,
      "num_input_tokens_seen": 108882470,
      "step": 5052,
      "time_per_iteration": 2.7047574520111084
    },
    {
      "auxiliary_loss_clip": 0.01180527,
      "auxiliary_loss_mlp": 0.01025095,
      "balance_loss_clip": 1.04677737,
      "balance_loss_mlp": 1.01800203,
      "epoch": 0.6075873263993267,
      "flos": 23769955572480.0,
      "grad_norm": 2.1586860286235723,
      "language_loss": 0.82618189,
      "learning_rate": 1.409142853097693e-06,
      "loss": 0.84823811,
      "num_input_tokens_seen": 108902710,
      "step": 5053,
      "time_per_iteration": 3.6650140285491943
    },
    {
      "auxiliary_loss_clip": 0.01278518,
      "auxiliary_loss_mlp": 0.01027451,
      "balance_loss_clip": 1.04855883,
      "balance_loss_mlp": 1.01957679,
      "epoch": 0.6077075692899657,
      "flos": 24454484035200.0,
      "grad_norm": 2.056654171044102,
      "language_loss": 0.79671979,
      "learning_rate": 1.408398695817553e-06,
      "loss": 0.81977946,
      "num_input_tokens_seen": 108919935,
      "step": 5054,
      "time_per_iteration": 2.812471628189087
    },
    {
      "auxiliary_loss_clip": 0.01279205,
      "auxiliary_loss_mlp": 0.01029177,
      "balance_loss_clip": 1.04618406,
      "balance_loss_mlp": 1.02042747,
      "epoch": 0.6078278121806048,
      "flos": 27382102041600.0,
      "grad_norm": 3.19384642144775,
      "language_loss": 0.70598334,
      "learning_rate": 1.4076546282828593e-06,
      "loss": 0.72906721,
      "num_input_tokens_seen": 108942790,
      "step": 5055,
      "time_per_iteration": 2.776175022125244
    },
    {
      "auxiliary_loss_clip": 0.0128233,
      "auxiliary_loss_mlp": 0.01027296,
      "balance_loss_clip": 1.04473627,
      "balance_loss_mlp": 1.01953495,
      "epoch": 0.6079480550712439,
      "flos": 38436447306240.0,
      "grad_norm": 3.8082812372556285,
      "language_loss": 0.66488147,
      "learning_rate": 1.4069106506064874e-06,
      "loss": 0.68797779,
      "num_input_tokens_seen": 108964215,
      "step": 5056,
      "time_per_iteration": 2.9406931400299072
    },
    {
      "auxiliary_loss_clip": 0.01275921,
      "auxiliary_loss_mlp": 0.01024484,
      "balance_loss_clip": 1.05092275,
      "balance_loss_mlp": 1.0178473,
      "epoch": 0.608068297961883,
      "flos": 25336271013120.0,
      "grad_norm": 2.153407766123738,
      "language_loss": 0.78668827,
      "learning_rate": 1.4061667629012989e-06,
      "loss": 0.80969232,
      "num_input_tokens_seen": 108984885,
      "step": 5057,
      "time_per_iteration": 2.7238872051239014
    },
    {
      "auxiliary_loss_clip": 0.01270663,
      "auxiliary_loss_mlp": 0.01026188,
      "balance_loss_clip": 1.04790413,
      "balance_loss_mlp": 1.01894581,
      "epoch": 0.608188540852522,
      "flos": 24202463235840.0,
      "grad_norm": 1.7336160507821639,
      "language_loss": 0.83341753,
      "learning_rate": 1.40542296528014e-06,
      "loss": 0.85638607,
      "num_input_tokens_seen": 109004545,
      "step": 5058,
      "time_per_iteration": 3.622704267501831
    },
    {
      "auxiliary_loss_clip": 0.01229735,
      "auxiliary_loss_mlp": 0.01031717,
      "balance_loss_clip": 1.0479095,
      "balance_loss_mlp": 1.02399206,
      "epoch": 0.6083087837431612,
      "flos": 21284146851840.0,
      "grad_norm": 2.2880985935399076,
      "language_loss": 0.76162452,
      "learning_rate": 1.4046792578558452e-06,
      "loss": 0.78423899,
      "num_input_tokens_seen": 109022440,
      "step": 5059,
      "time_per_iteration": 2.7045700550079346
    },
    {
      "auxiliary_loss_clip": 0.0127794,
      "auxiliary_loss_mlp": 0.01026913,
      "balance_loss_clip": 1.04683256,
      "balance_loss_mlp": 1.02023149,
      "epoch": 0.6084290266338003,
      "flos": 16471435178880.0,
      "grad_norm": 2.6860940550840278,
      "language_loss": 0.76101631,
      "learning_rate": 1.4039356407412325e-06,
      "loss": 0.78406489,
      "num_input_tokens_seen": 109035680,
      "step": 5060,
      "time_per_iteration": 2.730055809020996
    },
    {
      "auxiliary_loss_clip": 0.01118837,
      "auxiliary_loss_mlp": 0.00999149,
      "balance_loss_clip": 1.00738192,
      "balance_loss_mlp": 0.99816573,
      "epoch": 0.6085492695244393,
      "flos": 66443574931200.0,
      "grad_norm": 0.7771161623019952,
      "language_loss": 0.57084423,
      "learning_rate": 1.40319211404911e-06,
      "loss": 0.59202409,
      "num_input_tokens_seen": 109090680,
      "step": 5061,
      "time_per_iteration": 3.218367576599121
    },
    {
      "auxiliary_loss_clip": 0.01182033,
      "auxiliary_loss_mlp": 0.01027899,
      "balance_loss_clip": 1.05204296,
      "balance_loss_mlp": 1.02093983,
      "epoch": 0.6086695124150785,
      "flos": 23618986709760.0,
      "grad_norm": 1.7604036922074076,
      "language_loss": 0.90510488,
      "learning_rate": 1.4024486778922691e-06,
      "loss": 0.92720419,
      "num_input_tokens_seen": 109108995,
      "step": 5062,
      "time_per_iteration": 2.684919834136963
    },
    {
      "auxiliary_loss_clip": 0.01284927,
      "auxiliary_loss_mlp": 0.01024403,
      "balance_loss_clip": 1.04760289,
      "balance_loss_mlp": 1.01724124,
      "epoch": 0.6087897553057176,
      "flos": 20157054917760.0,
      "grad_norm": 2.4989852901873943,
      "language_loss": 0.77817535,
      "learning_rate": 1.4017053323834884e-06,
      "loss": 0.80126864,
      "num_input_tokens_seen": 109128825,
      "step": 5063,
      "time_per_iteration": 2.6925270557403564
    },
    {
      "auxiliary_loss_clip": 0.01279401,
      "auxiliary_loss_mlp": 0.01026013,
      "balance_loss_clip": 1.04566479,
      "balance_loss_mlp": 1.01909876,
      "epoch": 0.6089099981963566,
      "flos": 25482535194240.0,
      "grad_norm": 2.0483751071076877,
      "language_loss": 0.76034105,
      "learning_rate": 1.4009620776355333e-06,
      "loss": 0.78339517,
      "num_input_tokens_seen": 109150425,
      "step": 5064,
      "time_per_iteration": 2.8085622787475586
    },
    {
      "auxiliary_loss_clip": 0.01228504,
      "auxiliary_loss_mlp": 0.01023567,
      "balance_loss_clip": 1.04720771,
      "balance_loss_mlp": 1.01616967,
      "epoch": 0.6090302410869958,
      "flos": 25332895134720.0,
      "grad_norm": 1.747191434384784,
      "language_loss": 0.79150331,
      "learning_rate": 1.4002189137611553e-06,
      "loss": 0.81402403,
      "num_input_tokens_seen": 109169765,
      "step": 5065,
      "time_per_iteration": 2.7434160709381104
    },
    {
      "auxiliary_loss_clip": 0.01227366,
      "auxiliary_loss_mlp": 0.01026951,
      "balance_loss_clip": 1.04738891,
      "balance_loss_mlp": 1.01976275,
      "epoch": 0.6091504839776348,
      "flos": 23987358639360.0,
      "grad_norm": 1.983116778925235,
      "language_loss": 0.69790727,
      "learning_rate": 1.3994758408730901e-06,
      "loss": 0.7204504,
      "num_input_tokens_seen": 109188950,
      "step": 5066,
      "time_per_iteration": 2.733985185623169
    },
    {
      "auxiliary_loss_clip": 0.01281335,
      "auxiliary_loss_mlp": 0.01026771,
      "balance_loss_clip": 1.05075538,
      "balance_loss_mlp": 1.01964247,
      "epoch": 0.6092707268682739,
      "flos": 29643037666560.0,
      "grad_norm": 24.778478681251023,
      "language_loss": 0.76264465,
      "learning_rate": 1.3987328590840629e-06,
      "loss": 0.78572571,
      "num_input_tokens_seen": 109209895,
      "step": 5067,
      "time_per_iteration": 2.7102322578430176
    },
    {
      "auxiliary_loss_clip": 0.01229304,
      "auxiliary_loss_mlp": 0.01032149,
      "balance_loss_clip": 1.04787052,
      "balance_loss_mlp": 1.02432275,
      "epoch": 0.609390969758913,
      "flos": 24024957200640.0,
      "grad_norm": 1.8250819287650555,
      "language_loss": 0.86709768,
      "learning_rate": 1.397989968506783e-06,
      "loss": 0.88971221,
      "num_input_tokens_seen": 109228905,
      "step": 5068,
      "time_per_iteration": 2.662245512008667
    },
    {
      "auxiliary_loss_clip": 0.01185095,
      "auxiliary_loss_mlp": 0.01027226,
      "balance_loss_clip": 1.05434442,
      "balance_loss_mlp": 1.01970053,
      "epoch": 0.6095112126495521,
      "flos": 11102143288320.0,
      "grad_norm": 2.4832002341530743,
      "language_loss": 0.72801512,
      "learning_rate": 1.3972471692539458e-06,
      "loss": 0.7501384,
      "num_input_tokens_seen": 109243620,
      "step": 5069,
      "time_per_iteration": 2.5485498905181885
    },
    {
      "auxiliary_loss_clip": 0.01276053,
      "auxiliary_loss_mlp": 0.01025908,
      "balance_loss_clip": 1.04706156,
      "balance_loss_mlp": 1.01842201,
      "epoch": 0.6096314555401912,
      "flos": 17265491187840.0,
      "grad_norm": 2.4331394185749806,
      "language_loss": 0.75345838,
      "learning_rate": 1.3965044614382348e-06,
      "loss": 0.77647799,
      "num_input_tokens_seen": 109259070,
      "step": 5070,
      "time_per_iteration": 2.6407153606414795
    },
    {
      "auxiliary_loss_clip": 0.01183915,
      "auxiliary_loss_mlp": 0.01023759,
      "balance_loss_clip": 1.05279875,
      "balance_loss_mlp": 1.01569438,
      "epoch": 0.6097516984308303,
      "flos": 21645910679040.0,
      "grad_norm": 3.1914612237330293,
      "language_loss": 0.75773156,
      "learning_rate": 1.3957618451723162e-06,
      "loss": 0.77980828,
      "num_input_tokens_seen": 109275100,
      "step": 5071,
      "time_per_iteration": 2.586203098297119
    },
    {
      "auxiliary_loss_clip": 0.01279355,
      "auxiliary_loss_mlp": 0.01032586,
      "balance_loss_clip": 1.04743218,
      "balance_loss_mlp": 1.02506638,
      "epoch": 0.6098719413214694,
      "flos": 27199208966400.0,
      "grad_norm": 2.339498541173769,
      "language_loss": 0.71637714,
      "learning_rate": 1.3950193205688457e-06,
      "loss": 0.73949659,
      "num_input_tokens_seen": 109294825,
      "step": 5072,
      "time_per_iteration": 2.799733877182007
    },
    {
      "auxiliary_loss_clip": 0.01277481,
      "auxiliary_loss_mlp": 0.01026017,
      "balance_loss_clip": 1.04876447,
      "balance_loss_mlp": 1.01918304,
      "epoch": 0.6099921842121084,
      "flos": 20412954385920.0,
      "grad_norm": 1.9241321393487125,
      "language_loss": 0.83883077,
      "learning_rate": 1.3942768877404627e-06,
      "loss": 0.86186576,
      "num_input_tokens_seen": 109313790,
      "step": 5073,
      "time_per_iteration": 2.6849329471588135
    },
    {
      "auxiliary_loss_clip": 0.01178805,
      "auxiliary_loss_mlp": 0.01022837,
      "balance_loss_clip": 1.04876351,
      "balance_loss_mlp": 1.01544023,
      "epoch": 0.6101124271027476,
      "flos": 23366139897600.0,
      "grad_norm": 1.7002385726661227,
      "language_loss": 0.73780036,
      "learning_rate": 1.393534546799795e-06,
      "loss": 0.75981677,
      "num_input_tokens_seen": 109333490,
      "step": 5074,
      "time_per_iteration": 2.6749420166015625
    },
    {
      "auxiliary_loss_clip": 0.01271104,
      "auxiliary_loss_mlp": 0.01029556,
      "balance_loss_clip": 1.04722333,
      "balance_loss_mlp": 1.02143168,
      "epoch": 0.6102326699933867,
      "flos": 26687840993280.0,
      "grad_norm": 1.6414674550216768,
      "language_loss": 0.68011057,
      "learning_rate": 1.3927922978594536e-06,
      "loss": 0.70311713,
      "num_input_tokens_seen": 109354575,
      "step": 5075,
      "time_per_iteration": 2.786829710006714
    },
    {
      "auxiliary_loss_clip": 0.01115416,
      "auxiliary_loss_mlp": 0.01002627,
      "balance_loss_clip": 1.00775933,
      "balance_loss_mlp": 1.00154257,
      "epoch": 0.6103529128840257,
      "flos": 60644612551680.0,
      "grad_norm": 0.7721370157636712,
      "language_loss": 0.57393837,
      "learning_rate": 1.3920501410320387e-06,
      "loss": 0.59511882,
      "num_input_tokens_seen": 109410690,
      "step": 5076,
      "time_per_iteration": 4.255307912826538
    },
    {
      "auxiliary_loss_clip": 0.01277201,
      "auxiliary_loss_mlp": 0.01022673,
      "balance_loss_clip": 1.04540157,
      "balance_loss_mlp": 1.01501966,
      "epoch": 0.6104731557746649,
      "flos": 19021307806080.0,
      "grad_norm": 2.6902751429848393,
      "language_loss": 0.75746793,
      "learning_rate": 1.3913080764301333e-06,
      "loss": 0.78046668,
      "num_input_tokens_seen": 109427650,
      "step": 5077,
      "time_per_iteration": 3.625964403152466
    },
    {
      "auxiliary_loss_clip": 0.01386087,
      "auxiliary_loss_mlp": 0.0103124,
      "balance_loss_clip": 1.04314232,
      "balance_loss_mlp": 1.02367306,
      "epoch": 0.6105933986653039,
      "flos": 23366894083200.0,
      "grad_norm": 1.7802188404077606,
      "language_loss": 0.71476889,
      "learning_rate": 1.3905661041663085e-06,
      "loss": 0.73894221,
      "num_input_tokens_seen": 109448835,
      "step": 5078,
      "time_per_iteration": 2.841189384460449
    },
    {
      "auxiliary_loss_clip": 0.01231068,
      "auxiliary_loss_mlp": 0.01030497,
      "balance_loss_clip": 1.05143046,
      "balance_loss_mlp": 1.02277219,
      "epoch": 0.610713641555943,
      "flos": 34637565006720.0,
      "grad_norm": 2.054643632275565,
      "language_loss": 0.64722079,
      "learning_rate": 1.389824224353122e-06,
      "loss": 0.66983646,
      "num_input_tokens_seen": 109470425,
      "step": 5079,
      "time_per_iteration": 3.659517526626587
    },
    {
      "auxiliary_loss_clip": 0.01227704,
      "auxiliary_loss_mlp": 0.01027715,
      "balance_loss_clip": 1.05134881,
      "balance_loss_mlp": 1.02026463,
      "epoch": 0.610833884446582,
      "flos": 26646471504000.0,
      "grad_norm": 1.6337457802526285,
      "language_loss": 0.76893508,
      "learning_rate": 1.389082437103115e-06,
      "loss": 0.79148924,
      "num_input_tokens_seen": 109489695,
      "step": 5080,
      "time_per_iteration": 2.769146203994751
    },
    {
      "auxiliary_loss_clip": 0.01327851,
      "auxiliary_loss_mlp": 0.01026775,
      "balance_loss_clip": 1.04477632,
      "balance_loss_mlp": 1.01913965,
      "epoch": 0.6109541273372212,
      "flos": 21215126868480.0,
      "grad_norm": 1.934015409492076,
      "language_loss": 0.78669846,
      "learning_rate": 1.3883407425288172e-06,
      "loss": 0.81024474,
      "num_input_tokens_seen": 109510030,
      "step": 5081,
      "time_per_iteration": 2.769529342651367
    },
    {
      "auxiliary_loss_clip": 0.01278341,
      "auxiliary_loss_mlp": 0.01026901,
      "balance_loss_clip": 1.04550993,
      "balance_loss_mlp": 1.01964092,
      "epoch": 0.6110743702278603,
      "flos": 20084084438400.0,
      "grad_norm": 2.8838529712910326,
      "language_loss": 0.79736185,
      "learning_rate": 1.3875991407427417e-06,
      "loss": 0.82041425,
      "num_input_tokens_seen": 109528255,
      "step": 5082,
      "time_per_iteration": 2.6802284717559814
    },
    {
      "auxiliary_loss_clip": 0.01231161,
      "auxiliary_loss_mlp": 0.01001822,
      "balance_loss_clip": 1.00893283,
      "balance_loss_mlp": 1.00076711,
      "epoch": 0.6111946131184993,
      "flos": 68302957438080.0,
      "grad_norm": 0.7831935192920694,
      "language_loss": 0.58144033,
      "learning_rate": 1.38685763185739e-06,
      "loss": 0.60377014,
      "num_input_tokens_seen": 109581915,
      "step": 5083,
      "time_per_iteration": 3.2711102962493896
    },
    {
      "auxiliary_loss_clip": 0.01177929,
      "auxiliary_loss_mlp": 0.01024386,
      "balance_loss_clip": 1.05007136,
      "balance_loss_mlp": 1.01696253,
      "epoch": 0.6113148560091385,
      "flos": 19937676602880.0,
      "grad_norm": 2.62657510639017,
      "language_loss": 0.6770398,
      "learning_rate": 1.3861162159852476e-06,
      "loss": 0.69906294,
      "num_input_tokens_seen": 109600050,
      "step": 5084,
      "time_per_iteration": 3.5393564701080322
    },
    {
      "auxiliary_loss_clip": 0.01284866,
      "auxiliary_loss_mlp": 0.01024148,
      "balance_loss_clip": 1.05095577,
      "balance_loss_mlp": 1.01670909,
      "epoch": 0.6114350988997775,
      "flos": 23731854220800.0,
      "grad_norm": 2.075124553832144,
      "language_loss": 0.79806733,
      "learning_rate": 1.3853748932387875e-06,
      "loss": 0.8211574,
      "num_input_tokens_seen": 109620690,
      "step": 5085,
      "time_per_iteration": 2.7660224437713623
    },
    {
      "auxiliary_loss_clip": 0.0127205,
      "auxiliary_loss_mlp": 0.01024575,
      "balance_loss_clip": 1.04629374,
      "balance_loss_mlp": 1.01711798,
      "epoch": 0.6115553417904166,
      "flos": 24023700224640.0,
      "grad_norm": 6.067525705034636,
      "language_loss": 0.75035137,
      "learning_rate": 1.3846336637304671e-06,
      "loss": 0.77331758,
      "num_input_tokens_seen": 109638960,
      "step": 5086,
      "time_per_iteration": 2.735663414001465
    },
    {
      "auxiliary_loss_clip": 0.01280817,
      "auxiliary_loss_mlp": 0.01024088,
      "balance_loss_clip": 1.05004752,
      "balance_loss_mlp": 1.01700687,
      "epoch": 0.6116755846810558,
      "flos": 23733542160000.0,
      "grad_norm": 2.096768264603153,
      "language_loss": 0.83327842,
      "learning_rate": 1.3838925275727316e-06,
      "loss": 0.85632747,
      "num_input_tokens_seen": 109659700,
      "step": 5087,
      "time_per_iteration": 2.8134238719940186
    },
    {
      "auxiliary_loss_clip": 0.01181002,
      "auxiliary_loss_mlp": 0.01029125,
      "balance_loss_clip": 1.05196285,
      "balance_loss_mlp": 1.02164507,
      "epoch": 0.6117958275716948,
      "flos": 18661626967680.0,
      "grad_norm": 1.7086477015624633,
      "language_loss": 0.78824651,
      "learning_rate": 1.3831514848780089e-06,
      "loss": 0.8103478,
      "num_input_tokens_seen": 109679275,
      "step": 5088,
      "time_per_iteration": 2.6036839485168457
    },
    {
      "auxiliary_loss_clip": 0.01226345,
      "auxiliary_loss_mlp": 0.01033034,
      "balance_loss_clip": 1.04780459,
      "balance_loss_mlp": 1.02561879,
      "epoch": 0.6119160704623339,
      "flos": 16471183783680.0,
      "grad_norm": 2.33555512071763,
      "language_loss": 0.91948283,
      "learning_rate": 1.3824105357587152e-06,
      "loss": 0.94207656,
      "num_input_tokens_seen": 109696380,
      "step": 5089,
      "time_per_iteration": 2.6591315269470215
    },
    {
      "auxiliary_loss_clip": 0.01271779,
      "auxiliary_loss_mlp": 0.01027277,
      "balance_loss_clip": 1.04431272,
      "balance_loss_mlp": 1.0199815,
      "epoch": 0.612036313352973,
      "flos": 23915465568000.0,
      "grad_norm": 1.734549861503467,
      "language_loss": 0.82519281,
      "learning_rate": 1.381669680327253e-06,
      "loss": 0.84818339,
      "num_input_tokens_seen": 109718060,
      "step": 5090,
      "time_per_iteration": 2.865927219390869
    },
    {
      "auxiliary_loss_clip": 0.01272866,
      "auxiliary_loss_mlp": 0.0102704,
      "balance_loss_clip": 1.04737306,
      "balance_loss_mlp": 1.019557,
      "epoch": 0.6121565562436121,
      "flos": 26974766833920.0,
      "grad_norm": 2.614178639255029,
      "language_loss": 0.7128185,
      "learning_rate": 1.380928918696008e-06,
      "loss": 0.73581755,
      "num_input_tokens_seen": 109736830,
      "step": 5091,
      "time_per_iteration": 2.6894822120666504
    },
    {
      "auxiliary_loss_clip": 0.01223882,
      "auxiliary_loss_mlp": 0.01024908,
      "balance_loss_clip": 1.04712176,
      "balance_loss_mlp": 1.01757407,
      "epoch": 0.6122767991342511,
      "flos": 15668867646720.0,
      "grad_norm": 2.3477450039779018,
      "language_loss": 0.71829361,
      "learning_rate": 1.3801882509773548e-06,
      "loss": 0.74078155,
      "num_input_tokens_seen": 109754690,
      "step": 5092,
      "time_per_iteration": 2.662302255630493
    },
    {
      "auxiliary_loss_clip": 0.01226464,
      "auxiliary_loss_mlp": 0.01029988,
      "balance_loss_clip": 1.04677665,
      "balance_loss_mlp": 1.02280021,
      "epoch": 0.6123970420248903,
      "flos": 27964321591680.0,
      "grad_norm": 1.8333544805554682,
      "language_loss": 0.81670219,
      "learning_rate": 1.3794476772836503e-06,
      "loss": 0.83926672,
      "num_input_tokens_seen": 109775790,
      "step": 5093,
      "time_per_iteration": 2.708315372467041
    },
    {
      "auxiliary_loss_clip": 0.01324966,
      "auxiliary_loss_mlp": 0.01025211,
      "balance_loss_clip": 1.04746175,
      "balance_loss_mlp": 1.01716387,
      "epoch": 0.6125172849155294,
      "flos": 21468727866240.0,
      "grad_norm": 1.6891738140571082,
      "language_loss": 0.84518534,
      "learning_rate": 1.3787071977272402e-06,
      "loss": 0.86868709,
      "num_input_tokens_seen": 109795050,
      "step": 5094,
      "time_per_iteration": 2.738004207611084
    },
    {
      "auxiliary_loss_clip": 0.01372547,
      "auxiliary_loss_mlp": 0.01025522,
      "balance_loss_clip": 1.04763174,
      "balance_loss_mlp": 1.01758575,
      "epoch": 0.6126375278061684,
      "flos": 16248321849600.0,
      "grad_norm": 2.5081761880912308,
      "language_loss": 0.72080177,
      "learning_rate": 1.3779668124204535e-06,
      "loss": 0.74478245,
      "num_input_tokens_seen": 109811465,
      "step": 5095,
      "time_per_iteration": 2.6716678142547607
    },
    {
      "auxiliary_loss_clip": 0.01271413,
      "auxiliary_loss_mlp": 0.01023229,
      "balance_loss_clip": 1.04971433,
      "balance_loss_mlp": 1.01674092,
      "epoch": 0.6127577706968076,
      "flos": 20448865008000.0,
      "grad_norm": 1.6724381783352378,
      "language_loss": 0.809461,
      "learning_rate": 1.3772265214756074e-06,
      "loss": 0.83240741,
      "num_input_tokens_seen": 109831225,
      "step": 5096,
      "time_per_iteration": 2.722831964492798
    },
    {
      "auxiliary_loss_clip": 0.01230669,
      "auxiliary_loss_mlp": 0.01029308,
      "balance_loss_clip": 1.04776382,
      "balance_loss_mlp": 1.02192628,
      "epoch": 0.6128780135874466,
      "flos": 18260397072000.0,
      "grad_norm": 1.8926896620040188,
      "language_loss": 0.75520289,
      "learning_rate": 1.3764863250050025e-06,
      "loss": 0.77780271,
      "num_input_tokens_seen": 109849465,
      "step": 5097,
      "time_per_iteration": 2.651766300201416
    },
    {
      "auxiliary_loss_clip": 0.01325393,
      "auxiliary_loss_mlp": 0.0102705,
      "balance_loss_clip": 1.04397845,
      "balance_loss_mlp": 1.02004623,
      "epoch": 0.6129982564780857,
      "flos": 24937088192640.0,
      "grad_norm": 2.0660251192747747,
      "language_loss": 0.80419147,
      "learning_rate": 1.3757462231209272e-06,
      "loss": 0.82771587,
      "num_input_tokens_seen": 109869770,
      "step": 5098,
      "time_per_iteration": 2.7281382083892822
    },
    {
      "auxiliary_loss_clip": 0.01269513,
      "auxiliary_loss_mlp": 0.01029217,
      "balance_loss_clip": 1.0448128,
      "balance_loss_mlp": 1.02142692,
      "epoch": 0.6131184993687249,
      "flos": 22492038430080.0,
      "grad_norm": 2.371458902252567,
      "language_loss": 0.88863951,
      "learning_rate": 1.3750062159356525e-06,
      "loss": 0.91162682,
      "num_input_tokens_seen": 109889120,
      "step": 5099,
      "time_per_iteration": 2.69254732131958
    },
    {
      "auxiliary_loss_clip": 0.01315366,
      "auxiliary_loss_mlp": 0.01020693,
      "balance_loss_clip": 1.04239547,
      "balance_loss_mlp": 1.01385617,
      "epoch": 0.6132387422593639,
      "flos": 15885839750400.0,
      "grad_norm": 1.8859435975993797,
      "language_loss": 0.83045381,
      "learning_rate": 1.3742663035614382e-06,
      "loss": 0.85381442,
      "num_input_tokens_seen": 109906490,
      "step": 5100,
      "time_per_iteration": 2.762586832046509
    },
    {
      "auxiliary_loss_clip": 0.01179121,
      "auxiliary_loss_mlp": 0.01029823,
      "balance_loss_clip": 1.05024481,
      "balance_loss_mlp": 1.02252102,
      "epoch": 0.613358985150003,
      "flos": 25411539962880.0,
      "grad_norm": 2.034664740934993,
      "language_loss": 0.80163676,
      "learning_rate": 1.3735264861105283e-06,
      "loss": 0.82372618,
      "num_input_tokens_seen": 109927130,
      "step": 5101,
      "time_per_iteration": 2.7184720039367676
    },
    {
      "auxiliary_loss_clip": 0.01328545,
      "auxiliary_loss_mlp": 0.0102511,
      "balance_loss_clip": 1.04625583,
      "balance_loss_mlp": 1.01807117,
      "epoch": 0.6134792280406421,
      "flos": 21361283308800.0,
      "grad_norm": 5.907338035391929,
      "language_loss": 0.78415602,
      "learning_rate": 1.372786763695152e-06,
      "loss": 0.80769253,
      "num_input_tokens_seen": 109945890,
      "step": 5102,
      "time_per_iteration": 3.7518231868743896
    },
    {
      "auxiliary_loss_clip": 0.01233429,
      "auxiliary_loss_mlp": 0.0102917,
      "balance_loss_clip": 1.05000186,
      "balance_loss_mlp": 1.0212009,
      "epoch": 0.6135994709312812,
      "flos": 21211248199680.0,
      "grad_norm": 2.443113099558924,
      "language_loss": 0.77373075,
      "learning_rate": 1.3720471364275257e-06,
      "loss": 0.79635674,
      "num_input_tokens_seen": 109965535,
      "step": 5103,
      "time_per_iteration": 3.6105997562408447
    },
    {
      "auxiliary_loss_clip": 0.01329178,
      "auxiliary_loss_mlp": 0.02568586,
      "balance_loss_clip": 1.04647756,
      "balance_loss_mlp": 0.99998236,
      "epoch": 0.6137197138219203,
      "flos": 14794047907200.0,
      "grad_norm": 2.0997370297787237,
      "language_loss": 0.78796077,
      "learning_rate": 1.3713076044198486e-06,
      "loss": 0.82693839,
      "num_input_tokens_seen": 109982345,
      "step": 5104,
      "time_per_iteration": 2.7435214519500732
    },
    {
      "auxiliary_loss_clip": 0.01274272,
      "auxiliary_loss_mlp": 0.01029668,
      "balance_loss_clip": 1.04676259,
      "balance_loss_mlp": 1.02200246,
      "epoch": 0.6138399567125594,
      "flos": 20084515401600.0,
      "grad_norm": 2.810606703306165,
      "language_loss": 0.81094229,
      "learning_rate": 1.3705681677843086e-06,
      "loss": 0.83398175,
      "num_input_tokens_seen": 110000940,
      "step": 5105,
      "time_per_iteration": 3.58648419380188
    },
    {
      "auxiliary_loss_clip": 0.01064833,
      "auxiliary_loss_mlp": 0.01000372,
      "balance_loss_clip": 1.00756872,
      "balance_loss_mlp": 0.9993704,
      "epoch": 0.6139601996031985,
      "flos": 60123838193280.0,
      "grad_norm": 0.7936162611209164,
      "language_loss": 0.60491866,
      "learning_rate": 1.3698288266330768e-06,
      "loss": 0.62557071,
      "num_input_tokens_seen": 110061565,
      "step": 5106,
      "time_per_iteration": 3.2810003757476807
    },
    {
      "auxiliary_loss_clip": 0.01275077,
      "auxiliary_loss_mlp": 0.01022894,
      "balance_loss_clip": 1.05172014,
      "balance_loss_mlp": 1.01603329,
      "epoch": 0.6140804424938375,
      "flos": 23586703361280.0,
      "grad_norm": 2.1929564730477136,
      "language_loss": 0.72329116,
      "learning_rate": 1.3690895810783113e-06,
      "loss": 0.74627084,
      "num_input_tokens_seen": 110080360,
      "step": 5107,
      "time_per_iteration": 2.776644468307495
    },
    {
      "auxiliary_loss_clip": 0.01329843,
      "auxiliary_loss_mlp": 0.02567217,
      "balance_loss_clip": 1.04040289,
      "balance_loss_mlp": 0.9999795,
      "epoch": 0.6142006853844767,
      "flos": 21398199511680.0,
      "grad_norm": 2.2964756821147727,
      "language_loss": 0.71614981,
      "learning_rate": 1.3683504312321543e-06,
      "loss": 0.7551204,
      "num_input_tokens_seen": 110100695,
      "step": 5108,
      "time_per_iteration": 2.9300575256347656
    },
    {
      "auxiliary_loss_clip": 0.01234873,
      "auxiliary_loss_mlp": 0.01022743,
      "balance_loss_clip": 1.05085647,
      "balance_loss_mlp": 1.01547468,
      "epoch": 0.6143209282751158,
      "flos": 12057367622400.0,
      "grad_norm": 1.9803998944057724,
      "language_loss": 0.80130732,
      "learning_rate": 1.3676113772067355e-06,
      "loss": 0.82388353,
      "num_input_tokens_seen": 110117750,
      "step": 5109,
      "time_per_iteration": 2.9907724857330322
    },
    {
      "auxiliary_loss_clip": 0.01383491,
      "auxiliary_loss_mlp": 0.01028166,
      "balance_loss_clip": 1.04642177,
      "balance_loss_mlp": 1.02039361,
      "epoch": 0.6144411711657548,
      "flos": 25082274965760.0,
      "grad_norm": 1.961649429490757,
      "language_loss": 0.72380131,
      "learning_rate": 1.3668724191141671e-06,
      "loss": 0.74791789,
      "num_input_tokens_seen": 110137020,
      "step": 5110,
      "time_per_iteration": 3.729192018508911
    },
    {
      "auxiliary_loss_clip": 0.01329212,
      "auxiliary_loss_mlp": 0.01032072,
      "balance_loss_clip": 1.05345905,
      "balance_loss_mlp": 1.02488422,
      "epoch": 0.6145614140563939,
      "flos": 20114069316480.0,
      "grad_norm": 2.4881895040003394,
      "language_loss": 0.66502643,
      "learning_rate": 1.3661335570665493e-06,
      "loss": 0.68863928,
      "num_input_tokens_seen": 110154930,
      "step": 5111,
      "time_per_iteration": 2.7587826251983643
    },
    {
      "auxiliary_loss_clip": 0.01283091,
      "auxiliary_loss_mlp": 0.01027299,
      "balance_loss_clip": 1.05135393,
      "balance_loss_mlp": 1.02039075,
      "epoch": 0.614681656947033,
      "flos": 16800376953600.0,
      "grad_norm": 2.6182252829555157,
      "language_loss": 0.69748318,
      "learning_rate": 1.3653947911759676e-06,
      "loss": 0.72058713,
      "num_input_tokens_seen": 110172480,
      "step": 5112,
      "time_per_iteration": 2.7654523849487305
    },
    {
      "auxiliary_loss_clip": 0.01373038,
      "auxiliary_loss_mlp": 0.01032129,
      "balance_loss_clip": 1.04583406,
      "balance_loss_mlp": 1.02497029,
      "epoch": 0.6148018998376721,
      "flos": 38801587011840.0,
      "grad_norm": 1.6807348041616355,
      "language_loss": 0.74602413,
      "learning_rate": 1.3646561215544904e-06,
      "loss": 0.7700758,
      "num_input_tokens_seen": 110197120,
      "step": 5113,
      "time_per_iteration": 2.9578561782836914
    },
    {
      "auxiliary_loss_clip": 0.01227424,
      "auxiliary_loss_mlp": 0.01024866,
      "balance_loss_clip": 1.05005574,
      "balance_loss_mlp": 1.01743889,
      "epoch": 0.6149221427283111,
      "flos": 23327032965120.0,
      "grad_norm": 2.3323324350479084,
      "language_loss": 0.79660273,
      "learning_rate": 1.363917548314176e-06,
      "loss": 0.81912565,
      "num_input_tokens_seen": 110216385,
      "step": 5114,
      "time_per_iteration": 2.775977373123169
    },
    {
      "auxiliary_loss_clip": 0.01134192,
      "auxiliary_loss_mlp": 0.01034406,
      "balance_loss_clip": 1.04886377,
      "balance_loss_mlp": 1.02700651,
      "epoch": 0.6150423856189503,
      "flos": 22379494141440.0,
      "grad_norm": 1.7972440066675273,
      "language_loss": 0.72958148,
      "learning_rate": 1.3631790715670626e-06,
      "loss": 0.75126743,
      "num_input_tokens_seen": 110234790,
      "step": 5115,
      "time_per_iteration": 2.7443110942840576
    },
    {
      "auxiliary_loss_clip": 0.01463147,
      "auxiliary_loss_mlp": 0.01023862,
      "balance_loss_clip": 1.04335523,
      "balance_loss_mlp": 1.01720452,
      "epoch": 0.6151626285095894,
      "flos": 18692078722560.0,
      "grad_norm": 2.173761382545158,
      "language_loss": 0.85850108,
      "learning_rate": 1.3624406914251783e-06,
      "loss": 0.88337123,
      "num_input_tokens_seen": 110251910,
      "step": 5116,
      "time_per_iteration": 2.9799575805664062
    },
    {
      "auxiliary_loss_clip": 0.01228751,
      "auxiliary_loss_mlp": 0.01023522,
      "balance_loss_clip": 1.04750562,
      "balance_loss_mlp": 1.01681685,
      "epoch": 0.6152828714002284,
      "flos": 15851688894720.0,
      "grad_norm": 1.8983580874208135,
      "language_loss": 0.88651681,
      "learning_rate": 1.3617024080005335e-06,
      "loss": 0.9090395,
      "num_input_tokens_seen": 110268810,
      "step": 5117,
      "time_per_iteration": 2.7704038619995117
    },
    {
      "auxiliary_loss_clip": 0.01178742,
      "auxiliary_loss_mlp": 0.02569393,
      "balance_loss_clip": 1.04594541,
      "balance_loss_mlp": 0.99995601,
      "epoch": 0.6154031142908676,
      "flos": 24869792062080.0,
      "grad_norm": 1.6271135355913813,
      "language_loss": 0.74328035,
      "learning_rate": 1.3609642214051266e-06,
      "loss": 0.78076166,
      "num_input_tokens_seen": 110293035,
      "step": 5118,
      "time_per_iteration": 2.8735740184783936
    },
    {
      "auxiliary_loss_clip": 0.01281828,
      "auxiliary_loss_mlp": 0.0102717,
      "balance_loss_clip": 1.0529449,
      "balance_loss_mlp": 1.01941574,
      "epoch": 0.6155233571815066,
      "flos": 19244744357760.0,
      "grad_norm": 1.8610784354118166,
      "language_loss": 0.66343129,
      "learning_rate": 1.3602261317509385e-06,
      "loss": 0.68652129,
      "num_input_tokens_seen": 110309695,
      "step": 5119,
      "time_per_iteration": 2.745975971221924
    },
    {
      "auxiliary_loss_clip": 0.01230423,
      "auxiliary_loss_mlp": 0.01022505,
      "balance_loss_clip": 1.05054808,
      "balance_loss_mlp": 1.01517069,
      "epoch": 0.6156436000721457,
      "flos": 18770077105920.0,
      "grad_norm": 3.1905826166171742,
      "language_loss": 0.82967532,
      "learning_rate": 1.3594881391499387e-06,
      "loss": 0.85220456,
      "num_input_tokens_seen": 110328610,
      "step": 5120,
      "time_per_iteration": 2.6698200702667236
    },
    {
      "auxiliary_loss_clip": 0.01277466,
      "auxiliary_loss_mlp": 0.01028885,
      "balance_loss_clip": 1.04678631,
      "balance_loss_mlp": 1.02164936,
      "epoch": 0.6157638429627849,
      "flos": 18041198325120.0,
      "grad_norm": 1.8514314356588815,
      "language_loss": 0.79604226,
      "learning_rate": 1.3587502437140778e-06,
      "loss": 0.81910574,
      "num_input_tokens_seen": 110346775,
      "step": 5121,
      "time_per_iteration": 2.7267770767211914
    },
    {
      "auxiliary_loss_clip": 0.01281474,
      "auxiliary_loss_mlp": 0.01031836,
      "balance_loss_clip": 1.04691994,
      "balance_loss_mlp": 1.02471876,
      "epoch": 0.6158840858534239,
      "flos": 25556726736000.0,
      "grad_norm": 2.5099835089305578,
      "language_loss": 0.85009122,
      "learning_rate": 1.3580124455552952e-06,
      "loss": 0.87322426,
      "num_input_tokens_seen": 110366140,
      "step": 5122,
      "time_per_iteration": 2.8027498722076416
    },
    {
      "auxiliary_loss_clip": 0.01226615,
      "auxiliary_loss_mlp": 0.02563958,
      "balance_loss_clip": 1.04950523,
      "balance_loss_mlp": 0.99996656,
      "epoch": 0.616004328744063,
      "flos": 24640788902400.0,
      "grad_norm": 1.7755147425596598,
      "language_loss": 0.8731975,
      "learning_rate": 1.3572747447855148e-06,
      "loss": 0.91110319,
      "num_input_tokens_seen": 110386550,
      "step": 5123,
      "time_per_iteration": 2.760728597640991
    },
    {
      "auxiliary_loss_clip": 0.01183512,
      "auxiliary_loss_mlp": 0.01029246,
      "balance_loss_clip": 1.05328405,
      "balance_loss_mlp": 1.02139604,
      "epoch": 0.6161245716347021,
      "flos": 21689686379520.0,
      "grad_norm": 1.8922802823042264,
      "language_loss": 0.69351304,
      "learning_rate": 1.356537141516644e-06,
      "loss": 0.71564054,
      "num_input_tokens_seen": 110403970,
      "step": 5124,
      "time_per_iteration": 2.6115829944610596
    },
    {
      "auxiliary_loss_clip": 0.01226506,
      "auxiliary_loss_mlp": 0.01028682,
      "balance_loss_clip": 1.05128622,
      "balance_loss_mlp": 1.02142787,
      "epoch": 0.6162448145253412,
      "flos": 35189225061120.0,
      "grad_norm": 2.489054613470497,
      "language_loss": 0.61631799,
      "learning_rate": 1.3557996358605775e-06,
      "loss": 0.63886982,
      "num_input_tokens_seen": 110423890,
      "step": 5125,
      "time_per_iteration": 2.819504737854004
    },
    {
      "auxiliary_loss_clip": 0.01228565,
      "auxiliary_loss_mlp": 0.01026568,
      "balance_loss_clip": 1.0485934,
      "balance_loss_mlp": 1.01973724,
      "epoch": 0.6163650574159802,
      "flos": 21615279356160.0,
      "grad_norm": 2.2466334089035707,
      "language_loss": 0.70352024,
      "learning_rate": 1.3550622279291941e-06,
      "loss": 0.7260716,
      "num_input_tokens_seen": 110442035,
      "step": 5126,
      "time_per_iteration": 2.672429323196411
    },
    {
      "auxiliary_loss_clip": 0.01367066,
      "auxiliary_loss_mlp": 0.01024251,
      "balance_loss_clip": 1.04225373,
      "balance_loss_mlp": 1.01709008,
      "epoch": 0.6164853003066194,
      "flos": 24572163968640.0,
      "grad_norm": 1.418581236876818,
      "language_loss": 0.83451629,
      "learning_rate": 1.354324917834358e-06,
      "loss": 0.85842949,
      "num_input_tokens_seen": 110463280,
      "step": 5127,
      "time_per_iteration": 2.848684310913086
    },
    {
      "auxiliary_loss_clip": 0.01426828,
      "auxiliary_loss_mlp": 0.02567051,
      "balance_loss_clip": 1.04590964,
      "balance_loss_mlp": 0.99992347,
      "epoch": 0.6166055431972585,
      "flos": 21835986474240.0,
      "grad_norm": 3.5088033132772725,
      "language_loss": 0.77156854,
      "learning_rate": 1.353587705687918e-06,
      "loss": 0.81150734,
      "num_input_tokens_seen": 110481455,
      "step": 5128,
      "time_per_iteration": 4.22551703453064
    },
    {
      "auxiliary_loss_clip": 0.01287001,
      "auxiliary_loss_mlp": 0.01032403,
      "balance_loss_clip": 1.05263257,
      "balance_loss_mlp": 1.02402818,
      "epoch": 0.6167257860878975,
      "flos": 17785262943360.0,
      "grad_norm": 7.130138144029775,
      "language_loss": 0.72627622,
      "learning_rate": 1.3528505916017096e-06,
      "loss": 0.74947023,
      "num_input_tokens_seen": 110499155,
      "step": 5129,
      "time_per_iteration": 3.6219944953918457
    },
    {
      "auxiliary_loss_clip": 0.01230193,
      "auxiliary_loss_mlp": 0.01024439,
      "balance_loss_clip": 1.04780316,
      "balance_loss_mlp": 1.01704192,
      "epoch": 0.6168460289785367,
      "flos": 23214811898880.0,
      "grad_norm": 2.0887777051120198,
      "language_loss": 0.88790703,
      "learning_rate": 1.3521135756875514e-06,
      "loss": 0.91045332,
      "num_input_tokens_seen": 110515470,
      "step": 5130,
      "time_per_iteration": 2.7215476036071777
    },
    {
      "auxiliary_loss_clip": 0.01419657,
      "auxiliary_loss_mlp": 0.01022724,
      "balance_loss_clip": 1.04278588,
      "balance_loss_mlp": 1.01589656,
      "epoch": 0.6169662718691757,
      "flos": 26213281482240.0,
      "grad_norm": 1.4668499678122036,
      "language_loss": 0.86190706,
      "learning_rate": 1.3513766580572496e-06,
      "loss": 0.88633084,
      "num_input_tokens_seen": 110538290,
      "step": 5131,
      "time_per_iteration": 3.7690961360931396
    },
    {
      "auxiliary_loss_clip": 0.01229049,
      "auxiliary_loss_mlp": 0.01027295,
      "balance_loss_clip": 1.05060124,
      "balance_loss_mlp": 1.02122188,
      "epoch": 0.6170865147598148,
      "flos": 19026120228480.0,
      "grad_norm": 2.1272319484678173,
      "language_loss": 0.76946324,
      "learning_rate": 1.3506398388225924e-06,
      "loss": 0.79202664,
      "num_input_tokens_seen": 110555610,
      "step": 5132,
      "time_per_iteration": 2.655075788497925
    },
    {
      "auxiliary_loss_clip": 0.011782,
      "auxiliary_loss_mlp": 0.01028956,
      "balance_loss_clip": 1.05032694,
      "balance_loss_mlp": 1.02223277,
      "epoch": 0.617206757650454,
      "flos": 18260361158400.0,
      "grad_norm": 2.0125856185551663,
      "language_loss": 0.72027028,
      "learning_rate": 1.349903118095355e-06,
      "loss": 0.74234188,
      "num_input_tokens_seen": 110574745,
      "step": 5133,
      "time_per_iteration": 2.636082887649536
    },
    {
      "auxiliary_loss_clip": 0.01233631,
      "auxiliary_loss_mlp": 0.01025224,
      "balance_loss_clip": 1.05049384,
      "balance_loss_mlp": 1.01786828,
      "epoch": 0.617327000541093,
      "flos": 18186959715840.0,
      "grad_norm": 2.0770400971114977,
      "language_loss": 0.73945928,
      "learning_rate": 1.349166495987298e-06,
      "loss": 0.76204783,
      "num_input_tokens_seen": 110593310,
      "step": 5134,
      "time_per_iteration": 2.6061668395996094
    },
    {
      "auxiliary_loss_clip": 0.01192672,
      "auxiliary_loss_mlp": 0.00998517,
      "balance_loss_clip": 1.02265882,
      "balance_loss_mlp": 0.99740809,
      "epoch": 0.6174472434317321,
      "flos": 61833796122240.0,
      "grad_norm": 0.8196722587977189,
      "language_loss": 0.60813844,
      "learning_rate": 1.348429972610166e-06,
      "loss": 0.6300503,
      "num_input_tokens_seen": 110657615,
      "step": 5135,
      "time_per_iteration": 3.3039610385894775
    },
    {
      "auxiliary_loss_clip": 0.01288837,
      "auxiliary_loss_mlp": 0.00999255,
      "balance_loss_clip": 1.02446675,
      "balance_loss_mlp": 0.99816382,
      "epoch": 0.6175674863223712,
      "flos": 71230970494080.0,
      "grad_norm": 0.8437007096204273,
      "language_loss": 0.57816678,
      "learning_rate": 1.3476935480756897e-06,
      "loss": 0.60104769,
      "num_input_tokens_seen": 110714365,
      "step": 5136,
      "time_per_iteration": 4.062406063079834
    },
    {
      "auxiliary_loss_clip": 0.01326372,
      "auxiliary_loss_mlp": 0.01025083,
      "balance_loss_clip": 1.04460621,
      "balance_loss_mlp": 1.0175786,
      "epoch": 0.6176877292130103,
      "flos": 21835447770240.0,
      "grad_norm": 2.3417702857954237,
      "language_loss": 0.7524969,
      "learning_rate": 1.346957222495583e-06,
      "loss": 0.77601147,
      "num_input_tokens_seen": 110732160,
      "step": 5137,
      "time_per_iteration": 2.724106788635254
    },
    {
      "auxiliary_loss_clip": 0.01183666,
      "auxiliary_loss_mlp": 0.02568968,
      "balance_loss_clip": 1.05090725,
      "balance_loss_mlp": 0.99993706,
      "epoch": 0.6178079721036493,
      "flos": 17741738638080.0,
      "grad_norm": 2.5238112744606442,
      "language_loss": 0.71018457,
      "learning_rate": 1.3462209959815466e-06,
      "loss": 0.74771094,
      "num_input_tokens_seen": 110746900,
      "step": 5138,
      "time_per_iteration": 2.682579755783081
    },
    {
      "auxiliary_loss_clip": 0.01279668,
      "auxiliary_loss_mlp": 0.01025857,
      "balance_loss_clip": 1.04887605,
      "balance_loss_mlp": 1.01882362,
      "epoch": 0.6179282149942885,
      "flos": 22633131052800.0,
      "grad_norm": 2.214197109132397,
      "language_loss": 0.74303365,
      "learning_rate": 1.345484868645265e-06,
      "loss": 0.76608884,
      "num_input_tokens_seen": 110765710,
      "step": 5139,
      "time_per_iteration": 2.707698106765747
    },
    {
      "auxiliary_loss_clip": 0.01232497,
      "auxiliary_loss_mlp": 0.01026158,
      "balance_loss_clip": 1.04474545,
      "balance_loss_mlp": 1.01869297,
      "epoch": 0.6180484578849276,
      "flos": 22310330503680.0,
      "grad_norm": 1.935779881278282,
      "language_loss": 0.78419697,
      "learning_rate": 1.3447488405984088e-06,
      "loss": 0.80678356,
      "num_input_tokens_seen": 110783970,
      "step": 5140,
      "time_per_iteration": 2.79646372795105
    },
    {
      "auxiliary_loss_clip": 0.01277792,
      "auxiliary_loss_mlp": 0.01025788,
      "balance_loss_clip": 1.04856277,
      "balance_loss_mlp": 1.01838207,
      "epoch": 0.6181687007755666,
      "flos": 35225458905600.0,
      "grad_norm": 2.5057163566772243,
      "language_loss": 0.70282161,
      "learning_rate": 1.3440129119526322e-06,
      "loss": 0.72585738,
      "num_input_tokens_seen": 110806395,
      "step": 5141,
      "time_per_iteration": 2.8213860988616943
    },
    {
      "auxiliary_loss_clip": 0.01065416,
      "auxiliary_loss_mlp": 0.01002901,
      "balance_loss_clip": 1.00838947,
      "balance_loss_mlp": 1.00191772,
      "epoch": 0.6182889436662057,
      "flos": 61547370094080.0,
      "grad_norm": 0.80676825491449,
      "language_loss": 0.51128531,
      "learning_rate": 1.3432770828195762e-06,
      "loss": 0.53196847,
      "num_input_tokens_seen": 110867380,
      "step": 5142,
      "time_per_iteration": 3.346045732498169
    },
    {
      "auxiliary_loss_clip": 0.01323207,
      "auxiliary_loss_mlp": 0.01027944,
      "balance_loss_clip": 1.04455066,
      "balance_loss_mlp": 1.02054143,
      "epoch": 0.6184091865568448,
      "flos": 19609991804160.0,
      "grad_norm": 2.465603132389759,
      "language_loss": 0.70284605,
      "learning_rate": 1.3425413533108635e-06,
      "loss": 0.72635758,
      "num_input_tokens_seen": 110885980,
      "step": 5143,
      "time_per_iteration": 2.7806894779205322
    },
    {
      "auxiliary_loss_clip": 0.0137812,
      "auxiliary_loss_mlp": 0.01029738,
      "balance_loss_clip": 1.04899263,
      "balance_loss_mlp": 1.02169752,
      "epoch": 0.6185294294474839,
      "flos": 23586882929280.0,
      "grad_norm": 3.5990455474269822,
      "language_loss": 0.70301932,
      "learning_rate": 1.341805723538105e-06,
      "loss": 0.72709787,
      "num_input_tokens_seen": 110906085,
      "step": 5144,
      "time_per_iteration": 2.7901458740234375
    },
    {
      "auxiliary_loss_clip": 0.01185658,
      "auxiliary_loss_mlp": 0.01028904,
      "balance_loss_clip": 1.04879022,
      "balance_loss_mlp": 1.02215934,
      "epoch": 0.618649672338123,
      "flos": 26762032535040.0,
      "grad_norm": 1.5945778671009538,
      "language_loss": 0.7723093,
      "learning_rate": 1.3410701936128948e-06,
      "loss": 0.79445493,
      "num_input_tokens_seen": 110928865,
      "step": 5145,
      "time_per_iteration": 2.756326675415039
    },
    {
      "auxiliary_loss_clip": 0.01228628,
      "auxiliary_loss_mlp": 0.01025341,
      "balance_loss_clip": 1.05052125,
      "balance_loss_mlp": 1.01820028,
      "epoch": 0.6187699152287621,
      "flos": 14456630522880.0,
      "grad_norm": 2.40433890243612,
      "language_loss": 0.85429478,
      "learning_rate": 1.340334763646812e-06,
      "loss": 0.87683445,
      "num_input_tokens_seen": 110943000,
      "step": 5146,
      "time_per_iteration": 2.615948438644409
    },
    {
      "auxiliary_loss_clip": 0.01181811,
      "auxiliary_loss_mlp": 0.01026074,
      "balance_loss_clip": 1.05045247,
      "balance_loss_mlp": 1.01834655,
      "epoch": 0.6188901581194012,
      "flos": 20084766796800.0,
      "grad_norm": 1.8257583671502058,
      "language_loss": 0.74144173,
      "learning_rate": 1.3395994337514218e-06,
      "loss": 0.7635206,
      "num_input_tokens_seen": 110963170,
      "step": 5147,
      "time_per_iteration": 2.673292636871338
    },
    {
      "auxiliary_loss_clip": 0.0122238,
      "auxiliary_loss_mlp": 0.0102744,
      "balance_loss_clip": 1.04573131,
      "balance_loss_mlp": 1.01998377,
      "epoch": 0.6190104010100402,
      "flos": 25700728360320.0,
      "grad_norm": 1.9065971433094664,
      "language_loss": 0.78674018,
      "learning_rate": 1.3388642040382725e-06,
      "loss": 0.80923843,
      "num_input_tokens_seen": 110983595,
      "step": 5148,
      "time_per_iteration": 2.710770845413208
    },
    {
      "auxiliary_loss_clip": 0.0133179,
      "auxiliary_loss_mlp": 0.0102733,
      "balance_loss_clip": 1.04314017,
      "balance_loss_mlp": 1.02051687,
      "epoch": 0.6191306439006794,
      "flos": 30442372974720.0,
      "grad_norm": 1.6437833153464387,
      "language_loss": 0.83941132,
      "learning_rate": 1.3381290746188975e-06,
      "loss": 0.86300254,
      "num_input_tokens_seen": 111002965,
      "step": 5149,
      "time_per_iteration": 2.810241222381592
    },
    {
      "auxiliary_loss_clip": 0.01230633,
      "auxiliary_loss_mlp": 0.01033657,
      "balance_loss_clip": 1.05348873,
      "balance_loss_mlp": 1.02665639,
      "epoch": 0.6192508867913185,
      "flos": 26685793918080.0,
      "grad_norm": 1.6715343022056415,
      "language_loss": 0.67448115,
      "learning_rate": 1.3373940456048152e-06,
      "loss": 0.69712406,
      "num_input_tokens_seen": 111022990,
      "step": 5150,
      "time_per_iteration": 2.7225418090820312
    },
    {
      "auxiliary_loss_clip": 0.01178223,
      "auxiliary_loss_mlp": 0.01026926,
      "balance_loss_clip": 1.05042207,
      "balance_loss_mlp": 1.01942778,
      "epoch": 0.6193711296819575,
      "flos": 36722036090880.0,
      "grad_norm": 1.6128307554999184,
      "language_loss": 0.59127796,
      "learning_rate": 1.3366591171075299e-06,
      "loss": 0.61332947,
      "num_input_tokens_seen": 111046495,
      "step": 5151,
      "time_per_iteration": 2.733884572982788
    },
    {
      "auxiliary_loss_clip": 0.01278503,
      "auxiliary_loss_mlp": 0.01028217,
      "balance_loss_clip": 1.05011201,
      "balance_loss_mlp": 1.02092481,
      "epoch": 0.6194913725725967,
      "flos": 25192556697600.0,
      "grad_norm": 1.9990743827208126,
      "language_loss": 0.91105294,
      "learning_rate": 1.335924289238529e-06,
      "loss": 0.93412024,
      "num_input_tokens_seen": 111065705,
      "step": 5152,
      "time_per_iteration": 2.7388012409210205
    },
    {
      "auxiliary_loss_clip": 0.01236512,
      "auxiliary_loss_mlp": 0.02568971,
      "balance_loss_clip": 1.05523515,
      "balance_loss_mlp": 0.99996525,
      "epoch": 0.6196116154632357,
      "flos": 21178821196800.0,
      "grad_norm": 1.6877600453122135,
      "language_loss": 0.76779288,
      "learning_rate": 1.3351895621092859e-06,
      "loss": 0.80584776,
      "num_input_tokens_seen": 111086050,
      "step": 5153,
      "time_per_iteration": 2.6688716411590576
    },
    {
      "auxiliary_loss_clip": 0.01524993,
      "auxiliary_loss_mlp": 0.01026513,
      "balance_loss_clip": 1.03589749,
      "balance_loss_mlp": 1.01842499,
      "epoch": 0.6197318583538748,
      "flos": 16253744803200.0,
      "grad_norm": 1.8058373888521724,
      "language_loss": 0.76798999,
      "learning_rate": 1.3344549358312567e-06,
      "loss": 0.79350507,
      "num_input_tokens_seen": 111104450,
      "step": 5154,
      "time_per_iteration": 4.180957794189453
    },
    {
      "auxiliary_loss_clip": 0.01231941,
      "auxiliary_loss_mlp": 0.01036059,
      "balance_loss_clip": 1.05124617,
      "balance_loss_mlp": 1.02851343,
      "epoch": 0.619852101244514,
      "flos": 24425612478720.0,
      "grad_norm": 2.6121029146967607,
      "language_loss": 0.78211188,
      "learning_rate": 1.3337204105158852e-06,
      "loss": 0.80479193,
      "num_input_tokens_seen": 111123320,
      "step": 5155,
      "time_per_iteration": 3.913201332092285
    },
    {
      "auxiliary_loss_clip": 0.0131982,
      "auxiliary_loss_mlp": 0.01032735,
      "balance_loss_clip": 1.03885245,
      "balance_loss_mlp": 1.0253861,
      "epoch": 0.619972344135153,
      "flos": 16727298733440.0,
      "grad_norm": 1.8835582428471784,
      "language_loss": 0.72564393,
      "learning_rate": 1.332985986274597e-06,
      "loss": 0.74916947,
      "num_input_tokens_seen": 111140950,
      "step": 5156,
      "time_per_iteration": 2.717623710632324
    },
    {
      "auxiliary_loss_clip": 0.01421813,
      "auxiliary_loss_mlp": 0.02565676,
      "balance_loss_clip": 1.0468595,
      "balance_loss_mlp": 0.99998248,
      "epoch": 0.6200925870257921,
      "flos": 12495190498560.0,
      "grad_norm": 2.0585769288183875,
      "language_loss": 0.75411177,
      "learning_rate": 1.3322516632188047e-06,
      "loss": 0.79398668,
      "num_input_tokens_seen": 111157845,
      "step": 5157,
      "time_per_iteration": 3.928401231765747
    },
    {
      "auxiliary_loss_clip": 0.01323846,
      "auxiliary_loss_mlp": 0.01023678,
      "balance_loss_clip": 1.04470432,
      "balance_loss_mlp": 1.01616192,
      "epoch": 0.6202128299164312,
      "flos": 26539350168960.0,
      "grad_norm": 3.890564585720792,
      "language_loss": 0.66819715,
      "learning_rate": 1.3315174414599045e-06,
      "loss": 0.69167244,
      "num_input_tokens_seen": 111179165,
      "step": 5158,
      "time_per_iteration": 2.7968170642852783
    },
    {
      "auxiliary_loss_clip": 0.01226453,
      "auxiliary_loss_mlp": 0.01034148,
      "balance_loss_clip": 1.04640496,
      "balance_loss_mlp": 1.02655995,
      "epoch": 0.6203330728070703,
      "flos": 18770508069120.0,
      "grad_norm": 1.7355635331986912,
      "language_loss": 0.75344729,
      "learning_rate": 1.3307833211092768e-06,
      "loss": 0.77605325,
      "num_input_tokens_seen": 111197830,
      "step": 5159,
      "time_per_iteration": 2.618452310562134
    },
    {
      "auxiliary_loss_clip": 0.01183503,
      "auxiliary_loss_mlp": 0.01028034,
      "balance_loss_clip": 1.05455685,
      "balance_loss_mlp": 1.0206579,
      "epoch": 0.6204533156977093,
      "flos": 20629782835200.0,
      "grad_norm": 1.7041583296043412,
      "language_loss": 0.75317687,
      "learning_rate": 1.3300493022782873e-06,
      "loss": 0.77529222,
      "num_input_tokens_seen": 111218400,
      "step": 5160,
      "time_per_iteration": 2.708855628967285
    },
    {
      "auxiliary_loss_clip": 0.01368004,
      "auxiliary_loss_mlp": 0.02572247,
      "balance_loss_clip": 1.04322314,
      "balance_loss_mlp": 0.99998617,
      "epoch": 0.6205735585883485,
      "flos": 17348050598400.0,
      "grad_norm": 2.887793092692339,
      "language_loss": 0.72628164,
      "learning_rate": 1.3293153850782855e-06,
      "loss": 0.76568413,
      "num_input_tokens_seen": 111236720,
      "step": 5161,
      "time_per_iteration": 2.709351062774658
    },
    {
      "auxiliary_loss_clip": 0.01321851,
      "auxiliary_loss_mlp": 0.0102822,
      "balance_loss_clip": 1.04454434,
      "balance_loss_mlp": 1.02023554,
      "epoch": 0.6206938014789876,
      "flos": 22965017742720.0,
      "grad_norm": 2.213024311017882,
      "language_loss": 0.71471488,
      "learning_rate": 1.3285815696206069e-06,
      "loss": 0.73821557,
      "num_input_tokens_seen": 111258265,
      "step": 5162,
      "time_per_iteration": 3.8434739112854004
    },
    {
      "auxiliary_loss_clip": 0.01329123,
      "auxiliary_loss_mlp": 0.01028307,
      "balance_loss_clip": 1.04506755,
      "balance_loss_mlp": 1.02085328,
      "epoch": 0.6208140443696266,
      "flos": 23983192661760.0,
      "grad_norm": 1.9361157871888601,
      "language_loss": 0.76934248,
      "learning_rate": 1.32784785601657e-06,
      "loss": 0.79291677,
      "num_input_tokens_seen": 111277675,
      "step": 5163,
      "time_per_iteration": 2.7412614822387695
    },
    {
      "auxiliary_loss_clip": 0.01282134,
      "auxiliary_loss_mlp": 0.01024814,
      "balance_loss_clip": 1.04841113,
      "balance_loss_mlp": 1.01748872,
      "epoch": 0.6209342872602658,
      "flos": 35077291303680.0,
      "grad_norm": 2.098074174440486,
      "language_loss": 0.7388038,
      "learning_rate": 1.3271142443774798e-06,
      "loss": 0.76187336,
      "num_input_tokens_seen": 111299910,
      "step": 5164,
      "time_per_iteration": 2.7375049591064453
    },
    {
      "auxiliary_loss_clip": 0.01273092,
      "auxiliary_loss_mlp": 0.01022584,
      "balance_loss_clip": 1.0490433,
      "balance_loss_mlp": 1.0155623,
      "epoch": 0.6210545301509048,
      "flos": 26979327861120.0,
      "grad_norm": 2.1665871642082313,
      "language_loss": 0.81863391,
      "learning_rate": 1.3263807348146228e-06,
      "loss": 0.84159064,
      "num_input_tokens_seen": 111319765,
      "step": 5165,
      "time_per_iteration": 2.6797335147857666
    },
    {
      "auxiliary_loss_clip": 0.01277095,
      "auxiliary_loss_mlp": 0.01025248,
      "balance_loss_clip": 1.0446341,
      "balance_loss_mlp": 1.0168736,
      "epoch": 0.6211747730415439,
      "flos": 33618240852480.0,
      "grad_norm": 1.8697945923739856,
      "language_loss": 0.73497301,
      "learning_rate": 1.3256473274392733e-06,
      "loss": 0.75799644,
      "num_input_tokens_seen": 111341110,
      "step": 5166,
      "time_per_iteration": 2.6757049560546875
    },
    {
      "auxiliary_loss_clip": 0.0118123,
      "auxiliary_loss_mlp": 0.01027196,
      "balance_loss_clip": 1.05261362,
      "balance_loss_mlp": 1.01923931,
      "epoch": 0.6212950159321831,
      "flos": 34167099646080.0,
      "grad_norm": 2.242564901148075,
      "language_loss": 0.70174891,
      "learning_rate": 1.3249140223626873e-06,
      "loss": 0.7238332,
      "num_input_tokens_seen": 111362730,
      "step": 5167,
      "time_per_iteration": 2.608325481414795
    },
    {
      "auxiliary_loss_clip": 0.01225788,
      "auxiliary_loss_mlp": 0.01024809,
      "balance_loss_clip": 1.05050254,
      "balance_loss_mlp": 1.01771951,
      "epoch": 0.6214152588228221,
      "flos": 27965758135680.0,
      "grad_norm": 2.3505204332417717,
      "language_loss": 0.7547614,
      "learning_rate": 1.3241808196961077e-06,
      "loss": 0.77726734,
      "num_input_tokens_seen": 111383855,
      "step": 5168,
      "time_per_iteration": 2.597623348236084
    },
    {
      "auxiliary_loss_clip": 0.01270181,
      "auxiliary_loss_mlp": 0.01028354,
      "balance_loss_clip": 1.04712796,
      "balance_loss_mlp": 1.02160692,
      "epoch": 0.6215355017134612,
      "flos": 20230204965120.0,
      "grad_norm": 1.8451329640585628,
      "language_loss": 0.70896196,
      "learning_rate": 1.3234477195507608e-06,
      "loss": 0.7319473,
      "num_input_tokens_seen": 111402685,
      "step": 5169,
      "time_per_iteration": 2.7126755714416504
    },
    {
      "auxiliary_loss_clip": 0.01327595,
      "auxiliary_loss_mlp": 0.01025741,
      "balance_loss_clip": 1.05054879,
      "balance_loss_mlp": 1.01820374,
      "epoch": 0.6216557446041003,
      "flos": 41428129219200.0,
      "grad_norm": 1.8530362377880827,
      "language_loss": 0.62816429,
      "learning_rate": 1.322714722037857e-06,
      "loss": 0.65169764,
      "num_input_tokens_seen": 111424130,
      "step": 5170,
      "time_per_iteration": 2.842797040939331
    },
    {
      "auxiliary_loss_clip": 0.01338388,
      "auxiliary_loss_mlp": 0.01028576,
      "balance_loss_clip": 1.04795551,
      "balance_loss_mlp": 1.02076733,
      "epoch": 0.6217759874947394,
      "flos": 27928770105600.0,
      "grad_norm": 3.30963765509893,
      "language_loss": 0.76940215,
      "learning_rate": 1.321981827268591e-06,
      "loss": 0.79307175,
      "num_input_tokens_seen": 111444785,
      "step": 5171,
      "time_per_iteration": 2.9384877681732178
    },
    {
      "auxiliary_loss_clip": 0.01280948,
      "auxiliary_loss_mlp": 0.01026844,
      "balance_loss_clip": 1.0473423,
      "balance_loss_mlp": 1.01934004,
      "epoch": 0.6218962303853784,
      "flos": 21765673601280.0,
      "grad_norm": 1.7103725717863467,
      "language_loss": 0.81408799,
      "learning_rate": 1.3212490353541426e-06,
      "loss": 0.83716583,
      "num_input_tokens_seen": 111467045,
      "step": 5172,
      "time_per_iteration": 2.7653162479400635
    },
    {
      "auxiliary_loss_clip": 0.01181833,
      "auxiliary_loss_mlp": 0.01026263,
      "balance_loss_clip": 1.05262268,
      "balance_loss_mlp": 1.01865411,
      "epoch": 0.6220164732760175,
      "flos": 21246260981760.0,
      "grad_norm": 2.0431840836157504,
      "language_loss": 0.80544585,
      "learning_rate": 1.3205163464056762e-06,
      "loss": 0.82752681,
      "num_input_tokens_seen": 111483650,
      "step": 5173,
      "time_per_iteration": 2.6027352809906006
    },
    {
      "auxiliary_loss_clip": 0.01226503,
      "auxiliary_loss_mlp": 0.01024947,
      "balance_loss_clip": 1.04924464,
      "balance_loss_mlp": 1.01813984,
      "epoch": 0.6221367161666567,
      "flos": 26136360506880.0,
      "grad_norm": 1.9340305325811804,
      "language_loss": 0.72704864,
      "learning_rate": 1.319783760534339e-06,
      "loss": 0.74956316,
      "num_input_tokens_seen": 111502895,
      "step": 5174,
      "time_per_iteration": 2.6610209941864014
    },
    {
      "auxiliary_loss_clip": 0.01230182,
      "auxiliary_loss_mlp": 0.01029371,
      "balance_loss_clip": 1.05012846,
      "balance_loss_mlp": 1.02186704,
      "epoch": 0.6222569590572957,
      "flos": 16284196558080.0,
      "grad_norm": 2.1270659508916143,
      "language_loss": 0.75668931,
      "learning_rate": 1.319051277851266e-06,
      "loss": 0.77928483,
      "num_input_tokens_seen": 111519180,
      "step": 5175,
      "time_per_iteration": 2.573702335357666
    },
    {
      "auxiliary_loss_clip": 0.01132172,
      "auxiliary_loss_mlp": 0.01028079,
      "balance_loss_clip": 1.05075049,
      "balance_loss_mlp": 1.02055109,
      "epoch": 0.6223772019479348,
      "flos": 18223840005120.0,
      "grad_norm": 1.8757151761914947,
      "language_loss": 0.84061658,
      "learning_rate": 1.3183188984675716e-06,
      "loss": 0.8622191,
      "num_input_tokens_seen": 111537545,
      "step": 5176,
      "time_per_iteration": 2.6659088134765625
    },
    {
      "auxiliary_loss_clip": 0.01277793,
      "auxiliary_loss_mlp": 0.01032056,
      "balance_loss_clip": 1.05070949,
      "balance_loss_mlp": 1.02477217,
      "epoch": 0.6224974448385739,
      "flos": 27489797994240.0,
      "grad_norm": 2.1480520963358565,
      "language_loss": 0.71681678,
      "learning_rate": 1.3175866224943586e-06,
      "loss": 0.73991531,
      "num_input_tokens_seen": 111556265,
      "step": 5177,
      "time_per_iteration": 2.676151752471924
    },
    {
      "auxiliary_loss_clip": 0.01284018,
      "auxiliary_loss_mlp": 0.01028273,
      "balance_loss_clip": 1.05092418,
      "balance_loss_mlp": 1.02076316,
      "epoch": 0.622617687729213,
      "flos": 19791951125760.0,
      "grad_norm": 2.2846404041403208,
      "language_loss": 0.73349935,
      "learning_rate": 1.316854450042712e-06,
      "loss": 0.7566222,
      "num_input_tokens_seen": 111574205,
      "step": 5178,
      "time_per_iteration": 2.6740732192993164
    },
    {
      "auxiliary_loss_clip": 0.01232915,
      "auxiliary_loss_mlp": 0.01031981,
      "balance_loss_clip": 1.05099905,
      "balance_loss_mlp": 1.02414894,
      "epoch": 0.622737930619852,
      "flos": 23038886062080.0,
      "grad_norm": 1.9291144280621508,
      "language_loss": 0.74416292,
      "learning_rate": 1.3161223812237024e-06,
      "loss": 0.76681191,
      "num_input_tokens_seen": 111593560,
      "step": 5179,
      "time_per_iteration": 3.5173895359039307
    },
    {
      "auxiliary_loss_clip": 0.01179974,
      "auxiliary_loss_mlp": 0.01030945,
      "balance_loss_clip": 1.04969859,
      "balance_loss_mlp": 1.02293694,
      "epoch": 0.6228581735104912,
      "flos": 12634271959680.0,
      "grad_norm": 2.89019033178809,
      "language_loss": 0.85476196,
      "learning_rate": 1.3153904161483842e-06,
      "loss": 0.87687123,
      "num_input_tokens_seen": 111608860,
      "step": 5180,
      "time_per_iteration": 2.5421395301818848
    },
    {
      "auxiliary_loss_clip": 0.01326616,
      "auxiliary_loss_mlp": 0.01024359,
      "balance_loss_clip": 1.0459609,
      "balance_loss_mlp": 1.01699781,
      "epoch": 0.6229784164011303,
      "flos": 23802813538560.0,
      "grad_norm": 2.288192458924868,
      "language_loss": 0.85684264,
      "learning_rate": 1.3146585549277953e-06,
      "loss": 0.88035238,
      "num_input_tokens_seen": 111627500,
      "step": 5181,
      "time_per_iteration": 3.587916851043701
    },
    {
      "auxiliary_loss_clip": 0.0119138,
      "auxiliary_loss_mlp": 0.01024439,
      "balance_loss_clip": 1.05265808,
      "balance_loss_mlp": 1.01730728,
      "epoch": 0.6230986592917693,
      "flos": 22414219614720.0,
      "grad_norm": 2.0657661511649277,
      "language_loss": 0.78731394,
      "learning_rate": 1.3139267976729591e-06,
      "loss": 0.8094722,
      "num_input_tokens_seen": 111647690,
      "step": 5182,
      "time_per_iteration": 2.6124725341796875
    },
    {
      "auxiliary_loss_clip": 0.01233559,
      "auxiliary_loss_mlp": 0.01024946,
      "balance_loss_clip": 1.05161548,
      "balance_loss_mlp": 1.0177846,
      "epoch": 0.6232189021824085,
      "flos": 34528217028480.0,
      "grad_norm": 1.9745758488708953,
      "language_loss": 0.7176975,
      "learning_rate": 1.3131951444948815e-06,
      "loss": 0.7402826,
      "num_input_tokens_seen": 111667090,
      "step": 5183,
      "time_per_iteration": 3.6561830043792725
    },
    {
      "auxiliary_loss_clip": 0.01285339,
      "auxiliary_loss_mlp": 0.01032713,
      "balance_loss_clip": 1.05320978,
      "balance_loss_mlp": 1.02491045,
      "epoch": 0.6233391450730476,
      "flos": 22237000888320.0,
      "grad_norm": 1.9103366897641774,
      "language_loss": 0.7642498,
      "learning_rate": 1.3124635955045546e-06,
      "loss": 0.78743035,
      "num_input_tokens_seen": 111686905,
      "step": 5184,
      "time_per_iteration": 2.7294771671295166
    },
    {
      "auxiliary_loss_clip": 0.0127125,
      "auxiliary_loss_mlp": 0.02567668,
      "balance_loss_clip": 1.04175615,
      "balance_loss_mlp": 0.99995434,
      "epoch": 0.6234593879636866,
      "flos": 20332693445760.0,
      "grad_norm": 1.969891007262231,
      "language_loss": 0.84271288,
      "learning_rate": 1.3117321508129537e-06,
      "loss": 0.88110209,
      "num_input_tokens_seen": 111704985,
      "step": 5185,
      "time_per_iteration": 2.764082431793213
    },
    {
      "auxiliary_loss_clip": 0.01281492,
      "auxiliary_loss_mlp": 0.01031103,
      "balance_loss_clip": 1.04946971,
      "balance_loss_mlp": 1.02371788,
      "epoch": 0.6235796308543258,
      "flos": 20664903358080.0,
      "grad_norm": 1.6974515740998122,
      "language_loss": 0.76268929,
      "learning_rate": 1.3110008105310388e-06,
      "loss": 0.78581524,
      "num_input_tokens_seen": 111724805,
      "step": 5186,
      "time_per_iteration": 2.709744930267334
    },
    {
      "auxiliary_loss_clip": 0.01183065,
      "auxiliary_loss_mlp": 0.01025044,
      "balance_loss_clip": 1.05122101,
      "balance_loss_mlp": 1.01740241,
      "epoch": 0.6236998737449648,
      "flos": 26618641441920.0,
      "grad_norm": 2.8923904763969612,
      "language_loss": 0.78076375,
      "learning_rate": 1.3102695747697526e-06,
      "loss": 0.80284476,
      "num_input_tokens_seen": 111747675,
      "step": 5187,
      "time_per_iteration": 2.670607566833496
    },
    {
      "auxiliary_loss_clip": 0.01427667,
      "auxiliary_loss_mlp": 0.01028597,
      "balance_loss_clip": 1.04726493,
      "balance_loss_mlp": 1.02039492,
      "epoch": 0.6238201166356039,
      "flos": 12674599954560.0,
      "grad_norm": 2.205005947132641,
      "language_loss": 0.90754461,
      "learning_rate": 1.3095384436400237e-06,
      "loss": 0.93210721,
      "num_input_tokens_seen": 111759205,
      "step": 5188,
      "time_per_iteration": 3.8011813163757324
    },
    {
      "auxiliary_loss_clip": 0.01187387,
      "auxiliary_loss_mlp": 0.01028554,
      "balance_loss_clip": 1.04876733,
      "balance_loss_mlp": 1.02080512,
      "epoch": 0.623940359526243,
      "flos": 10452160730880.0,
      "grad_norm": 2.1058535734879684,
      "language_loss": 0.82734048,
      "learning_rate": 1.3088074172527633e-06,
      "loss": 0.84949994,
      "num_input_tokens_seen": 111776335,
      "step": 5189,
      "time_per_iteration": 2.6194944381713867
    },
    {
      "auxiliary_loss_clip": 0.01284781,
      "auxiliary_loss_mlp": 0.01032071,
      "balance_loss_clip": 1.04957616,
      "balance_loss_mlp": 1.02432227,
      "epoch": 0.6240606024168821,
      "flos": 29059525226880.0,
      "grad_norm": 2.221692202759293,
      "language_loss": 0.7168895,
      "learning_rate": 1.3080764957188684e-06,
      "loss": 0.74005806,
      "num_input_tokens_seen": 111796580,
      "step": 5190,
      "time_per_iteration": 2.7706222534179688
    },
    {
      "auxiliary_loss_clip": 0.012857,
      "auxiliary_loss_mlp": 0.01030169,
      "balance_loss_clip": 1.04545879,
      "balance_loss_mlp": 1.0230819,
      "epoch": 0.6241808453075212,
      "flos": 22018089450240.0,
      "grad_norm": 1.8103341700646838,
      "language_loss": 0.70599806,
      "learning_rate": 1.3073456791492192e-06,
      "loss": 0.72915673,
      "num_input_tokens_seen": 111816290,
      "step": 5191,
      "time_per_iteration": 2.7717013359069824
    },
    {
      "auxiliary_loss_clip": 0.01281926,
      "auxiliary_loss_mlp": 0.01030474,
      "balance_loss_clip": 1.0480082,
      "balance_loss_mlp": 1.02340245,
      "epoch": 0.6243010881981603,
      "flos": 21138708683520.0,
      "grad_norm": 1.842808891261161,
      "language_loss": 0.78428793,
      "learning_rate": 1.3066149676546801e-06,
      "loss": 0.80741197,
      "num_input_tokens_seen": 111834470,
      "step": 5192,
      "time_per_iteration": 2.660034656524658
    },
    {
      "auxiliary_loss_clip": 0.01284699,
      "auxiliary_loss_mlp": 0.01024654,
      "balance_loss_clip": 1.0551666,
      "balance_loss_mlp": 1.01751888,
      "epoch": 0.6244213310887994,
      "flos": 22344948236160.0,
      "grad_norm": 7.61983536006757,
      "language_loss": 0.66427243,
      "learning_rate": 1.3058843613460985e-06,
      "loss": 0.68736595,
      "num_input_tokens_seen": 111852410,
      "step": 5193,
      "time_per_iteration": 2.723572015762329
    },
    {
      "auxiliary_loss_clip": 0.01239102,
      "auxiliary_loss_mlp": 0.01028982,
      "balance_loss_clip": 1.0494467,
      "balance_loss_mlp": 1.02162659,
      "epoch": 0.6245415739794384,
      "flos": 15231978524160.0,
      "grad_norm": 2.334462620099171,
      "language_loss": 0.74450731,
      "learning_rate": 1.3051538603343075e-06,
      "loss": 0.76718819,
      "num_input_tokens_seen": 111870340,
      "step": 5194,
      "time_per_iteration": 2.7307798862457275
    },
    {
      "auxiliary_loss_clip": 0.01227499,
      "auxiliary_loss_mlp": 0.01026477,
      "balance_loss_clip": 1.05135751,
      "balance_loss_mlp": 1.01923466,
      "epoch": 0.6246618168700776,
      "flos": 18879891960960.0,
      "grad_norm": 2.0457238825788147,
      "language_loss": 0.67678607,
      "learning_rate": 1.3044234647301235e-06,
      "loss": 0.69932586,
      "num_input_tokens_seen": 111888365,
      "step": 5195,
      "time_per_iteration": 2.6403303146362305
    },
    {
      "auxiliary_loss_clip": 0.01225466,
      "auxiliary_loss_mlp": 0.01029998,
      "balance_loss_clip": 1.05149782,
      "balance_loss_mlp": 1.02255893,
      "epoch": 0.6247820597607167,
      "flos": 14319201087360.0,
      "grad_norm": 1.7156315443861265,
      "language_loss": 0.72776508,
      "learning_rate": 1.303693174644347e-06,
      "loss": 0.75031972,
      "num_input_tokens_seen": 111905840,
      "step": 5196,
      "time_per_iteration": 2.6161599159240723
    },
    {
      "auxiliary_loss_clip": 0.01272262,
      "auxiliary_loss_mlp": 0.0102456,
      "balance_loss_clip": 1.04532743,
      "balance_loss_mlp": 1.01690054,
      "epoch": 0.6249023026513557,
      "flos": 22637979388800.0,
      "grad_norm": 2.029532668543354,
      "language_loss": 0.80717981,
      "learning_rate": 1.3029629901877625e-06,
      "loss": 0.83014798,
      "num_input_tokens_seen": 111925215,
      "step": 5197,
      "time_per_iteration": 2.7429633140563965
    },
    {
      "auxiliary_loss_clip": 0.01137751,
      "auxiliary_loss_mlp": 0.01029809,
      "balance_loss_clip": 1.0507803,
      "balance_loss_mlp": 1.02282667,
      "epoch": 0.6250225455419949,
      "flos": 20266690204800.0,
      "grad_norm": 4.249011547311132,
      "language_loss": 0.77396667,
      "learning_rate": 1.3022329114711376e-06,
      "loss": 0.79564226,
      "num_input_tokens_seen": 111943925,
      "step": 5198,
      "time_per_iteration": 2.6942427158355713
    },
    {
      "auxiliary_loss_clip": 0.01276326,
      "auxiliary_loss_mlp": 0.01028715,
      "balance_loss_clip": 1.04841995,
      "balance_loss_mlp": 1.02135921,
      "epoch": 0.6251427884326339,
      "flos": 23437853400960.0,
      "grad_norm": 1.868194934020265,
      "language_loss": 0.6968022,
      "learning_rate": 1.3015029386052256e-06,
      "loss": 0.71985269,
      "num_input_tokens_seen": 111964095,
      "step": 5199,
      "time_per_iteration": 2.838588237762451
    },
    {
      "auxiliary_loss_clip": 0.01242112,
      "auxiliary_loss_mlp": 0.01029085,
      "balance_loss_clip": 1.05128682,
      "balance_loss_mlp": 1.02180719,
      "epoch": 0.625263031323273,
      "flos": 31723055464320.0,
      "grad_norm": 2.6019472876714422,
      "language_loss": 0.72895747,
      "learning_rate": 1.3007730717007622e-06,
      "loss": 0.75166947,
      "num_input_tokens_seen": 111984910,
      "step": 5200,
      "time_per_iteration": 2.7884914875030518
    },
    {
      "auxiliary_loss_clip": 0.01186305,
      "auxiliary_loss_mlp": 0.01029607,
      "balance_loss_clip": 1.05355215,
      "balance_loss_mlp": 1.02130973,
      "epoch": 0.6253832742139122,
      "flos": 24134341092480.0,
      "grad_norm": 1.982189716871274,
      "language_loss": 0.75760674,
      "learning_rate": 1.3000433108684676e-06,
      "loss": 0.77976584,
      "num_input_tokens_seen": 112005410,
      "step": 5201,
      "time_per_iteration": 2.666295051574707
    },
    {
      "auxiliary_loss_clip": 0.01227432,
      "auxiliary_loss_mlp": 0.01027327,
      "balance_loss_clip": 1.05048203,
      "balance_loss_mlp": 1.01982892,
      "epoch": 0.6255035171045512,
      "flos": 27668812400640.0,
      "grad_norm": 2.371057201589831,
      "language_loss": 0.80644453,
      "learning_rate": 1.2993136562190467e-06,
      "loss": 0.82899213,
      "num_input_tokens_seen": 112024530,
      "step": 5202,
      "time_per_iteration": 2.6810905933380127
    },
    {
      "auxiliary_loss_clip": 0.01280967,
      "auxiliary_loss_mlp": 0.010269,
      "balance_loss_clip": 1.04903197,
      "balance_loss_mlp": 1.01965809,
      "epoch": 0.6256237599951903,
      "flos": 20227798753920.0,
      "grad_norm": 1.6107068382200134,
      "language_loss": 0.70357275,
      "learning_rate": 1.2985841078631871e-06,
      "loss": 0.72665143,
      "num_input_tokens_seen": 112043850,
      "step": 5203,
      "time_per_iteration": 2.7017014026641846
    },
    {
      "auxiliary_loss_clip": 0.01426394,
      "auxiliary_loss_mlp": 0.01024183,
      "balance_loss_clip": 1.04036295,
      "balance_loss_mlp": 1.01657176,
      "epoch": 0.6257440028858293,
      "flos": 24170574936960.0,
      "grad_norm": 1.6946631554952534,
      "language_loss": 0.781528,
      "learning_rate": 1.2978546659115608e-06,
      "loss": 0.80603373,
      "num_input_tokens_seen": 112061930,
      "step": 5204,
      "time_per_iteration": 2.8244731426239014
    },
    {
      "auxiliary_loss_clip": 0.01282019,
      "auxiliary_loss_mlp": 0.01027633,
      "balance_loss_clip": 1.04928863,
      "balance_loss_mlp": 1.02057004,
      "epoch": 0.6258642457764685,
      "flos": 15851940289920.0,
      "grad_norm": 1.8291364951069515,
      "language_loss": 0.84975839,
      "learning_rate": 1.2971253304748228e-06,
      "loss": 0.87285489,
      "num_input_tokens_seen": 112079645,
      "step": 5205,
      "time_per_iteration": 3.6474146842956543
    },
    {
      "auxiliary_loss_clip": 0.01233389,
      "auxiliary_loss_mlp": 0.01031344,
      "balance_loss_clip": 1.05276132,
      "balance_loss_mlp": 1.02292776,
      "epoch": 0.6259844886671075,
      "flos": 11911354836480.0,
      "grad_norm": 1.9337071688269172,
      "language_loss": 0.75007188,
      "learning_rate": 1.296396101663614e-06,
      "loss": 0.77271926,
      "num_input_tokens_seen": 112096205,
      "step": 5206,
      "time_per_iteration": 2.65181303024292
    },
    {
      "auxiliary_loss_clip": 0.01234778,
      "auxiliary_loss_mlp": 0.01021093,
      "balance_loss_clip": 1.05160522,
      "balance_loss_mlp": 1.01422358,
      "epoch": 0.6261047315577466,
      "flos": 15887958652800.0,
      "grad_norm": 6.066796636192211,
      "language_loss": 0.84447604,
      "learning_rate": 1.2956669795885565e-06,
      "loss": 0.86703479,
      "num_input_tokens_seen": 112112835,
      "step": 5207,
      "time_per_iteration": 3.560659646987915
    },
    {
      "auxiliary_loss_clip": 0.01322944,
      "auxiliary_loss_mlp": 0.01030105,
      "balance_loss_clip": 1.04858685,
      "balance_loss_mlp": 1.02283287,
      "epoch": 0.6262249744483858,
      "flos": 31248926916480.0,
      "grad_norm": 1.6792332758772577,
      "language_loss": 0.68574381,
      "learning_rate": 1.294937964360259e-06,
      "loss": 0.70927429,
      "num_input_tokens_seen": 112133105,
      "step": 5208,
      "time_per_iteration": 3.7277519702911377
    },
    {
      "auxiliary_loss_clip": 0.01185757,
      "auxiliary_loss_mlp": 0.01030481,
      "balance_loss_clip": 1.04892087,
      "balance_loss_mlp": 1.0222733,
      "epoch": 0.6263452173390248,
      "flos": 27198598435200.0,
      "grad_norm": 2.302988849097869,
      "language_loss": 0.71555918,
      "learning_rate": 1.2942090560893108e-06,
      "loss": 0.73772156,
      "num_input_tokens_seen": 112152510,
      "step": 5209,
      "time_per_iteration": 2.7351410388946533
    },
    {
      "auxiliary_loss_clip": 0.01179645,
      "auxiliary_loss_mlp": 0.0102425,
      "balance_loss_clip": 1.05172396,
      "balance_loss_mlp": 1.01753545,
      "epoch": 0.6264654602296639,
      "flos": 37342069683840.0,
      "grad_norm": 2.002068888090633,
      "language_loss": 0.60676467,
      "learning_rate": 1.2934802548862882e-06,
      "loss": 0.62880361,
      "num_input_tokens_seen": 112175295,
      "step": 5210,
      "time_per_iteration": 2.705451726913452
    },
    {
      "auxiliary_loss_clip": 0.01276181,
      "auxiliary_loss_mlp": 0.01025531,
      "balance_loss_clip": 1.04669738,
      "balance_loss_mlp": 1.01819336,
      "epoch": 0.626585703120303,
      "flos": 14756952136320.0,
      "grad_norm": 1.9284353282893616,
      "language_loss": 0.82640642,
      "learning_rate": 1.292751560861749e-06,
      "loss": 0.84942359,
      "num_input_tokens_seen": 112190200,
      "step": 5211,
      "time_per_iteration": 2.714510440826416
    },
    {
      "auxiliary_loss_clip": 0.01184253,
      "auxiliary_loss_mlp": 0.01027813,
      "balance_loss_clip": 1.05230856,
      "balance_loss_mlp": 1.02032375,
      "epoch": 0.6267059460109421,
      "flos": 22347318533760.0,
      "grad_norm": 1.8368973995079338,
      "language_loss": 0.79722935,
      "learning_rate": 1.2920229741262354e-06,
      "loss": 0.81935,
      "num_input_tokens_seen": 112208205,
      "step": 5212,
      "time_per_iteration": 2.598170757293701
    },
    {
      "auxiliary_loss_clip": 0.01278148,
      "auxiliary_loss_mlp": 0.0102821,
      "balance_loss_clip": 1.04873872,
      "balance_loss_mlp": 1.02066445,
      "epoch": 0.6268261889015811,
      "flos": 17748813617280.0,
      "grad_norm": 2.0939179913414883,
      "language_loss": 0.7512297,
      "learning_rate": 1.2912944947902739e-06,
      "loss": 0.7742933,
      "num_input_tokens_seen": 112224690,
      "step": 5213,
      "time_per_iteration": 3.597698926925659
    },
    {
      "auxiliary_loss_clip": 0.01286469,
      "auxiliary_loss_mlp": 0.01027159,
      "balance_loss_clip": 1.04905844,
      "balance_loss_mlp": 1.01929116,
      "epoch": 0.6269464317922203,
      "flos": 32846484211200.0,
      "grad_norm": 2.5395770104276134,
      "language_loss": 0.71853673,
      "learning_rate": 1.2905661229643742e-06,
      "loss": 0.74167299,
      "num_input_tokens_seen": 112244450,
      "step": 5214,
      "time_per_iteration": 2.8033905029296875
    },
    {
      "auxiliary_loss_clip": 0.01179926,
      "auxiliary_loss_mlp": 0.01025233,
      "balance_loss_clip": 1.05004597,
      "balance_loss_mlp": 1.01764488,
      "epoch": 0.6270666746828594,
      "flos": 17929192740480.0,
      "grad_norm": 2.4503313023663167,
      "language_loss": 0.83980465,
      "learning_rate": 1.2898378587590299e-06,
      "loss": 0.86185622,
      "num_input_tokens_seen": 112261050,
      "step": 5215,
      "time_per_iteration": 2.5678577423095703
    },
    {
      "auxiliary_loss_clip": 0.01229179,
      "auxiliary_loss_mlp": 0.010274,
      "balance_loss_clip": 1.05122018,
      "balance_loss_mlp": 1.02010477,
      "epoch": 0.6271869175734984,
      "flos": 17457326749440.0,
      "grad_norm": 2.0411573757529347,
      "language_loss": 0.87129951,
      "learning_rate": 1.2891097022847173e-06,
      "loss": 0.89386523,
      "num_input_tokens_seen": 112278395,
      "step": 5216,
      "time_per_iteration": 2.7097458839416504
    },
    {
      "auxiliary_loss_clip": 0.01276205,
      "auxiliary_loss_mlp": 0.01031115,
      "balance_loss_clip": 1.04763198,
      "balance_loss_mlp": 1.0236882,
      "epoch": 0.6273071604641376,
      "flos": 26868615166080.0,
      "grad_norm": 1.9269455583442034,
      "language_loss": 0.66565782,
      "learning_rate": 1.2883816536518978e-06,
      "loss": 0.68873101,
      "num_input_tokens_seen": 112299535,
      "step": 5217,
      "time_per_iteration": 2.7054436206817627
    },
    {
      "auxiliary_loss_clip": 0.01224365,
      "auxiliary_loss_mlp": 0.0102505,
      "balance_loss_clip": 1.04935503,
      "balance_loss_mlp": 1.01791811,
      "epoch": 0.6274274033547766,
      "flos": 26062384446720.0,
      "grad_norm": 2.2171834010980986,
      "language_loss": 0.81656194,
      "learning_rate": 1.2876537129710155e-06,
      "loss": 0.83905613,
      "num_input_tokens_seen": 112317265,
      "step": 5218,
      "time_per_iteration": 2.672982931137085
    },
    {
      "auxiliary_loss_clip": 0.01284211,
      "auxiliary_loss_mlp": 0.01030949,
      "balance_loss_clip": 1.05570459,
      "balance_loss_mlp": 1.02259851,
      "epoch": 0.6275476462454157,
      "flos": 20266259241600.0,
      "grad_norm": 3.78961055908765,
      "language_loss": 0.75605667,
      "learning_rate": 1.286925880352499e-06,
      "loss": 0.77920824,
      "num_input_tokens_seen": 112336125,
      "step": 5219,
      "time_per_iteration": 2.7014381885528564
    },
    {
      "auxiliary_loss_clip": 0.01282454,
      "auxiliary_loss_mlp": 0.01028451,
      "balance_loss_clip": 1.05009139,
      "balance_loss_mlp": 1.02093506,
      "epoch": 0.6276678891360549,
      "flos": 26320402817280.0,
      "grad_norm": 1.7785320595914071,
      "language_loss": 0.71172941,
      "learning_rate": 1.2861981559067592e-06,
      "loss": 0.73483849,
      "num_input_tokens_seen": 112356730,
      "step": 5220,
      "time_per_iteration": 2.7380990982055664
    },
    {
      "auxiliary_loss_clip": 0.01373036,
      "auxiliary_loss_mlp": 0.01024402,
      "balance_loss_clip": 1.04599416,
      "balance_loss_mlp": 1.01708889,
      "epoch": 0.6277881320266939,
      "flos": 13912512324480.0,
      "grad_norm": 1.9203878163776462,
      "language_loss": 0.80535662,
      "learning_rate": 1.2854705397441917e-06,
      "loss": 0.82933098,
      "num_input_tokens_seen": 112372270,
      "step": 5221,
      "time_per_iteration": 2.7552683353424072
    },
    {
      "auxiliary_loss_clip": 0.01321501,
      "auxiliary_loss_mlp": 0.01029642,
      "balance_loss_clip": 1.04384589,
      "balance_loss_mlp": 1.02182829,
      "epoch": 0.627908374917333,
      "flos": 27048922462080.0,
      "grad_norm": 2.2601976505738857,
      "language_loss": 0.77521902,
      "learning_rate": 1.2847430319751747e-06,
      "loss": 0.79873043,
      "num_input_tokens_seen": 112390365,
      "step": 5222,
      "time_per_iteration": 2.729901075363159
    },
    {
      "auxiliary_loss_clip": 0.01226782,
      "auxiliary_loss_mlp": 0.01030752,
      "balance_loss_clip": 1.051512,
      "balance_loss_mlp": 1.02342689,
      "epoch": 0.6280286178079721,
      "flos": 23769201386880.0,
      "grad_norm": 3.4046659177686522,
      "language_loss": 0.67519516,
      "learning_rate": 1.2840156327100712e-06,
      "loss": 0.69777054,
      "num_input_tokens_seen": 112407490,
      "step": 5223,
      "time_per_iteration": 2.6693599224090576
    },
    {
      "auxiliary_loss_clip": 0.01180726,
      "auxiliary_loss_mlp": 0.01026641,
      "balance_loss_clip": 1.05274868,
      "balance_loss_mlp": 1.01902986,
      "epoch": 0.6281488606986112,
      "flos": 26359150613760.0,
      "grad_norm": 1.7129232632631066,
      "language_loss": 0.72301424,
      "learning_rate": 1.2832883420592272e-06,
      "loss": 0.74508786,
      "num_input_tokens_seen": 112426385,
      "step": 5224,
      "time_per_iteration": 2.6220614910125732
    },
    {
      "auxiliary_loss_clip": 0.01276586,
      "auxiliary_loss_mlp": 0.01030435,
      "balance_loss_clip": 1.04826832,
      "balance_loss_mlp": 1.02274942,
      "epoch": 0.6282691035892503,
      "flos": 36137194848000.0,
      "grad_norm": 2.1736297850715633,
      "language_loss": 0.6416012,
      "learning_rate": 1.282561160132972e-06,
      "loss": 0.66467142,
      "num_input_tokens_seen": 112446905,
      "step": 5225,
      "time_per_iteration": 2.857266664505005
    },
    {
      "auxiliary_loss_clip": 0.01284876,
      "auxiliary_loss_mlp": 0.01029333,
      "balance_loss_clip": 1.04609144,
      "balance_loss_mlp": 1.02181077,
      "epoch": 0.6283893464798894,
      "flos": 26537231266560.0,
      "grad_norm": 2.444198818965886,
      "language_loss": 0.80891258,
      "learning_rate": 1.2818340870416186e-06,
      "loss": 0.83205473,
      "num_input_tokens_seen": 112468040,
      "step": 5226,
      "time_per_iteration": 2.72715163230896
    },
    {
      "auxiliary_loss_clip": 0.01337256,
      "auxiliary_loss_mlp": 0.01027544,
      "balance_loss_clip": 1.04593253,
      "balance_loss_mlp": 1.01992095,
      "epoch": 0.6285095893705285,
      "flos": 22237216369920.0,
      "grad_norm": 4.409693665180063,
      "language_loss": 0.75872684,
      "learning_rate": 1.2811071228954626e-06,
      "loss": 0.78237486,
      "num_input_tokens_seen": 112486675,
      "step": 5227,
      "time_per_iteration": 2.768714666366577
    },
    {
      "auxiliary_loss_clip": 0.01278262,
      "auxiliary_loss_mlp": 0.0102567,
      "balance_loss_clip": 1.04979944,
      "balance_loss_mlp": 1.01852012,
      "epoch": 0.6286298322611675,
      "flos": 26542259170560.0,
      "grad_norm": 2.0823425975974876,
      "language_loss": 0.80741644,
      "learning_rate": 1.2803802678047846e-06,
      "loss": 0.83045572,
      "num_input_tokens_seen": 112506825,
      "step": 5228,
      "time_per_iteration": 2.698317050933838
    },
    {
      "auxiliary_loss_clip": 0.01285698,
      "auxiliary_loss_mlp": 0.01032797,
      "balance_loss_clip": 1.0524826,
      "balance_loss_mlp": 1.02371299,
      "epoch": 0.6287500751518067,
      "flos": 21795227516160.0,
      "grad_norm": 1.7463572881287777,
      "language_loss": 0.74092978,
      "learning_rate": 1.279653521879848e-06,
      "loss": 0.76411474,
      "num_input_tokens_seen": 112526890,
      "step": 5229,
      "time_per_iteration": 2.6969852447509766
    },
    {
      "auxiliary_loss_clip": 0.01464817,
      "auxiliary_loss_mlp": 0.01027109,
      "balance_loss_clip": 1.03931236,
      "balance_loss_mlp": 1.01981306,
      "epoch": 0.6288703180424458,
      "flos": 20009605587840.0,
      "grad_norm": 2.20501053662743,
      "language_loss": 0.83993769,
      "learning_rate": 1.2789268852308997e-06,
      "loss": 0.86485696,
      "num_input_tokens_seen": 112542100,
      "step": 5230,
      "time_per_iteration": 2.7914962768554688
    },
    {
      "auxiliary_loss_clip": 0.01226974,
      "auxiliary_loss_mlp": 0.01033333,
      "balance_loss_clip": 1.04956698,
      "balance_loss_mlp": 1.02590632,
      "epoch": 0.6289905609330848,
      "flos": 22124923476480.0,
      "grad_norm": 2.021888885453244,
      "language_loss": 0.70989525,
      "learning_rate": 1.2782003579681688e-06,
      "loss": 0.73249829,
      "num_input_tokens_seen": 112561630,
      "step": 5231,
      "time_per_iteration": 3.607811450958252
    },
    {
      "auxiliary_loss_clip": 0.01183896,
      "auxiliary_loss_mlp": 0.01028841,
      "balance_loss_clip": 1.0522325,
      "balance_loss_mlp": 1.02156937,
      "epoch": 0.629110803823724,
      "flos": 25518481729920.0,
      "grad_norm": 1.8426764453027429,
      "language_loss": 0.7416178,
      "learning_rate": 1.2774739402018701e-06,
      "loss": 0.76374513,
      "num_input_tokens_seen": 112582465,
      "step": 5232,
      "time_per_iteration": 2.700361490249634
    },
    {
      "auxiliary_loss_clip": 0.0123854,
      "auxiliary_loss_mlp": 0.01028457,
      "balance_loss_clip": 1.05767632,
      "balance_loss_mlp": 1.02052402,
      "epoch": 0.629231046714363,
      "flos": 20886616056960.0,
      "grad_norm": 1.7823799680677446,
      "language_loss": 0.73080593,
      "learning_rate": 1.2767476320422002e-06,
      "loss": 0.7534759,
      "num_input_tokens_seen": 112602390,
      "step": 5233,
      "time_per_iteration": 3.55440092086792
    },
    {
      "auxiliary_loss_clip": 0.01234997,
      "auxiliary_loss_mlp": 0.01001097,
      "balance_loss_clip": 1.01045299,
      "balance_loss_mlp": 1.00007808,
      "epoch": 0.6293512896050021,
      "flos": 65050027908480.0,
      "grad_norm": 0.6837399785350393,
      "language_loss": 0.57146657,
      "learning_rate": 1.2760214335993392e-06,
      "loss": 0.59382749,
      "num_input_tokens_seen": 112669035,
      "step": 5234,
      "time_per_iteration": 3.35624098777771
    },
    {
      "auxiliary_loss_clip": 0.01227738,
      "auxiliary_loss_mlp": 0.01025654,
      "balance_loss_clip": 1.04887044,
      "balance_loss_mlp": 1.01877594,
      "epoch": 0.6294715324956413,
      "flos": 34677857088000.0,
      "grad_norm": 2.1483736888928004,
      "language_loss": 0.58242559,
      "learning_rate": 1.2752953449834514e-06,
      "loss": 0.60495949,
      "num_input_tokens_seen": 112691485,
      "step": 5235,
      "time_per_iteration": 3.6812658309936523
    },
    {
      "auxiliary_loss_clip": 0.01181427,
      "auxiliary_loss_mlp": 0.01026593,
      "balance_loss_clip": 1.052356,
      "balance_loss_mlp": 1.01957774,
      "epoch": 0.6295917753862803,
      "flos": 22784207656320.0,
      "grad_norm": 2.216772044019813,
      "language_loss": 0.80476153,
      "learning_rate": 1.2745693663046836e-06,
      "loss": 0.82684171,
      "num_input_tokens_seen": 112710555,
      "step": 5236,
      "time_per_iteration": 2.626579999923706
    },
    {
      "auxiliary_loss_clip": 0.01224565,
      "auxiliary_loss_mlp": 0.01024977,
      "balance_loss_clip": 1.05017364,
      "balance_loss_mlp": 1.01774669,
      "epoch": 0.6297120182769194,
      "flos": 20850454039680.0,
      "grad_norm": 2.6793175924225854,
      "language_loss": 0.81050617,
      "learning_rate": 1.2738434976731662e-06,
      "loss": 0.83300161,
      "num_input_tokens_seen": 112728740,
      "step": 5237,
      "time_per_iteration": 2.6783668994903564
    },
    {
      "auxiliary_loss_clip": 0.01278971,
      "auxiliary_loss_mlp": 0.01028019,
      "balance_loss_clip": 1.04943061,
      "balance_loss_mlp": 1.02059484,
      "epoch": 0.6298322611675584,
      "flos": 19497662997120.0,
      "grad_norm": 3.8363346495975077,
      "language_loss": 0.75039554,
      "learning_rate": 1.2731177391990125e-06,
      "loss": 0.77346539,
      "num_input_tokens_seen": 112748665,
      "step": 5238,
      "time_per_iteration": 2.6868369579315186
    },
    {
      "auxiliary_loss_clip": 0.01279937,
      "auxiliary_loss_mlp": 0.01025604,
      "balance_loss_clip": 1.04714429,
      "balance_loss_mlp": 1.01817465,
      "epoch": 0.6299525040581976,
      "flos": 12604466649600.0,
      "grad_norm": 1.9173752880746324,
      "language_loss": 0.81794739,
      "learning_rate": 1.2723920909923203e-06,
      "loss": 0.84100282,
      "num_input_tokens_seen": 112764410,
      "step": 5239,
      "time_per_iteration": 2.6843395233154297
    },
    {
      "auxiliary_loss_clip": 0.01067986,
      "auxiliary_loss_mlp": 0.01006305,
      "balance_loss_clip": 1.0113008,
      "balance_loss_mlp": 1.0052979,
      "epoch": 0.6300727469488366,
      "flos": 57725685636480.0,
      "grad_norm": 1.0004861188553915,
      "language_loss": 0.60396427,
      "learning_rate": 1.2716665531631688e-06,
      "loss": 0.62470716,
      "num_input_tokens_seen": 112818695,
      "step": 5240,
      "time_per_iteration": 3.974187135696411
    },
    {
      "auxiliary_loss_clip": 0.0113416,
      "auxiliary_loss_mlp": 0.01033068,
      "balance_loss_clip": 1.04911494,
      "balance_loss_mlp": 1.02533185,
      "epoch": 0.6301929898394757,
      "flos": 22527302607360.0,
      "grad_norm": 2.4079822232954613,
      "language_loss": 0.77429599,
      "learning_rate": 1.270941125821623e-06,
      "loss": 0.79596823,
      "num_input_tokens_seen": 112839120,
      "step": 5241,
      "time_per_iteration": 2.70611310005188
    },
    {
      "auxiliary_loss_clip": 0.01227386,
      "auxiliary_loss_mlp": 0.01025736,
      "balance_loss_clip": 1.04767108,
      "balance_loss_mlp": 1.01900923,
      "epoch": 0.6303132327301149,
      "flos": 28293550675200.0,
      "grad_norm": 1.818001164859896,
      "language_loss": 0.75621247,
      "learning_rate": 1.2702158090777278e-06,
      "loss": 0.77874368,
      "num_input_tokens_seen": 112860210,
      "step": 5242,
      "time_per_iteration": 2.7473702430725098
    },
    {
      "auxiliary_loss_clip": 0.01326662,
      "auxiliary_loss_mlp": 0.01032447,
      "balance_loss_clip": 1.04576719,
      "balance_loss_mlp": 1.02548838,
      "epoch": 0.6304334756207539,
      "flos": 25264521596160.0,
      "grad_norm": 2.813112057760493,
      "language_loss": 0.7515735,
      "learning_rate": 1.2694906030415148e-06,
      "loss": 0.7751646,
      "num_input_tokens_seen": 112877955,
      "step": 5243,
      "time_per_iteration": 2.7301151752471924
    },
    {
      "auxiliary_loss_clip": 0.01285846,
      "auxiliary_loss_mlp": 0.01029922,
      "balance_loss_clip": 1.04724777,
      "balance_loss_mlp": 1.0221796,
      "epoch": 0.630553718511393,
      "flos": 18033548728320.0,
      "grad_norm": 3.4664413119164865,
      "language_loss": 0.82258189,
      "learning_rate": 1.2687655078229958e-06,
      "loss": 0.8457396,
      "num_input_tokens_seen": 112892285,
      "step": 5244,
      "time_per_iteration": 2.717881679534912
    },
    {
      "auxiliary_loss_clip": 0.01277244,
      "auxiliary_loss_mlp": 0.01025758,
      "balance_loss_clip": 1.05146706,
      "balance_loss_mlp": 1.01854849,
      "epoch": 0.6306739614020321,
      "flos": 27304103658240.0,
      "grad_norm": 2.8456645586947897,
      "language_loss": 0.69128603,
      "learning_rate": 1.2680405235321678e-06,
      "loss": 0.71431607,
      "num_input_tokens_seen": 112913620,
      "step": 5245,
      "time_per_iteration": 2.7474567890167236
    },
    {
      "auxiliary_loss_clip": 0.01286453,
      "auxiliary_loss_mlp": 0.02570352,
      "balance_loss_clip": 1.05560398,
      "balance_loss_mlp": 0.99988598,
      "epoch": 0.6307942042926712,
      "flos": 15341434243200.0,
      "grad_norm": 2.0632607644560816,
      "language_loss": 0.78449005,
      "learning_rate": 1.267315650279011e-06,
      "loss": 0.82305807,
      "num_input_tokens_seen": 112932090,
      "step": 5246,
      "time_per_iteration": 2.6856634616851807
    },
    {
      "auxiliary_loss_clip": 0.01321156,
      "auxiliary_loss_mlp": 0.01031118,
      "balance_loss_clip": 1.04765201,
      "balance_loss_mlp": 1.02315438,
      "epoch": 0.6309144471833102,
      "flos": 19606400444160.0,
      "grad_norm": 1.9669207060561098,
      "language_loss": 0.73973566,
      "learning_rate": 1.2665908881734874e-06,
      "loss": 0.76325834,
      "num_input_tokens_seen": 112950925,
      "step": 5247,
      "time_per_iteration": 2.7034502029418945
    },
    {
      "auxiliary_loss_clip": 0.01227234,
      "auxiliary_loss_mlp": 0.010256,
      "balance_loss_clip": 1.04886651,
      "balance_loss_mlp": 1.01860225,
      "epoch": 0.6310346900739494,
      "flos": 17493345112320.0,
      "grad_norm": 2.255887720296965,
      "language_loss": 0.85421461,
      "learning_rate": 1.2658662373255432e-06,
      "loss": 0.87674296,
      "num_input_tokens_seen": 112969315,
      "step": 5248,
      "time_per_iteration": 2.6199700832366943
    },
    {
      "auxiliary_loss_clip": 0.0117256,
      "auxiliary_loss_mlp": 0.01006581,
      "balance_loss_clip": 1.01031876,
      "balance_loss_mlp": 1.00560975,
      "epoch": 0.6311549329645885,
      "flos": 55070164131840.0,
      "grad_norm": 0.7117239665389015,
      "language_loss": 0.52194118,
      "learning_rate": 1.2651416978451063e-06,
      "loss": 0.54373258,
      "num_input_tokens_seen": 113034700,
      "step": 5249,
      "time_per_iteration": 3.3058671951293945
    },
    {
      "auxiliary_loss_clip": 0.01186323,
      "auxiliary_loss_mlp": 0.01026321,
      "balance_loss_clip": 1.05386317,
      "balance_loss_mlp": 1.01845372,
      "epoch": 0.6312751758552275,
      "flos": 41902545075840.0,
      "grad_norm": 2.5629892148116684,
      "language_loss": 0.65270674,
      "learning_rate": 1.2644172698420903e-06,
      "loss": 0.67483318,
      "num_input_tokens_seen": 113056805,
      "step": 5250,
      "time_per_iteration": 2.7889795303344727
    },
    {
      "auxiliary_loss_clip": 0.01328652,
      "auxiliary_loss_mlp": 0.01034586,
      "balance_loss_clip": 1.04693031,
      "balance_loss_mlp": 1.02722144,
      "epoch": 0.6313954187458667,
      "flos": 19646800266240.0,
      "grad_norm": 1.9926496973763286,
      "language_loss": 0.85196859,
      "learning_rate": 1.2636929534263892e-06,
      "loss": 0.87560093,
      "num_input_tokens_seen": 113075790,
      "step": 5251,
      "time_per_iteration": 2.7114148139953613
    },
    {
      "auxiliary_loss_clip": 0.01331079,
      "auxiliary_loss_mlp": 0.0102676,
      "balance_loss_clip": 1.04313207,
      "balance_loss_mlp": 1.01963091,
      "epoch": 0.6315156616365057,
      "flos": 22894273906560.0,
      "grad_norm": 4.4396584599345,
      "language_loss": 0.77206576,
      "learning_rate": 1.2629687487078821e-06,
      "loss": 0.79564416,
      "num_input_tokens_seen": 113094600,
      "step": 5252,
      "time_per_iteration": 2.757223129272461
    },
    {
      "auxiliary_loss_clip": 0.01231865,
      "auxiliary_loss_mlp": 0.01030509,
      "balance_loss_clip": 1.04732025,
      "balance_loss_mlp": 1.02246547,
      "epoch": 0.6316359045271448,
      "flos": 23726251699200.0,
      "grad_norm": 2.389215926509584,
      "language_loss": 0.76536751,
      "learning_rate": 1.2622446557964293e-06,
      "loss": 0.78799129,
      "num_input_tokens_seen": 113112605,
      "step": 5253,
      "time_per_iteration": 2.68548583984375
    },
    {
      "auxiliary_loss_clip": 0.01278353,
      "auxiliary_loss_mlp": 0.01019873,
      "balance_loss_clip": 1.04542851,
      "balance_loss_mlp": 1.01240444,
      "epoch": 0.631756147417784,
      "flos": 33108417164160.0,
      "grad_norm": 2.136544840285888,
      "language_loss": 0.71831954,
      "learning_rate": 1.261520674801876e-06,
      "loss": 0.74130177,
      "num_input_tokens_seen": 113133200,
      "step": 5254,
      "time_per_iteration": 2.784088134765625
    },
    {
      "auxiliary_loss_clip": 0.01285631,
      "auxiliary_loss_mlp": 0.01035564,
      "balance_loss_clip": 1.05625749,
      "balance_loss_mlp": 1.02712381,
      "epoch": 0.631876390308423,
      "flos": 31248424126080.0,
      "grad_norm": 2.3634801171899755,
      "language_loss": 0.72594404,
      "learning_rate": 1.2607968058340488e-06,
      "loss": 0.749156,
      "num_input_tokens_seen": 113152895,
      "step": 5255,
      "time_per_iteration": 2.7550034523010254
    },
    {
      "auxiliary_loss_clip": 0.0127498,
      "auxiliary_loss_mlp": 0.01026884,
      "balance_loss_clip": 1.0464381,
      "balance_loss_mlp": 1.01962972,
      "epoch": 0.6319966331990621,
      "flos": 24681152810880.0,
      "grad_norm": 1.984211437842188,
      "language_loss": 0.73435342,
      "learning_rate": 1.2600730490027583e-06,
      "loss": 0.75737208,
      "num_input_tokens_seen": 113173135,
      "step": 5256,
      "time_per_iteration": 2.7152047157287598
    },
    {
      "auxiliary_loss_clip": 0.0132509,
      "auxiliary_loss_mlp": 0.01032785,
      "balance_loss_clip": 1.04585958,
      "balance_loss_mlp": 1.02549803,
      "epoch": 0.6321168760897012,
      "flos": 17491764913920.0,
      "grad_norm": 10.538864755527547,
      "language_loss": 0.80564594,
      "learning_rate": 1.2593494044177984e-06,
      "loss": 0.82922471,
      "num_input_tokens_seen": 113191440,
      "step": 5257,
      "time_per_iteration": 3.620846748352051
    },
    {
      "auxiliary_loss_clip": 0.0118185,
      "auxiliary_loss_mlp": 0.01025594,
      "balance_loss_clip": 1.04990959,
      "balance_loss_mlp": 1.01805401,
      "epoch": 0.6322371189803403,
      "flos": 18295373940480.0,
      "grad_norm": 2.67724874632327,
      "language_loss": 0.80525362,
      "learning_rate": 1.2586258721889448e-06,
      "loss": 0.82732809,
      "num_input_tokens_seen": 113208790,
      "step": 5258,
      "time_per_iteration": 2.5657198429107666
    },
    {
      "auxiliary_loss_clip": 0.01372077,
      "auxiliary_loss_mlp": 0.01027591,
      "balance_loss_clip": 1.04523015,
      "balance_loss_mlp": 1.0198487,
      "epoch": 0.6323573618709794,
      "flos": 20157270399360.0,
      "grad_norm": 2.124776939129925,
      "language_loss": 0.81699967,
      "learning_rate": 1.2579024524259573e-06,
      "loss": 0.84099633,
      "num_input_tokens_seen": 113225050,
      "step": 5259,
      "time_per_iteration": 3.63861083984375
    },
    {
      "auxiliary_loss_clip": 0.01274109,
      "auxiliary_loss_mlp": 0.01026397,
      "balance_loss_clip": 1.04296684,
      "balance_loss_mlp": 1.0190326,
      "epoch": 0.6324776047616185,
      "flos": 20042391726720.0,
      "grad_norm": 2.1155244453714763,
      "language_loss": 0.91460067,
      "learning_rate": 1.2571791452385768e-06,
      "loss": 0.93760574,
      "num_input_tokens_seen": 113242315,
      "step": 5260,
      "time_per_iteration": 2.6852428913116455
    },
    {
      "auxiliary_loss_clip": 0.01277485,
      "auxiliary_loss_mlp": 0.01029851,
      "balance_loss_clip": 1.04940939,
      "balance_loss_mlp": 1.02329493,
      "epoch": 0.6325978476522576,
      "flos": 30848235724800.0,
      "grad_norm": 1.6300720456763362,
      "language_loss": 0.77230424,
      "learning_rate": 1.2564559507365301e-06,
      "loss": 0.79537755,
      "num_input_tokens_seen": 113264720,
      "step": 5261,
      "time_per_iteration": 3.6515378952026367
    },
    {
      "auxiliary_loss_clip": 0.01283459,
      "auxiliary_loss_mlp": 0.0102467,
      "balance_loss_clip": 1.0506916,
      "balance_loss_mlp": 1.01700461,
      "epoch": 0.6327180905428966,
      "flos": 24535104111360.0,
      "grad_norm": 2.9184297903601872,
      "language_loss": 0.78998768,
      "learning_rate": 1.2557328690295244e-06,
      "loss": 0.81306899,
      "num_input_tokens_seen": 113282910,
      "step": 5262,
      "time_per_iteration": 2.6835968494415283
    },
    {
      "auxiliary_loss_clip": 0.012304,
      "auxiliary_loss_mlp": 0.01031439,
      "balance_loss_clip": 1.04773486,
      "balance_loss_mlp": 1.02426827,
      "epoch": 0.6328383334335358,
      "flos": 21575274583680.0,
      "grad_norm": 1.6763201392426013,
      "language_loss": 0.76232147,
      "learning_rate": 1.255009900227251e-06,
      "loss": 0.78493983,
      "num_input_tokens_seen": 113301935,
      "step": 5263,
      "time_per_iteration": 2.761054515838623
    },
    {
      "auxiliary_loss_clip": 0.01176434,
      "auxiliary_loss_mlp": 0.01028465,
      "balance_loss_clip": 1.05067718,
      "balance_loss_mlp": 1.02124047,
      "epoch": 0.6329585763241748,
      "flos": 22929861306240.0,
      "grad_norm": 2.2159413090715536,
      "language_loss": 0.79588884,
      "learning_rate": 1.254287044439383e-06,
      "loss": 0.81793785,
      "num_input_tokens_seen": 113321540,
      "step": 5264,
      "time_per_iteration": 2.635645866394043
    },
    {
      "auxiliary_loss_clip": 0.01065272,
      "auxiliary_loss_mlp": 0.01007707,
      "balance_loss_clip": 1.00872195,
      "balance_loss_mlp": 1.00680149,
      "epoch": 0.6330788192148139,
      "flos": 70936897847040.0,
      "grad_norm": 0.773971778910318,
      "language_loss": 0.54372203,
      "learning_rate": 1.2535643017755776e-06,
      "loss": 0.56445181,
      "num_input_tokens_seen": 113383730,
      "step": 5265,
      "time_per_iteration": 3.2790462970733643
    },
    {
      "auxiliary_loss_clip": 0.01225899,
      "auxiliary_loss_mlp": 0.01028717,
      "balance_loss_clip": 1.04361844,
      "balance_loss_mlp": 1.02129626,
      "epoch": 0.6331990621054531,
      "flos": 21244501215360.0,
      "grad_norm": 3.4492790560881,
      "language_loss": 0.72267789,
      "learning_rate": 1.2528416723454737e-06,
      "loss": 0.74522412,
      "num_input_tokens_seen": 113400400,
      "step": 5266,
      "time_per_iteration": 3.681333303451538
    },
    {
      "auxiliary_loss_clip": 0.01177162,
      "auxiliary_loss_mlp": 0.0102405,
      "balance_loss_clip": 1.05089021,
      "balance_loss_mlp": 1.01702619,
      "epoch": 0.6333193049960921,
      "flos": 34459412526720.0,
      "grad_norm": 1.933363118868206,
      "language_loss": 0.70972544,
      "learning_rate": 1.2521191562586945e-06,
      "loss": 0.73173755,
      "num_input_tokens_seen": 113424050,
      "step": 5267,
      "time_per_iteration": 2.729631185531616
    },
    {
      "auxiliary_loss_clip": 0.0117753,
      "auxiliary_loss_mlp": 0.02566665,
      "balance_loss_clip": 1.05020952,
      "balance_loss_mlp": 0.99990064,
      "epoch": 0.6334395478867312,
      "flos": 18329883932160.0,
      "grad_norm": 2.6562279168059817,
      "language_loss": 0.76717275,
      "learning_rate": 1.2513967536248445e-06,
      "loss": 0.80461466,
      "num_input_tokens_seen": 113440370,
      "step": 5268,
      "time_per_iteration": 2.6455459594726562
    },
    {
      "auxiliary_loss_clip": 0.01227248,
      "auxiliary_loss_mlp": 0.01033837,
      "balance_loss_clip": 1.05121887,
      "balance_loss_mlp": 1.02618384,
      "epoch": 0.6335597907773702,
      "flos": 23623152687360.0,
      "grad_norm": 1.684609454335387,
      "language_loss": 0.81184131,
      "learning_rate": 1.2506744645535117e-06,
      "loss": 0.83445215,
      "num_input_tokens_seen": 113460800,
      "step": 5269,
      "time_per_iteration": 2.646491289138794
    },
    {
      "auxiliary_loss_clip": 0.01275716,
      "auxiliary_loss_mlp": 0.0102365,
      "balance_loss_clip": 1.04309106,
      "balance_loss_mlp": 1.01624131,
      "epoch": 0.6336800336680094,
      "flos": 22710913954560.0,
      "grad_norm": 2.6779841249830016,
      "language_loss": 0.60509878,
      "learning_rate": 1.249952289154267e-06,
      "loss": 0.62809241,
      "num_input_tokens_seen": 113480840,
      "step": 5270,
      "time_per_iteration": 2.7346575260162354
    },
    {
      "auxiliary_loss_clip": 0.01418272,
      "auxiliary_loss_mlp": 0.01028678,
      "balance_loss_clip": 1.04132748,
      "balance_loss_mlp": 1.02145362,
      "epoch": 0.6338002765586485,
      "flos": 23622757637760.0,
      "grad_norm": 1.8383011465736623,
      "language_loss": 0.76768851,
      "learning_rate": 1.2492302275366635e-06,
      "loss": 0.79215801,
      "num_input_tokens_seen": 113500515,
      "step": 5271,
      "time_per_iteration": 2.8717143535614014
    },
    {
      "auxiliary_loss_clip": 0.01227777,
      "auxiliary_loss_mlp": 0.01025046,
      "balance_loss_clip": 1.04877019,
      "balance_loss_mlp": 1.01674342,
      "epoch": 0.6339205194492875,
      "flos": 26505450708480.0,
      "grad_norm": 1.8906100038454539,
      "language_loss": 0.65720475,
      "learning_rate": 1.2485082798102377e-06,
      "loss": 0.67973292,
      "num_input_tokens_seen": 113520930,
      "step": 5272,
      "time_per_iteration": 2.66744327545166
    },
    {
      "auxiliary_loss_clip": 0.01335672,
      "auxiliary_loss_mlp": 0.01025218,
      "balance_loss_clip": 1.04652488,
      "balance_loss_mlp": 1.01782715,
      "epoch": 0.6340407623399267,
      "flos": 18544306170240.0,
      "grad_norm": 4.401367770023357,
      "language_loss": 0.6867938,
      "learning_rate": 1.2477864460845084e-06,
      "loss": 0.71040267,
      "num_input_tokens_seen": 113537330,
      "step": 5273,
      "time_per_iteration": 2.7111310958862305
    },
    {
      "auxiliary_loss_clip": 0.01280128,
      "auxiliary_loss_mlp": 0.01029879,
      "balance_loss_clip": 1.04845524,
      "balance_loss_mlp": 1.02192223,
      "epoch": 0.6341610052305657,
      "flos": 17712579772800.0,
      "grad_norm": 2.9346408429888506,
      "language_loss": 0.74403274,
      "learning_rate": 1.2470647264689776e-06,
      "loss": 0.76713282,
      "num_input_tokens_seen": 113555810,
      "step": 5274,
      "time_per_iteration": 2.644277334213257
    },
    {
      "auxiliary_loss_clip": 0.01428488,
      "auxiliary_loss_mlp": 0.01026236,
      "balance_loss_clip": 1.04045057,
      "balance_loss_mlp": 1.01851177,
      "epoch": 0.6342812481212048,
      "flos": 23587026583680.0,
      "grad_norm": 2.25228408086617,
      "language_loss": 0.71544713,
      "learning_rate": 1.2463431210731282e-06,
      "loss": 0.73999435,
      "num_input_tokens_seen": 113575395,
      "step": 5275,
      "time_per_iteration": 2.777639627456665
    },
    {
      "auxiliary_loss_clip": 0.01283966,
      "auxiliary_loss_mlp": 0.01025368,
      "balance_loss_clip": 1.04167843,
      "balance_loss_mlp": 1.01837587,
      "epoch": 0.634401491011844,
      "flos": 17821927751040.0,
      "grad_norm": 2.605796107315974,
      "language_loss": 0.7661404,
      "learning_rate": 1.2456216300064289e-06,
      "loss": 0.78923368,
      "num_input_tokens_seen": 113592945,
      "step": 5276,
      "time_per_iteration": 2.7237155437469482
    },
    {
      "auxiliary_loss_clip": 0.01275743,
      "auxiliary_loss_mlp": 0.01024254,
      "balance_loss_clip": 1.04569292,
      "balance_loss_mlp": 1.01686025,
      "epoch": 0.634521733902483,
      "flos": 21358158825600.0,
      "grad_norm": 1.6593548572924075,
      "language_loss": 0.78519452,
      "learning_rate": 1.244900253378328e-06,
      "loss": 0.80819452,
      "num_input_tokens_seen": 113613000,
      "step": 5277,
      "time_per_iteration": 2.726276397705078
    },
    {
      "auxiliary_loss_clip": 0.01425948,
      "auxiliary_loss_mlp": 0.01028537,
      "balance_loss_clip": 1.04473138,
      "balance_loss_mlp": 1.02142572,
      "epoch": 0.6346419767931221,
      "flos": 16545052103040.0,
      "grad_norm": 3.4604724031503413,
      "language_loss": 0.69620162,
      "learning_rate": 1.2441789912982583e-06,
      "loss": 0.72074646,
      "num_input_tokens_seen": 113630085,
      "step": 5278,
      "time_per_iteration": 2.8789350986480713
    },
    {
      "auxiliary_loss_clip": 0.01231589,
      "auxiliary_loss_mlp": 0.01029578,
      "balance_loss_clip": 1.05105352,
      "balance_loss_mlp": 1.02156091,
      "epoch": 0.6347622196837612,
      "flos": 24350989973760.0,
      "grad_norm": 1.7433787369038627,
      "language_loss": 0.64892191,
      "learning_rate": 1.2434578438756346e-06,
      "loss": 0.67153358,
      "num_input_tokens_seen": 113650515,
      "step": 5279,
      "time_per_iteration": 2.785865068435669
    },
    {
      "auxiliary_loss_clip": 0.01230458,
      "auxiliary_loss_mlp": 0.01025166,
      "balance_loss_clip": 1.0477016,
      "balance_loss_mlp": 1.01814139,
      "epoch": 0.6348824625744003,
      "flos": 64523178195840.0,
      "grad_norm": 1.9244817901241218,
      "language_loss": 0.77934837,
      "learning_rate": 1.242736811219855e-06,
      "loss": 0.80190468,
      "num_input_tokens_seen": 113676475,
      "step": 5280,
      "time_per_iteration": 3.053905725479126
    },
    {
      "auxiliary_loss_clip": 0.01224829,
      "auxiliary_loss_mlp": 0.01026235,
      "balance_loss_clip": 1.04686999,
      "balance_loss_mlp": 1.0190227,
      "epoch": 0.6350027054650393,
      "flos": 28622133313920.0,
      "grad_norm": 1.9938468459570362,
      "language_loss": 0.81865847,
      "learning_rate": 1.2420158934402988e-06,
      "loss": 0.84116912,
      "num_input_tokens_seen": 113697090,
      "step": 5281,
      "time_per_iteration": 2.7481939792633057
    },
    {
      "auxiliary_loss_clip": 0.01317511,
      "auxiliary_loss_mlp": 0.01023225,
      "balance_loss_clip": 1.04174614,
      "balance_loss_mlp": 1.01545238,
      "epoch": 0.6351229483556785,
      "flos": 23002544476800.0,
      "grad_norm": 2.6564719437766575,
      "language_loss": 0.84740782,
      "learning_rate": 1.2412950906463286e-06,
      "loss": 0.87081516,
      "num_input_tokens_seen": 113714395,
      "step": 5282,
      "time_per_iteration": 2.8220255374908447
    },
    {
      "auxiliary_loss_clip": 0.01371923,
      "auxiliary_loss_mlp": 0.0102662,
      "balance_loss_clip": 1.04397786,
      "balance_loss_mlp": 1.01966739,
      "epoch": 0.6352431912463176,
      "flos": 21939300967680.0,
      "grad_norm": 2.225390811335096,
      "language_loss": 0.89883804,
      "learning_rate": 1.2405744029472902e-06,
      "loss": 0.92282355,
      "num_input_tokens_seen": 113733880,
      "step": 5283,
      "time_per_iteration": 2.7432243824005127
    },
    {
      "auxiliary_loss_clip": 0.01276179,
      "auxiliary_loss_mlp": 0.01025042,
      "balance_loss_clip": 1.04707134,
      "balance_loss_mlp": 1.01791632,
      "epoch": 0.6353634341369566,
      "flos": 13735257684480.0,
      "grad_norm": 1.9389407649095132,
      "language_loss": 0.7634353,
      "learning_rate": 1.2398538304525108e-06,
      "loss": 0.78644753,
      "num_input_tokens_seen": 113752505,
      "step": 5284,
      "time_per_iteration": 3.617347478866577
    },
    {
      "auxiliary_loss_clip": 0.01335617,
      "auxiliary_loss_mlp": 0.01032284,
      "balance_loss_clip": 1.04982948,
      "balance_loss_mlp": 1.0243988,
      "epoch": 0.6354836770275958,
      "flos": 19316170552320.0,
      "grad_norm": 2.5036315219263514,
      "language_loss": 0.75693536,
      "learning_rate": 1.2391333732713016e-06,
      "loss": 0.78061438,
      "num_input_tokens_seen": 113770310,
      "step": 5285,
      "time_per_iteration": 3.629286766052246
    },
    {
      "auxiliary_loss_clip": 0.01335075,
      "auxiliary_loss_mlp": 0.01027891,
      "balance_loss_clip": 1.04631329,
      "balance_loss_mlp": 1.02003467,
      "epoch": 0.6356039199182348,
      "flos": 21613375935360.0,
      "grad_norm": 3.783730187925394,
      "language_loss": 0.78629208,
      "learning_rate": 1.2384130315129543e-06,
      "loss": 0.80992174,
      "num_input_tokens_seen": 113788635,
      "step": 5286,
      "time_per_iteration": 3.609495162963867
    },
    {
      "auxiliary_loss_clip": 0.01577718,
      "auxiliary_loss_mlp": 0.01035982,
      "balance_loss_clip": 1.0377872,
      "balance_loss_mlp": 1.02857625,
      "epoch": 0.6357241628088739,
      "flos": 18111978074880.0,
      "grad_norm": 2.2767841656411343,
      "language_loss": 0.73352516,
      "learning_rate": 1.2376928052867447e-06,
      "loss": 0.75966215,
      "num_input_tokens_seen": 113807755,
      "step": 5287,
      "time_per_iteration": 3.0749175548553467
    },
    {
      "auxiliary_loss_clip": 0.01279211,
      "auxiliary_loss_mlp": 0.01027614,
      "balance_loss_clip": 1.05007708,
      "balance_loss_mlp": 1.02047062,
      "epoch": 0.6358444056995131,
      "flos": 24935256599040.0,
      "grad_norm": 4.622598863238778,
      "language_loss": 0.77545387,
      "learning_rate": 1.2369726947019299e-06,
      "loss": 0.79852211,
      "num_input_tokens_seen": 113828230,
      "step": 5288,
      "time_per_iteration": 3.867518663406372
    },
    {
      "auxiliary_loss_clip": 0.01228078,
      "auxiliary_loss_mlp": 0.01023677,
      "balance_loss_clip": 1.04720545,
      "balance_loss_mlp": 1.01660764,
      "epoch": 0.6359646485901521,
      "flos": 23293348986240.0,
      "grad_norm": 2.6290673734833123,
      "language_loss": 0.67141831,
      "learning_rate": 1.2362526998677511e-06,
      "loss": 0.69393587,
      "num_input_tokens_seen": 113844595,
      "step": 5289,
      "time_per_iteration": 2.738909959793091
    },
    {
      "auxiliary_loss_clip": 0.0128417,
      "auxiliary_loss_mlp": 0.01029213,
      "balance_loss_clip": 1.04931843,
      "balance_loss_mlp": 1.02213228,
      "epoch": 0.6360848914807912,
      "flos": 20887442069760.0,
      "grad_norm": 1.958280311406575,
      "language_loss": 0.84136766,
      "learning_rate": 1.2355328208934301e-06,
      "loss": 0.86450148,
      "num_input_tokens_seen": 113863470,
      "step": 5290,
      "time_per_iteration": 2.723400831222534
    },
    {
      "auxiliary_loss_clip": 0.01230447,
      "auxiliary_loss_mlp": 0.02566996,
      "balance_loss_clip": 1.04834008,
      "balance_loss_mlp": 0.9999457,
      "epoch": 0.6362051343714303,
      "flos": 18479775386880.0,
      "grad_norm": 1.7548591254531911,
      "language_loss": 0.72765803,
      "learning_rate": 1.2348130578881728e-06,
      "loss": 0.76563245,
      "num_input_tokens_seen": 113881690,
      "step": 5291,
      "time_per_iteration": 3.5798356533050537
    },
    {
      "auxiliary_loss_clip": 0.01183418,
      "auxiliary_loss_mlp": 0.01026681,
      "balance_loss_clip": 1.0526371,
      "balance_loss_mlp": 1.01898599,
      "epoch": 0.6363253772620694,
      "flos": 24389594115840.0,
      "grad_norm": 3.976229429780192,
      "language_loss": 0.76433283,
      "learning_rate": 1.2340934109611664e-06,
      "loss": 0.78643382,
      "num_input_tokens_seen": 113902450,
      "step": 5292,
      "time_per_iteration": 2.659642457962036
    },
    {
      "auxiliary_loss_clip": 0.01290766,
      "auxiliary_loss_mlp": 0.01031627,
      "balance_loss_clip": 1.05176747,
      "balance_loss_mlp": 1.0234251,
      "epoch": 0.6364456201527084,
      "flos": 25958243940480.0,
      "grad_norm": 2.6556450841890316,
      "language_loss": 0.6845867,
      "learning_rate": 1.2333738802215798e-06,
      "loss": 0.70781064,
      "num_input_tokens_seen": 113922670,
      "step": 5293,
      "time_per_iteration": 2.7764925956726074
    },
    {
      "auxiliary_loss_clip": 0.01374372,
      "auxiliary_loss_mlp": 0.0102664,
      "balance_loss_clip": 1.04250097,
      "balance_loss_mlp": 1.01977086,
      "epoch": 0.6365658630433476,
      "flos": 20740711011840.0,
      "grad_norm": 1.8053087493572912,
      "language_loss": 0.81016296,
      "learning_rate": 1.2326544657785668e-06,
      "loss": 0.83417308,
      "num_input_tokens_seen": 113942360,
      "step": 5294,
      "time_per_iteration": 2.763246774673462
    },
    {
      "auxiliary_loss_clip": 0.01326943,
      "auxiliary_loss_mlp": 0.01030708,
      "balance_loss_clip": 1.04406643,
      "balance_loss_mlp": 1.0235498,
      "epoch": 0.6366861059339867,
      "flos": 21434146047360.0,
      "grad_norm": 3.2903584290296863,
      "language_loss": 0.74745631,
      "learning_rate": 1.2319351677412608e-06,
      "loss": 0.77103281,
      "num_input_tokens_seen": 113959405,
      "step": 5295,
      "time_per_iteration": 2.754897356033325
    },
    {
      "auxiliary_loss_clip": 0.01339776,
      "auxiliary_loss_mlp": 0.01025539,
      "balance_loss_clip": 1.0502311,
      "balance_loss_mlp": 1.0178138,
      "epoch": 0.6368063488246257,
      "flos": 22267093507200.0,
      "grad_norm": 3.0228096761891092,
      "language_loss": 0.74169385,
      "learning_rate": 1.2312159862187796e-06,
      "loss": 0.765347,
      "num_input_tokens_seen": 113977815,
      "step": 5296,
      "time_per_iteration": 2.7006120681762695
    },
    {
      "auxiliary_loss_clip": 0.01183189,
      "auxiliary_loss_mlp": 0.01033933,
      "balance_loss_clip": 1.0526979,
      "balance_loss_mlp": 1.02651787,
      "epoch": 0.6369265917152649,
      "flos": 22420719976320.0,
      "grad_norm": 1.57614599811213,
      "language_loss": 0.76277381,
      "learning_rate": 1.2304969213202217e-06,
      "loss": 0.78494507,
      "num_input_tokens_seen": 113999075,
      "step": 5297,
      "time_per_iteration": 2.6557743549346924
    },
    {
      "auxiliary_loss_clip": 0.01277922,
      "auxiliary_loss_mlp": 0.01024208,
      "balance_loss_clip": 1.04710948,
      "balance_loss_mlp": 1.01710296,
      "epoch": 0.6370468346059039,
      "flos": 24718176754560.0,
      "grad_norm": 2.881974843776227,
      "language_loss": 0.79669458,
      "learning_rate": 1.2297779731546692e-06,
      "loss": 0.81971586,
      "num_input_tokens_seen": 114018170,
      "step": 5298,
      "time_per_iteration": 2.6683907508850098
    },
    {
      "auxiliary_loss_clip": 0.01275798,
      "auxiliary_loss_mlp": 0.01031109,
      "balance_loss_clip": 1.04994333,
      "balance_loss_mlp": 1.02375984,
      "epoch": 0.637167077496543,
      "flos": 25296589463040.0,
      "grad_norm": 3.6600555054667767,
      "language_loss": 0.78095341,
      "learning_rate": 1.2290591418311853e-06,
      "loss": 0.80402249,
      "num_input_tokens_seen": 114035565,
      "step": 5299,
      "time_per_iteration": 2.7273497581481934
    },
    {
      "auxiliary_loss_clip": 0.01236541,
      "auxiliary_loss_mlp": 0.01025339,
      "balance_loss_clip": 1.05429935,
      "balance_loss_mlp": 1.01797748,
      "epoch": 0.637287320387182,
      "flos": 27671110871040.0,
      "grad_norm": 1.7119532676464184,
      "language_loss": 0.72348732,
      "learning_rate": 1.2283404274588172e-06,
      "loss": 0.74610615,
      "num_input_tokens_seen": 114054510,
      "step": 5300,
      "time_per_iteration": 2.704105854034424
    },
    {
      "auxiliary_loss_clip": 0.01379952,
      "auxiliary_loss_mlp": 0.01003266,
      "balance_loss_clip": 1.01305842,
      "balance_loss_mlp": 1.00230014,
      "epoch": 0.6374075632778212,
      "flos": 63173406873600.0,
      "grad_norm": 0.7424864654086032,
      "language_loss": 0.52722347,
      "learning_rate": 1.227621830146592e-06,
      "loss": 0.55105567,
      "num_input_tokens_seen": 114109875,
      "step": 5301,
      "time_per_iteration": 3.32707142829895
    },
    {
      "auxiliary_loss_clip": 0.01340744,
      "auxiliary_loss_mlp": 0.0103116,
      "balance_loss_clip": 1.05211771,
      "balance_loss_mlp": 1.02355456,
      "epoch": 0.6375278061684603,
      "flos": 25558127366400.0,
      "grad_norm": 1.8781768836186468,
      "language_loss": 0.78994977,
      "learning_rate": 1.2269033500035217e-06,
      "loss": 0.81366879,
      "num_input_tokens_seen": 114130010,
      "step": 5302,
      "time_per_iteration": 2.9176735877990723
    },
    {
      "auxiliary_loss_clip": 0.01227879,
      "auxiliary_loss_mlp": 0.01029359,
      "balance_loss_clip": 1.04790187,
      "balance_loss_mlp": 1.02178335,
      "epoch": 0.6376480490590993,
      "flos": 25666362023040.0,
      "grad_norm": 7.050854867166453,
      "language_loss": 0.73587561,
      "learning_rate": 1.2261849871385988e-06,
      "loss": 0.75844795,
      "num_input_tokens_seen": 114151115,
      "step": 5303,
      "time_per_iteration": 2.6849355697631836
    },
    {
      "auxiliary_loss_clip": 0.0117909,
      "auxiliary_loss_mlp": 0.01024813,
      "balance_loss_clip": 1.04996634,
      "balance_loss_mlp": 1.01760101,
      "epoch": 0.6377682919497385,
      "flos": 31537684350720.0,
      "grad_norm": 2.375572084102643,
      "language_loss": 0.6293323,
      "learning_rate": 1.2254667416607972e-06,
      "loss": 0.65137136,
      "num_input_tokens_seen": 114172715,
      "step": 5304,
      "time_per_iteration": 2.7112810611724854
    },
    {
      "auxiliary_loss_clip": 0.01226877,
      "auxiliary_loss_mlp": 0.01029648,
      "balance_loss_clip": 1.04875851,
      "balance_loss_mlp": 1.02211964,
      "epoch": 0.6378885348403776,
      "flos": 23039209284480.0,
      "grad_norm": 2.4206745456629153,
      "language_loss": 0.83108926,
      "learning_rate": 1.2247486136790756e-06,
      "loss": 0.8536545,
      "num_input_tokens_seen": 114192195,
      "step": 5305,
      "time_per_iteration": 2.651848554611206
    },
    {
      "auxiliary_loss_clip": 0.01233971,
      "auxiliary_loss_mlp": 0.01029945,
      "balance_loss_clip": 1.05219877,
      "balance_loss_mlp": 1.02261353,
      "epoch": 0.6380087777310166,
      "flos": 18697070712960.0,
      "grad_norm": 2.3200955888049486,
      "language_loss": 0.80098975,
      "learning_rate": 1.2240306033023726e-06,
      "loss": 0.82362884,
      "num_input_tokens_seen": 114210020,
      "step": 5306,
      "time_per_iteration": 2.6729204654693604
    },
    {
      "auxiliary_loss_clip": 0.01328401,
      "auxiliary_loss_mlp": 0.01024998,
      "balance_loss_clip": 1.04139209,
      "balance_loss_mlp": 1.01733899,
      "epoch": 0.6381290206216558,
      "flos": 23331558078720.0,
      "grad_norm": 1.8287691826387416,
      "language_loss": 0.71887654,
      "learning_rate": 1.223312710639611e-06,
      "loss": 0.74241054,
      "num_input_tokens_seen": 114228740,
      "step": 5307,
      "time_per_iteration": 2.6990292072296143
    },
    {
      "auxiliary_loss_clip": 0.0128004,
      "auxiliary_loss_mlp": 0.01025798,
      "balance_loss_clip": 1.05002117,
      "balance_loss_mlp": 1.01839852,
      "epoch": 0.6382492635122948,
      "flos": 18880466578560.0,
      "grad_norm": 9.148277954893828,
      "language_loss": 0.86992294,
      "learning_rate": 1.2225949357996928e-06,
      "loss": 0.89298135,
      "num_input_tokens_seen": 114246865,
      "step": 5308,
      "time_per_iteration": 2.6827967166900635
    },
    {
      "auxiliary_loss_clip": 0.01224944,
      "auxiliary_loss_mlp": 0.0102385,
      "balance_loss_clip": 1.04951203,
      "balance_loss_mlp": 1.01681399,
      "epoch": 0.6383695064029339,
      "flos": 27819134818560.0,
      "grad_norm": 1.6276328091586416,
      "language_loss": 0.80488217,
      "learning_rate": 1.221877278891505e-06,
      "loss": 0.82737017,
      "num_input_tokens_seen": 114266120,
      "step": 5309,
      "time_per_iteration": 3.9019930362701416
    },
    {
      "auxiliary_loss_clip": 0.01136217,
      "auxiliary_loss_mlp": 0.01028107,
      "balance_loss_clip": 1.05216789,
      "balance_loss_mlp": 1.02027524,
      "epoch": 0.638489749293573,
      "flos": 26395635853440.0,
      "grad_norm": 2.1254113099500227,
      "language_loss": 0.71100497,
      "learning_rate": 1.221159740023915e-06,
      "loss": 0.73264819,
      "num_input_tokens_seen": 114285950,
      "step": 5310,
      "time_per_iteration": 3.627065658569336
    },
    {
      "auxiliary_loss_clip": 0.01340415,
      "auxiliary_loss_mlp": 0.02572309,
      "balance_loss_clip": 1.05035043,
      "balance_loss_mlp": 0.99994391,
      "epoch": 0.6386099921842121,
      "flos": 23988328306560.0,
      "grad_norm": 1.9645677604340959,
      "language_loss": 0.72886324,
      "learning_rate": 1.2204423193057735e-06,
      "loss": 0.76799047,
      "num_input_tokens_seen": 114304780,
      "step": 5311,
      "time_per_iteration": 2.745328187942505
    },
    {
      "auxiliary_loss_clip": 0.01171094,
      "auxiliary_loss_mlp": 0.01006057,
      "balance_loss_clip": 1.00769019,
      "balance_loss_mlp": 1.00510311,
      "epoch": 0.6387302350748512,
      "flos": 71731169337600.0,
      "grad_norm": 0.8453016799504164,
      "language_loss": 0.63358885,
      "learning_rate": 1.2197250168459122e-06,
      "loss": 0.65536034,
      "num_input_tokens_seen": 114361180,
      "step": 5312,
      "time_per_iteration": 4.310146808624268
    },
    {
      "auxiliary_loss_clip": 0.01230352,
      "auxiliary_loss_mlp": 0.01022881,
      "balance_loss_clip": 1.04808497,
      "balance_loss_mlp": 1.01598477,
      "epoch": 0.6388504779654903,
      "flos": 14535778141440.0,
      "grad_norm": 2.070153284371048,
      "language_loss": 0.74610418,
      "learning_rate": 1.2190078327531454e-06,
      "loss": 0.76863647,
      "num_input_tokens_seen": 114377425,
      "step": 5313,
      "time_per_iteration": 2.6409835815429688
    },
    {
      "auxiliary_loss_clip": 0.01230427,
      "auxiliary_loss_mlp": 0.01029252,
      "balance_loss_clip": 1.04962349,
      "balance_loss_mlp": 1.02264488,
      "epoch": 0.6389707208561294,
      "flos": 22346133384960.0,
      "grad_norm": 2.074891438118566,
      "language_loss": 0.7294749,
      "learning_rate": 1.2182907671362697e-06,
      "loss": 0.75207162,
      "num_input_tokens_seen": 114398120,
      "step": 5314,
      "time_per_iteration": 2.6993775367736816
    },
    {
      "auxiliary_loss_clip": 0.01224753,
      "auxiliary_loss_mlp": 0.01025753,
      "balance_loss_clip": 1.05029035,
      "balance_loss_mlp": 1.01813591,
      "epoch": 0.6390909637467684,
      "flos": 19426883247360.0,
      "grad_norm": 2.155721714967466,
      "language_loss": 0.78890157,
      "learning_rate": 1.2175738201040626e-06,
      "loss": 0.81140661,
      "num_input_tokens_seen": 114415160,
      "step": 5315,
      "time_per_iteration": 2.591069459915161
    },
    {
      "auxiliary_loss_clip": 0.01228544,
      "auxiliary_loss_mlp": 0.01034898,
      "balance_loss_clip": 1.05011809,
      "balance_loss_mlp": 1.02724779,
      "epoch": 0.6392112066374076,
      "flos": 24090852700800.0,
      "grad_norm": 1.8746230010381824,
      "language_loss": 0.78611332,
      "learning_rate": 1.2168569917652855e-06,
      "loss": 0.80874777,
      "num_input_tokens_seen": 114435015,
      "step": 5316,
      "time_per_iteration": 2.707594633102417
    },
    {
      "auxiliary_loss_clip": 0.01225242,
      "auxiliary_loss_mlp": 0.01029437,
      "balance_loss_clip": 1.04828978,
      "balance_loss_mlp": 1.02234375,
      "epoch": 0.6393314495280467,
      "flos": 26795141896320.0,
      "grad_norm": 9.432821893734403,
      "language_loss": 0.63907576,
      "learning_rate": 1.2161402822286797e-06,
      "loss": 0.66162258,
      "num_input_tokens_seen": 114455700,
      "step": 5317,
      "time_per_iteration": 3.5787999629974365
    },
    {
      "auxiliary_loss_clip": 0.01329658,
      "auxiliary_loss_mlp": 0.01023797,
      "balance_loss_clip": 1.0481993,
      "balance_loss_mlp": 1.017133,
      "epoch": 0.6394516924186857,
      "flos": 20260692633600.0,
      "grad_norm": 2.3702539679344388,
      "language_loss": 0.79382002,
      "learning_rate": 1.2154236916029703e-06,
      "loss": 0.81735456,
      "num_input_tokens_seen": 114473675,
      "step": 5318,
      "time_per_iteration": 2.683284044265747
    },
    {
      "auxiliary_loss_clip": 0.01376356,
      "auxiliary_loss_mlp": 0.01031012,
      "balance_loss_clip": 1.04133296,
      "balance_loss_mlp": 1.02366006,
      "epoch": 0.6395719353093249,
      "flos": 18368847210240.0,
      "grad_norm": 2.4637363562091554,
      "language_loss": 0.73552692,
      "learning_rate": 1.2147072199968627e-06,
      "loss": 0.75960052,
      "num_input_tokens_seen": 114492310,
      "step": 5319,
      "time_per_iteration": 2.7734501361846924
    },
    {
      "auxiliary_loss_clip": 0.01226509,
      "auxiliary_loss_mlp": 0.01029856,
      "balance_loss_clip": 1.04923844,
      "balance_loss_mlp": 1.02301908,
      "epoch": 0.6396921781999639,
      "flos": 17566315591680.0,
      "grad_norm": 1.862433203879789,
      "language_loss": 0.71987844,
      "learning_rate": 1.2139908675190454e-06,
      "loss": 0.74244213,
      "num_input_tokens_seen": 114511520,
      "step": 5320,
      "time_per_iteration": 2.5943119525909424
    },
    {
      "auxiliary_loss_clip": 0.01416625,
      "auxiliary_loss_mlp": 0.0102649,
      "balance_loss_clip": 1.04022503,
      "balance_loss_mlp": 1.01908147,
      "epoch": 0.639812421090603,
      "flos": 21251252972160.0,
      "grad_norm": 2.7623210186736347,
      "language_loss": 0.75253767,
      "learning_rate": 1.2132746342781883e-06,
      "loss": 0.77696884,
      "num_input_tokens_seen": 114532680,
      "step": 5321,
      "time_per_iteration": 2.8414394855499268
    },
    {
      "auxiliary_loss_clip": 0.01179229,
      "auxiliary_loss_mlp": 0.01027898,
      "balance_loss_clip": 1.05036378,
      "balance_loss_mlp": 1.02044797,
      "epoch": 0.6399326639812422,
      "flos": 11180967684480.0,
      "grad_norm": 3.047870338701369,
      "language_loss": 0.80004287,
      "learning_rate": 1.2125585203829442e-06,
      "loss": 0.82211411,
      "num_input_tokens_seen": 114548320,
      "step": 5322,
      "time_per_iteration": 2.550372838973999
    },
    {
      "auxiliary_loss_clip": 0.01316651,
      "auxiliary_loss_mlp": 0.01027944,
      "balance_loss_clip": 1.04529929,
      "balance_loss_mlp": 1.02055264,
      "epoch": 0.6400529068718812,
      "flos": 23911048195200.0,
      "grad_norm": 1.8283726804263651,
      "language_loss": 0.74109644,
      "learning_rate": 1.211842525941946e-06,
      "loss": 0.76454234,
      "num_input_tokens_seen": 114568115,
      "step": 5323,
      "time_per_iteration": 2.7672312259674072
    },
    {
      "auxiliary_loss_clip": 0.01369604,
      "auxiliary_loss_mlp": 0.01026705,
      "balance_loss_clip": 1.04580283,
      "balance_loss_mlp": 1.01953447,
      "epoch": 0.6401731497625203,
      "flos": 44018724890880.0,
      "grad_norm": 1.825817199706977,
      "language_loss": 0.79146278,
      "learning_rate": 1.2111266510638105e-06,
      "loss": 0.81542587,
      "num_input_tokens_seen": 114591040,
      "step": 5324,
      "time_per_iteration": 3.008723735809326
    },
    {
      "auxiliary_loss_clip": 0.01415866,
      "auxiliary_loss_mlp": 0.01026504,
      "balance_loss_clip": 1.0408721,
      "balance_loss_mlp": 1.01885414,
      "epoch": 0.6402933926531594,
      "flos": 20662209838080.0,
      "grad_norm": 2.140801100020516,
      "language_loss": 0.80098367,
      "learning_rate": 1.2104108958571346e-06,
      "loss": 0.82540739,
      "num_input_tokens_seen": 114609310,
      "step": 5325,
      "time_per_iteration": 2.79460072517395
    },
    {
      "auxiliary_loss_clip": 0.01223413,
      "auxiliary_loss_mlp": 0.01032883,
      "balance_loss_clip": 1.04979348,
      "balance_loss_mlp": 1.02576935,
      "epoch": 0.6404136355437985,
      "flos": 24863327614080.0,
      "grad_norm": 1.458621368437152,
      "language_loss": 0.75491369,
      "learning_rate": 1.2096952604304975e-06,
      "loss": 0.77747667,
      "num_input_tokens_seen": 114629740,
      "step": 5326,
      "time_per_iteration": 2.7047767639160156
    },
    {
      "auxiliary_loss_clip": 0.01229211,
      "auxiliary_loss_mlp": 0.01026812,
      "balance_loss_clip": 1.04833412,
      "balance_loss_mlp": 1.01932561,
      "epoch": 0.6405338784344375,
      "flos": 40479548901120.0,
      "grad_norm": 2.2676051287849788,
      "language_loss": 0.70444167,
      "learning_rate": 1.2089797448924616e-06,
      "loss": 0.72700179,
      "num_input_tokens_seen": 114653615,
      "step": 5327,
      "time_per_iteration": 2.7771782875061035
    },
    {
      "auxiliary_loss_clip": 0.0128388,
      "auxiliary_loss_mlp": 0.01032524,
      "balance_loss_clip": 1.04107738,
      "balance_loss_mlp": 1.02472186,
      "epoch": 0.6406541213250767,
      "flos": 20886041439360.0,
      "grad_norm": 6.254847309407252,
      "language_loss": 0.66257668,
      "learning_rate": 1.2082643493515692e-06,
      "loss": 0.68574071,
      "num_input_tokens_seen": 114671935,
      "step": 5328,
      "time_per_iteration": 2.7917044162750244
    },
    {
      "auxiliary_loss_clip": 0.01227695,
      "auxiliary_loss_mlp": 0.01024791,
      "balance_loss_clip": 1.05037832,
      "balance_loss_mlp": 1.01734066,
      "epoch": 0.6407743642157158,
      "flos": 23295970679040.0,
      "grad_norm": 2.0696929043392256,
      "language_loss": 0.82075852,
      "learning_rate": 1.207549073916346e-06,
      "loss": 0.84328336,
      "num_input_tokens_seen": 114692870,
      "step": 5329,
      "time_per_iteration": 2.6529016494750977
    },
    {
      "auxiliary_loss_clip": 0.01272538,
      "auxiliary_loss_mlp": 0.0102621,
      "balance_loss_clip": 1.04760504,
      "balance_loss_mlp": 1.01924539,
      "epoch": 0.6408946071063548,
      "flos": 15012636122880.0,
      "grad_norm": 2.1317507050475566,
      "language_loss": 0.78030324,
      "learning_rate": 1.2068339186952976e-06,
      "loss": 0.80329078,
      "num_input_tokens_seen": 114710410,
      "step": 5330,
      "time_per_iteration": 2.6093974113464355
    },
    {
      "auxiliary_loss_clip": 0.01231919,
      "auxiliary_loss_mlp": 0.01028015,
      "balance_loss_clip": 1.05038023,
      "balance_loss_mlp": 1.02081156,
      "epoch": 0.6410148499969939,
      "flos": 22528595496960.0,
      "grad_norm": 1.8263896813610085,
      "language_loss": 0.73330009,
      "learning_rate": 1.2061188837969136e-06,
      "loss": 0.75589937,
      "num_input_tokens_seen": 114730020,
      "step": 5331,
      "time_per_iteration": 2.645265579223633
    },
    {
      "auxiliary_loss_clip": 0.01322756,
      "auxiliary_loss_mlp": 0.01029349,
      "balance_loss_clip": 1.04172075,
      "balance_loss_mlp": 1.02195764,
      "epoch": 0.641135092887633,
      "flos": 12422004537600.0,
      "grad_norm": 3.9640093134470415,
      "language_loss": 0.84484422,
      "learning_rate": 1.2054039693296631e-06,
      "loss": 0.86836535,
      "num_input_tokens_seen": 114748015,
      "step": 5332,
      "time_per_iteration": 2.6965737342834473
    },
    {
      "auxiliary_loss_clip": 0.01330345,
      "auxiliary_loss_mlp": 0.01027537,
      "balance_loss_clip": 1.04777288,
      "balance_loss_mlp": 1.02029836,
      "epoch": 0.6412553357782721,
      "flos": 22127329687680.0,
      "grad_norm": 2.284399082318156,
      "language_loss": 0.81534171,
      "learning_rate": 1.2046891754019992e-06,
      "loss": 0.83892053,
      "num_input_tokens_seen": 114768625,
      "step": 5333,
      "time_per_iteration": 2.6886987686157227
    },
    {
      "auxiliary_loss_clip": 0.01232679,
      "auxiliary_loss_mlp": 0.01030663,
      "balance_loss_clip": 1.05097818,
      "balance_loss_mlp": 1.02349544,
      "epoch": 0.6413755786689112,
      "flos": 15888605097600.0,
      "grad_norm": 2.0349333454385454,
      "language_loss": 0.82725275,
      "learning_rate": 1.2039745021223548e-06,
      "loss": 0.84988606,
      "num_input_tokens_seen": 114786045,
      "step": 5334,
      "time_per_iteration": 2.65878963470459
    },
    {
      "auxiliary_loss_clip": 0.01270238,
      "auxiliary_loss_mlp": 0.01003842,
      "balance_loss_clip": 1.00811672,
      "balance_loss_mlp": 1.00296021,
      "epoch": 0.6414958215595503,
      "flos": 68039159955840.0,
      "grad_norm": 0.7910189274494626,
      "language_loss": 0.57009828,
      "learning_rate": 1.2032599495991456e-06,
      "loss": 0.59283912,
      "num_input_tokens_seen": 114850785,
      "step": 5335,
      "time_per_iteration": 4.429994344711304
    },
    {
      "auxiliary_loss_clip": 0.01225924,
      "auxiliary_loss_mlp": 0.01026514,
      "balance_loss_clip": 1.04986441,
      "balance_loss_mlp": 1.01916444,
      "epoch": 0.6416160644501894,
      "flos": 44091300320640.0,
      "grad_norm": 1.9210795253401722,
      "language_loss": 0.69570345,
      "learning_rate": 1.2025455179407685e-06,
      "loss": 0.71822786,
      "num_input_tokens_seen": 114871945,
      "step": 5336,
      "time_per_iteration": 3.7220423221588135
    },
    {
      "auxiliary_loss_clip": 0.01223077,
      "auxiliary_loss_mlp": 0.02567919,
      "balance_loss_clip": 1.04731226,
      "balance_loss_mlp": 0.99993193,
      "epoch": 0.6417363073408284,
      "flos": 20959837931520.0,
      "grad_norm": 3.1876058874053483,
      "language_loss": 0.73820984,
      "learning_rate": 1.2018312072556022e-06,
      "loss": 0.77611983,
      "num_input_tokens_seen": 114890445,
      "step": 5337,
      "time_per_iteration": 2.653197765350342
    },
    {
      "auxiliary_loss_clip": 0.01178015,
      "auxiliary_loss_mlp": 0.0256696,
      "balance_loss_clip": 1.05112588,
      "balance_loss_mlp": 0.99994761,
      "epoch": 0.6418565502314676,
      "flos": 22455122227200.0,
      "grad_norm": 2.95149149268209,
      "language_loss": 0.74899352,
      "learning_rate": 1.2011170176520077e-06,
      "loss": 0.78644323,
      "num_input_tokens_seen": 114911360,
      "step": 5338,
      "time_per_iteration": 3.463630199432373
    },
    {
      "auxiliary_loss_clip": 0.01409338,
      "auxiliary_loss_mlp": 0.01023384,
      "balance_loss_clip": 1.03995204,
      "balance_loss_mlp": 1.01663327,
      "epoch": 0.6419767931221066,
      "flos": 25045502417280.0,
      "grad_norm": 1.5414715759350448,
      "language_loss": 0.81390941,
      "learning_rate": 1.2004029492383256e-06,
      "loss": 0.83823669,
      "num_input_tokens_seen": 114932700,
      "step": 5339,
      "time_per_iteration": 2.8309359550476074
    },
    {
      "auxiliary_loss_clip": 0.01227818,
      "auxiliary_loss_mlp": 0.01026233,
      "balance_loss_clip": 1.05111265,
      "balance_loss_mlp": 1.01877379,
      "epoch": 0.6420970360127457,
      "flos": 19463691709440.0,
      "grad_norm": 2.0765680730243252,
      "language_loss": 0.73506361,
      "learning_rate": 1.1996890021228814e-06,
      "loss": 0.75760412,
      "num_input_tokens_seen": 114949475,
      "step": 5340,
      "time_per_iteration": 2.6715340614318848
    },
    {
      "auxiliary_loss_clip": 0.01273984,
      "auxiliary_loss_mlp": 0.0103054,
      "balance_loss_clip": 1.0447886,
      "balance_loss_mlp": 1.02360857,
      "epoch": 0.6422172789033849,
      "flos": 40406147458560.0,
      "grad_norm": 1.8266681848311934,
      "language_loss": 0.69944024,
      "learning_rate": 1.1989751764139785e-06,
      "loss": 0.72248548,
      "num_input_tokens_seen": 114973125,
      "step": 5341,
      "time_per_iteration": 2.9005038738250732
    },
    {
      "auxiliary_loss_clip": 0.01375219,
      "auxiliary_loss_mlp": 0.01028541,
      "balance_loss_clip": 1.03998959,
      "balance_loss_mlp": 1.02163231,
      "epoch": 0.6423375217940239,
      "flos": 27672870637440.0,
      "grad_norm": 1.7497912918768581,
      "language_loss": 0.83254623,
      "learning_rate": 1.1982614722199044e-06,
      "loss": 0.85658377,
      "num_input_tokens_seen": 114994300,
      "step": 5342,
      "time_per_iteration": 2.788990020751953
    },
    {
      "auxiliary_loss_clip": 0.01281391,
      "auxiliary_loss_mlp": 0.01027931,
      "balance_loss_clip": 1.04532671,
      "balance_loss_mlp": 1.02093959,
      "epoch": 0.642457764684663,
      "flos": 18369242259840.0,
      "grad_norm": 2.4343447924437114,
      "language_loss": 0.77818197,
      "learning_rate": 1.1975478896489276e-06,
      "loss": 0.80127519,
      "num_input_tokens_seen": 115012135,
      "step": 5343,
      "time_per_iteration": 3.593946695327759
    },
    {
      "auxiliary_loss_clip": 0.01176639,
      "auxiliary_loss_mlp": 0.01025318,
      "balance_loss_clip": 1.04941106,
      "balance_loss_mlp": 1.01855016,
      "epoch": 0.6425780075753021,
      "flos": 19750509809280.0,
      "grad_norm": 2.306817174436465,
      "language_loss": 0.76225352,
      "learning_rate": 1.1968344288092981e-06,
      "loss": 0.78427309,
      "num_input_tokens_seen": 115028715,
      "step": 5344,
      "time_per_iteration": 2.5889034271240234
    },
    {
      "auxiliary_loss_clip": 0.01226182,
      "auxiliary_loss_mlp": 0.02566624,
      "balance_loss_clip": 1.04989588,
      "balance_loss_mlp": 0.99998045,
      "epoch": 0.6426982504659412,
      "flos": 20558536208640.0,
      "grad_norm": 1.9843278051346342,
      "language_loss": 0.64798766,
      "learning_rate": 1.1961210898092468e-06,
      "loss": 0.68591571,
      "num_input_tokens_seen": 115047665,
      "step": 5345,
      "time_per_iteration": 2.669842481613159
    },
    {
      "auxiliary_loss_clip": 0.01283959,
      "auxiliary_loss_mlp": 0.01032671,
      "balance_loss_clip": 1.05046904,
      "balance_loss_mlp": 1.0252831,
      "epoch": 0.6428184933565803,
      "flos": 17851984456320.0,
      "grad_norm": 4.141509608895586,
      "language_loss": 0.79022843,
      "learning_rate": 1.1954078727569874e-06,
      "loss": 0.81339478,
      "num_input_tokens_seen": 115064965,
      "step": 5346,
      "time_per_iteration": 2.6981024742126465
    },
    {
      "auxiliary_loss_clip": 0.01332893,
      "auxiliary_loss_mlp": 0.02565682,
      "balance_loss_clip": 1.04620123,
      "balance_loss_mlp": 0.99994588,
      "epoch": 0.6429387362472194,
      "flos": 22456953820800.0,
      "grad_norm": 1.607009078671201,
      "language_loss": 0.78207588,
      "learning_rate": 1.1946947777607141e-06,
      "loss": 0.82106167,
      "num_input_tokens_seen": 115086100,
      "step": 5347,
      "time_per_iteration": 2.7820684909820557
    },
    {
      "auxiliary_loss_clip": 0.01370123,
      "auxiliary_loss_mlp": 0.01030775,
      "balance_loss_clip": 1.04205418,
      "balance_loss_mlp": 1.02306819,
      "epoch": 0.6430589791378585,
      "flos": 24752579005440.0,
      "grad_norm": 1.9588895379546838,
      "language_loss": 0.80246627,
      "learning_rate": 1.1939818049286024e-06,
      "loss": 0.82647526,
      "num_input_tokens_seen": 115104260,
      "step": 5348,
      "time_per_iteration": 2.8068578243255615
    },
    {
      "auxiliary_loss_clip": 0.01414496,
      "auxiliary_loss_mlp": 0.01029484,
      "balance_loss_clip": 1.04117012,
      "balance_loss_mlp": 1.02190185,
      "epoch": 0.6431792220284975,
      "flos": 24901249397760.0,
      "grad_norm": 1.671189822093965,
      "language_loss": 0.75499475,
      "learning_rate": 1.1932689543688101e-06,
      "loss": 0.77943456,
      "num_input_tokens_seen": 115125365,
      "step": 5349,
      "time_per_iteration": 2.8449716567993164
    },
    {
      "auxiliary_loss_clip": 0.01275471,
      "auxiliary_loss_mlp": 0.01025915,
      "balance_loss_clip": 1.0487442,
      "balance_loss_mlp": 1.01873875,
      "epoch": 0.6432994649191367,
      "flos": 21032305620480.0,
      "grad_norm": 1.9031050842654673,
      "language_loss": 0.72462022,
      "learning_rate": 1.1925562261894756e-06,
      "loss": 0.74763405,
      "num_input_tokens_seen": 115144445,
      "step": 5350,
      "time_per_iteration": 2.707843542098999
    },
    {
      "auxiliary_loss_clip": 0.01274651,
      "auxiliary_loss_mlp": 0.0102889,
      "balance_loss_clip": 1.04629564,
      "balance_loss_mlp": 1.02175248,
      "epoch": 0.6434197078097758,
      "flos": 30884433655680.0,
      "grad_norm": 1.8083097374239043,
      "language_loss": 0.77531129,
      "learning_rate": 1.1918436204987207e-06,
      "loss": 0.79834664,
      "num_input_tokens_seen": 115166305,
      "step": 5351,
      "time_per_iteration": 2.8311479091644287
    },
    {
      "auxiliary_loss_clip": 0.01231923,
      "auxiliary_loss_mlp": 0.01029401,
      "balance_loss_clip": 1.05327809,
      "balance_loss_mlp": 1.02210212,
      "epoch": 0.6435399507004148,
      "flos": 15012492468480.0,
      "grad_norm": 2.2759310312235175,
      "language_loss": 0.82223141,
      "learning_rate": 1.191131137404645e-06,
      "loss": 0.84484458,
      "num_input_tokens_seen": 115183045,
      "step": 5352,
      "time_per_iteration": 2.7492055892944336
    },
    {
      "auxiliary_loss_clip": 0.01321933,
      "auxiliary_loss_mlp": 0.01030043,
      "balance_loss_clip": 1.04541469,
      "balance_loss_mlp": 1.02266967,
      "epoch": 0.643660193591054,
      "flos": 19901981462400.0,
      "grad_norm": 1.8575178301040514,
      "language_loss": 0.77082479,
      "learning_rate": 1.190418777015333e-06,
      "loss": 0.79434454,
      "num_input_tokens_seen": 115201955,
      "step": 5353,
      "time_per_iteration": 2.666023015975952
    },
    {
      "auxiliary_loss_clip": 0.01276569,
      "auxiliary_loss_mlp": 0.01025903,
      "balance_loss_clip": 1.04735518,
      "balance_loss_mlp": 1.01902175,
      "epoch": 0.643780436481693,
      "flos": 24133622820480.0,
      "grad_norm": 2.02905636660288,
      "language_loss": 0.73628998,
      "learning_rate": 1.1897065394388487e-06,
      "loss": 0.75931472,
      "num_input_tokens_seen": 115222395,
      "step": 5354,
      "time_per_iteration": 2.817009449005127
    },
    {
      "auxiliary_loss_clip": 0.01278498,
      "auxiliary_loss_mlp": 0.01033406,
      "balance_loss_clip": 1.05293894,
      "balance_loss_mlp": 1.0261457,
      "epoch": 0.6439006793723321,
      "flos": 23148808657920.0,
      "grad_norm": 1.6293159810393822,
      "language_loss": 0.76637399,
      "learning_rate": 1.1889944247832385e-06,
      "loss": 0.78949296,
      "num_input_tokens_seen": 115242635,
      "step": 5355,
      "time_per_iteration": 2.6723968982696533
    },
    {
      "auxiliary_loss_clip": 0.01229441,
      "auxiliary_loss_mlp": 0.01029315,
      "balance_loss_clip": 1.04646063,
      "balance_loss_mlp": 1.02213001,
      "epoch": 0.6440209222629713,
      "flos": 23617909301760.0,
      "grad_norm": 2.088411334325203,
      "language_loss": 0.70783746,
      "learning_rate": 1.1882824331565283e-06,
      "loss": 0.730425,
      "num_input_tokens_seen": 115262095,
      "step": 5356,
      "time_per_iteration": 2.682340621948242
    },
    {
      "auxiliary_loss_clip": 0.01327074,
      "auxiliary_loss_mlp": 0.01026426,
      "balance_loss_clip": 1.04334235,
      "balance_loss_mlp": 1.01956844,
      "epoch": 0.6441411651536103,
      "flos": 16544872535040.0,
      "grad_norm": 2.2736883216079637,
      "language_loss": 0.88878089,
      "learning_rate": 1.1875705646667287e-06,
      "loss": 0.91231585,
      "num_input_tokens_seen": 115279985,
      "step": 5357,
      "time_per_iteration": 2.7104852199554443
    },
    {
      "auxiliary_loss_clip": 0.0122643,
      "auxiliary_loss_mlp": 0.01019169,
      "balance_loss_clip": 1.047521,
      "balance_loss_mlp": 1.0120461,
      "epoch": 0.6442614080442494,
      "flos": 25410965345280.0,
      "grad_norm": 2.2511954163330747,
      "language_loss": 0.7511307,
      "learning_rate": 1.1868588194218282e-06,
      "loss": 0.77358663,
      "num_input_tokens_seen": 115300365,
      "step": 5358,
      "time_per_iteration": 2.708192825317383
    },
    {
      "auxiliary_loss_clip": 0.01282283,
      "auxiliary_loss_mlp": 0.01034072,
      "balance_loss_clip": 1.04663801,
      "balance_loss_mlp": 1.02644265,
      "epoch": 0.6443816509348885,
      "flos": 28294017552000.0,
      "grad_norm": 1.6925277234657443,
      "language_loss": 0.73895746,
      "learning_rate": 1.1861471975297979e-06,
      "loss": 0.76212096,
      "num_input_tokens_seen": 115322060,
      "step": 5359,
      "time_per_iteration": 2.8025894165039062
    },
    {
      "auxiliary_loss_clip": 0.01321892,
      "auxiliary_loss_mlp": 0.01030421,
      "balance_loss_clip": 1.04891777,
      "balance_loss_mlp": 1.02312505,
      "epoch": 0.6445018938255276,
      "flos": 36690075964800.0,
      "grad_norm": 2.206159393892351,
      "language_loss": 0.71043539,
      "learning_rate": 1.185435699098591e-06,
      "loss": 0.7339586,
      "num_input_tokens_seen": 115348255,
      "step": 5360,
      "time_per_iteration": 2.8623263835906982
    },
    {
      "auxiliary_loss_clip": 0.01279517,
      "auxiliary_loss_mlp": 0.01027589,
      "balance_loss_clip": 1.04837704,
      "balance_loss_mlp": 1.02044201,
      "epoch": 0.6446221367161666,
      "flos": 14501411804160.0,
      "grad_norm": 2.2598051967358534,
      "language_loss": 0.78921866,
      "learning_rate": 1.1847243242361403e-06,
      "loss": 0.81228971,
      "num_input_tokens_seen": 115366845,
      "step": 5361,
      "time_per_iteration": 3.6350767612457275
    },
    {
      "auxiliary_loss_clip": 0.01286049,
      "auxiliary_loss_mlp": 0.0102714,
      "balance_loss_clip": 1.05044973,
      "balance_loss_mlp": 1.02001977,
      "epoch": 0.6447423796068057,
      "flos": 24609367480320.0,
      "grad_norm": 1.7891082558215068,
      "language_loss": 0.7755214,
      "learning_rate": 1.1840130730503624e-06,
      "loss": 0.7986533,
      "num_input_tokens_seen": 115388125,
      "step": 5362,
      "time_per_iteration": 3.5857667922973633
    },
    {
      "auxiliary_loss_clip": 0.01179367,
      "auxiliary_loss_mlp": 0.01028556,
      "balance_loss_clip": 1.05046284,
      "balance_loss_mlp": 1.02104914,
      "epoch": 0.6448626224974449,
      "flos": 25047298097280.0,
      "grad_norm": 1.7660727028670387,
      "language_loss": 0.74972528,
      "learning_rate": 1.1833019456491518e-06,
      "loss": 0.77180451,
      "num_input_tokens_seen": 115409655,
      "step": 5363,
      "time_per_iteration": 2.6265995502471924
    },
    {
      "auxiliary_loss_clip": 0.0122944,
      "auxiliary_loss_mlp": 0.0102619,
      "balance_loss_clip": 1.05025554,
      "balance_loss_mlp": 1.01930559,
      "epoch": 0.6449828653880839,
      "flos": 22530355263360.0,
      "grad_norm": 8.985661777713332,
      "language_loss": 0.78447819,
      "learning_rate": 1.1825909421403871e-06,
      "loss": 0.80703443,
      "num_input_tokens_seen": 115428750,
      "step": 5364,
      "time_per_iteration": 2.6992149353027344
    },
    {
      "auxiliary_loss_clip": 0.01230321,
      "auxiliary_loss_mlp": 0.01028907,
      "balance_loss_clip": 1.05063748,
      "balance_loss_mlp": 1.02196872,
      "epoch": 0.645103108278723,
      "flos": 25695736369920.0,
      "grad_norm": 6.476611330423354,
      "language_loss": 0.76526976,
      "learning_rate": 1.181880062631926e-06,
      "loss": 0.78786206,
      "num_input_tokens_seen": 115448085,
      "step": 5365,
      "time_per_iteration": 3.556283473968506
    },
    {
      "auxiliary_loss_clip": 0.01277299,
      "auxiliary_loss_mlp": 0.01027111,
      "balance_loss_clip": 1.04844809,
      "balance_loss_mlp": 1.01898694,
      "epoch": 0.6452233511693621,
      "flos": 27450331925760.0,
      "grad_norm": 2.531395611345167,
      "language_loss": 0.85035801,
      "learning_rate": 1.1811693072316093e-06,
      "loss": 0.87340206,
      "num_input_tokens_seen": 115465765,
      "step": 5366,
      "time_per_iteration": 2.6717875003814697
    },
    {
      "auxiliary_loss_clip": 0.01180461,
      "auxiliary_loss_mlp": 0.02568662,
      "balance_loss_clip": 1.05140781,
      "balance_loss_mlp": 0.99996006,
      "epoch": 0.6453435940600012,
      "flos": 19208618254080.0,
      "grad_norm": 2.688501113884709,
      "language_loss": 0.83988339,
      "learning_rate": 1.1804586760472574e-06,
      "loss": 0.87737465,
      "num_input_tokens_seen": 115482230,
      "step": 5367,
      "time_per_iteration": 2.558027744293213
    },
    {
      "auxiliary_loss_clip": 0.01325121,
      "auxiliary_loss_mlp": 0.01025032,
      "balance_loss_clip": 1.04642391,
      "balance_loss_mlp": 1.01818347,
      "epoch": 0.6454638369506402,
      "flos": 25737680476800.0,
      "grad_norm": 2.4442832692307532,
      "language_loss": 0.80593359,
      "learning_rate": 1.1797481691866736e-06,
      "loss": 0.82943511,
      "num_input_tokens_seen": 115499455,
      "step": 5368,
      "time_per_iteration": 2.8676185607910156
    },
    {
      "auxiliary_loss_clip": 0.01279141,
      "auxiliary_loss_mlp": 0.01025732,
      "balance_loss_clip": 1.05091083,
      "balance_loss_mlp": 1.01790047,
      "epoch": 0.6455840798412794,
      "flos": 20989176364800.0,
      "grad_norm": 2.312555763473352,
      "language_loss": 0.83032167,
      "learning_rate": 1.1790377867576393e-06,
      "loss": 0.85337043,
      "num_input_tokens_seen": 115517205,
      "step": 5369,
      "time_per_iteration": 3.9158096313476562
    },
    {
      "auxiliary_loss_clip": 0.0127847,
      "auxiliary_loss_mlp": 0.0103164,
      "balance_loss_clip": 1.04567003,
      "balance_loss_mlp": 1.02427578,
      "epoch": 0.6457043227319185,
      "flos": 26067556005120.0,
      "grad_norm": 2.675385620108355,
      "language_loss": 0.76739192,
      "learning_rate": 1.1783275288679203e-06,
      "loss": 0.79049301,
      "num_input_tokens_seen": 115534370,
      "step": 5370,
      "time_per_iteration": 2.6878902912139893
    },
    {
      "auxiliary_loss_clip": 0.01121286,
      "auxiliary_loss_mlp": 0.0100246,
      "balance_loss_clip": 1.00901008,
      "balance_loss_mlp": 1.00145233,
      "epoch": 0.6458245656225575,
      "flos": 60370831088640.0,
      "grad_norm": 0.8457571277618732,
      "language_loss": 0.57114094,
      "learning_rate": 1.177617395625262e-06,
      "loss": 0.59237844,
      "num_input_tokens_seen": 115592345,
      "step": 5371,
      "time_per_iteration": 3.1920483112335205
    },
    {
      "auxiliary_loss_clip": 0.01225855,
      "auxiliary_loss_mlp": 0.01029658,
      "balance_loss_clip": 1.04978991,
      "balance_loss_mlp": 1.02219522,
      "epoch": 0.6459448085131967,
      "flos": 23076771932160.0,
      "grad_norm": 1.9086016240199255,
      "language_loss": 0.75413489,
      "learning_rate": 1.1769073871373908e-06,
      "loss": 0.77669001,
      "num_input_tokens_seen": 115612550,
      "step": 5372,
      "time_per_iteration": 2.6086840629577637
    },
    {
      "auxiliary_loss_clip": 0.01327057,
      "auxiliary_loss_mlp": 0.01023506,
      "balance_loss_clip": 1.04460764,
      "balance_loss_mlp": 1.01577187,
      "epoch": 0.6460650514038357,
      "flos": 22598190097920.0,
      "grad_norm": 1.6636978835221703,
      "language_loss": 0.83597755,
      "learning_rate": 1.176197503512015e-06,
      "loss": 0.85948324,
      "num_input_tokens_seen": 115632265,
      "step": 5373,
      "time_per_iteration": 2.704448938369751
    },
    {
      "auxiliary_loss_clip": 0.01278112,
      "auxiliary_loss_mlp": 0.01027707,
      "balance_loss_clip": 1.0479672,
      "balance_loss_mlp": 1.0212276,
      "epoch": 0.6461852942944748,
      "flos": 20266726118400.0,
      "grad_norm": 2.013839878944226,
      "language_loss": 0.81944561,
      "learning_rate": 1.1754877448568223e-06,
      "loss": 0.84250379,
      "num_input_tokens_seen": 115651720,
      "step": 5374,
      "time_per_iteration": 2.67710018157959
    },
    {
      "auxiliary_loss_clip": 0.01276058,
      "auxiliary_loss_mlp": 0.01026299,
      "balance_loss_clip": 1.04772878,
      "balance_loss_mlp": 1.01907802,
      "epoch": 0.646305537185114,
      "flos": 23367109564800.0,
      "grad_norm": 2.3404292361603374,
      "language_loss": 0.90275306,
      "learning_rate": 1.1747781112794837e-06,
      "loss": 0.9257766,
      "num_input_tokens_seen": 115668215,
      "step": 5375,
      "time_per_iteration": 2.6170248985290527
    },
    {
      "auxiliary_loss_clip": 0.01327311,
      "auxiliary_loss_mlp": 0.01029558,
      "balance_loss_clip": 1.04671907,
      "balance_loss_mlp": 1.02237546,
      "epoch": 0.646425780075753,
      "flos": 24277480790400.0,
      "grad_norm": 1.6096175715666006,
      "language_loss": 0.83236104,
      "learning_rate": 1.1740686028876487e-06,
      "loss": 0.85592973,
      "num_input_tokens_seen": 115687080,
      "step": 5376,
      "time_per_iteration": 2.718742609024048
    },
    {
      "auxiliary_loss_clip": 0.01225902,
      "auxiliary_loss_mlp": 0.01022982,
      "balance_loss_clip": 1.05187333,
      "balance_loss_mlp": 1.01616359,
      "epoch": 0.6465460229663921,
      "flos": 20813968800000.0,
      "grad_norm": 2.8525027511356766,
      "language_loss": 0.74911296,
      "learning_rate": 1.1733592197889507e-06,
      "loss": 0.77160186,
      "num_input_tokens_seen": 115703990,
      "step": 5377,
      "time_per_iteration": 2.6101062297821045
    },
    {
      "auxiliary_loss_clip": 0.01222714,
      "auxiliary_loss_mlp": 0.01024353,
      "balance_loss_clip": 1.04948461,
      "balance_loss_mlp": 1.0180794,
      "epoch": 0.6466662658570312,
      "flos": 22853299466880.0,
      "grad_norm": 2.565692617394344,
      "language_loss": 0.72882169,
      "learning_rate": 1.1726499620910014e-06,
      "loss": 0.75129235,
      "num_input_tokens_seen": 115724270,
      "step": 5378,
      "time_per_iteration": 2.6990389823913574
    },
    {
      "auxiliary_loss_clip": 0.01225325,
      "auxiliary_loss_mlp": 0.01028289,
      "balance_loss_clip": 1.04901457,
      "balance_loss_mlp": 1.02061796,
      "epoch": 0.6467865087476703,
      "flos": 15304553953920.0,
      "grad_norm": 3.430765639134419,
      "language_loss": 0.78471041,
      "learning_rate": 1.1719408299013955e-06,
      "loss": 0.80724645,
      "num_input_tokens_seen": 115742995,
      "step": 5379,
      "time_per_iteration": 2.6492514610290527
    },
    {
      "auxiliary_loss_clip": 0.01179078,
      "auxiliary_loss_mlp": 0.01027233,
      "balance_loss_clip": 1.05277228,
      "balance_loss_mlp": 1.01989245,
      "epoch": 0.6469067516383094,
      "flos": 19573650218880.0,
      "grad_norm": 2.2659522421839844,
      "language_loss": 0.76389015,
      "learning_rate": 1.1712318233277067e-06,
      "loss": 0.78595328,
      "num_input_tokens_seen": 115762015,
      "step": 5380,
      "time_per_iteration": 2.6937906742095947
    },
    {
      "auxiliary_loss_clip": 0.0112144,
      "auxiliary_loss_mlp": 0.01003221,
      "balance_loss_clip": 1.01114058,
      "balance_loss_mlp": 1.00214827,
      "epoch": 0.6470269945289485,
      "flos": 65098002522240.0,
      "grad_norm": 0.75160815782135,
      "language_loss": 0.57840848,
      "learning_rate": 1.1705229424774916e-06,
      "loss": 0.59965515,
      "num_input_tokens_seen": 115816285,
      "step": 5381,
      "time_per_iteration": 3.045072317123413
    },
    {
      "auxiliary_loss_clip": 0.01277431,
      "auxiliary_loss_mlp": 0.01034649,
      "balance_loss_clip": 1.0471741,
      "balance_loss_mlp": 1.02754402,
      "epoch": 0.6471472374195876,
      "flos": 30696943639680.0,
      "grad_norm": 2.0202776617357094,
      "language_loss": 0.64344656,
      "learning_rate": 1.1698141874582867e-06,
      "loss": 0.66656744,
      "num_input_tokens_seen": 115837330,
      "step": 5382,
      "time_per_iteration": 2.7180795669555664
    },
    {
      "auxiliary_loss_clip": 0.01178821,
      "auxiliary_loss_mlp": 0.01028386,
      "balance_loss_clip": 1.05262566,
      "balance_loss_mlp": 1.02101922,
      "epoch": 0.6472674803102266,
      "flos": 20521835487360.0,
      "grad_norm": 2.034748749924749,
      "language_loss": 0.72390306,
      "learning_rate": 1.169105558377609e-06,
      "loss": 0.74597514,
      "num_input_tokens_seen": 115857420,
      "step": 5383,
      "time_per_iteration": 2.5870606899261475
    },
    {
      "auxiliary_loss_clip": 0.01278582,
      "auxiliary_loss_mlp": 0.02565853,
      "balance_loss_clip": 1.05117393,
      "balance_loss_mlp": 0.99992359,
      "epoch": 0.6473877232008658,
      "flos": 24715447320960.0,
      "grad_norm": 2.708943284296548,
      "language_loss": 0.78717041,
      "learning_rate": 1.1683970553429587e-06,
      "loss": 0.82561475,
      "num_input_tokens_seen": 115878875,
      "step": 5384,
      "time_per_iteration": 2.7570509910583496
    },
    {
      "auxiliary_loss_clip": 0.01329172,
      "auxiliary_loss_mlp": 0.01026639,
      "balance_loss_clip": 1.04736793,
      "balance_loss_mlp": 1.0194422,
      "epoch": 0.6475079660915048,
      "flos": 15885552441600.0,
      "grad_norm": 5.653869737408974,
      "language_loss": 0.82087165,
      "learning_rate": 1.1676886784618128e-06,
      "loss": 0.84442973,
      "num_input_tokens_seen": 115895540,
      "step": 5385,
      "time_per_iteration": 2.654784917831421
    },
    {
      "auxiliary_loss_clip": 0.01226996,
      "auxiliary_loss_mlp": 0.01029774,
      "balance_loss_clip": 1.04971743,
      "balance_loss_mlp": 1.02268362,
      "epoch": 0.6476282089821439,
      "flos": 17381590922880.0,
      "grad_norm": 2.5817260941130957,
      "language_loss": 0.83703548,
      "learning_rate": 1.1669804278416332e-06,
      "loss": 0.85960317,
      "num_input_tokens_seen": 115910265,
      "step": 5386,
      "time_per_iteration": 2.6135501861572266
    },
    {
      "auxiliary_loss_clip": 0.01280067,
      "auxiliary_loss_mlp": 0.01022618,
      "balance_loss_clip": 1.04822266,
      "balance_loss_mlp": 1.01509607,
      "epoch": 0.6477484518727831,
      "flos": 20194078861440.0,
      "grad_norm": 1.9821179943034815,
      "language_loss": 0.71588248,
      "learning_rate": 1.1662723035898602e-06,
      "loss": 0.73890936,
      "num_input_tokens_seen": 115930025,
      "step": 5387,
      "time_per_iteration": 3.600111484527588
    },
    {
      "auxiliary_loss_clip": 0.01227508,
      "auxiliary_loss_mlp": 0.01023685,
      "balance_loss_clip": 1.05032039,
      "balance_loss_mlp": 1.01644325,
      "epoch": 0.6478686947634221,
      "flos": 25410426641280.0,
      "grad_norm": 2.1140001741649836,
      "language_loss": 0.82215172,
      "learning_rate": 1.165564305813915e-06,
      "loss": 0.84466362,
      "num_input_tokens_seen": 115949025,
      "step": 5388,
      "time_per_iteration": 3.5675952434539795
    },
    {
      "auxiliary_loss_clip": 0.01225622,
      "auxiliary_loss_mlp": 0.01022907,
      "balance_loss_clip": 1.0497731,
      "balance_loss_mlp": 1.01607597,
      "epoch": 0.6479889376540612,
      "flos": 20083581648000.0,
      "grad_norm": 1.856256842358605,
      "language_loss": 0.8139317,
      "learning_rate": 1.1648564346212019e-06,
      "loss": 0.83641696,
      "num_input_tokens_seen": 115968145,
      "step": 5389,
      "time_per_iteration": 2.6461098194122314
    },
    {
      "auxiliary_loss_clip": 0.01222353,
      "auxiliary_loss_mlp": 0.0102845,
      "balance_loss_clip": 1.05021536,
      "balance_loss_mlp": 1.02160764,
      "epoch": 0.6481091805447003,
      "flos": 26758082039040.0,
      "grad_norm": 2.5120302109855177,
      "language_loss": 0.75994474,
      "learning_rate": 1.164148690119104e-06,
      "loss": 0.78245282,
      "num_input_tokens_seen": 115989425,
      "step": 5390,
      "time_per_iteration": 3.6221113204956055
    },
    {
      "auxiliary_loss_clip": 0.01175864,
      "auxiliary_loss_mlp": 0.01026678,
      "balance_loss_clip": 1.05009258,
      "balance_loss_mlp": 1.01981747,
      "epoch": 0.6482294234353394,
      "flos": 23952094462080.0,
      "grad_norm": 1.9658036618130383,
      "language_loss": 0.74014735,
      "learning_rate": 1.163441072414985e-06,
      "loss": 0.76217276,
      "num_input_tokens_seen": 116009630,
      "step": 5391,
      "time_per_iteration": 2.580493688583374
    },
    {
      "auxiliary_loss_clip": 0.01225376,
      "auxiliary_loss_mlp": 0.01025615,
      "balance_loss_clip": 1.05036151,
      "balance_loss_mlp": 1.01870072,
      "epoch": 0.6483496663259785,
      "flos": 26209833776640.0,
      "grad_norm": 1.7519620705742083,
      "language_loss": 0.70102823,
      "learning_rate": 1.16273358161619e-06,
      "loss": 0.72353816,
      "num_input_tokens_seen": 116029965,
      "step": 5392,
      "time_per_iteration": 2.697293758392334
    },
    {
      "auxiliary_loss_clip": 0.01188735,
      "auxiliary_loss_mlp": 0.01025642,
      "balance_loss_clip": 1.05256462,
      "balance_loss_mlp": 1.01875782,
      "epoch": 0.6484699092166175,
      "flos": 20922239370240.0,
      "grad_norm": 2.1035885663388534,
      "language_loss": 0.83482474,
      "learning_rate": 1.1620262178300446e-06,
      "loss": 0.85696846,
      "num_input_tokens_seen": 116048580,
      "step": 5393,
      "time_per_iteration": 2.6464905738830566
    },
    {
      "auxiliary_loss_clip": 0.01326949,
      "auxiliary_loss_mlp": 0.01021068,
      "balance_loss_clip": 1.04554951,
      "balance_loss_mlp": 1.01434433,
      "epoch": 0.6485901521072567,
      "flos": 33072865678080.0,
      "grad_norm": 1.860060521558231,
      "language_loss": 0.76033175,
      "learning_rate": 1.1613189811638563e-06,
      "loss": 0.78381193,
      "num_input_tokens_seen": 116070305,
      "step": 5394,
      "time_per_iteration": 2.825573444366455
    },
    {
      "auxiliary_loss_clip": 0.01229912,
      "auxiliary_loss_mlp": 0.010252,
      "balance_loss_clip": 1.05102873,
      "balance_loss_mlp": 1.0184319,
      "epoch": 0.6487103949978957,
      "flos": 22274060745600.0,
      "grad_norm": 1.9498922523056847,
      "language_loss": 0.7799421,
      "learning_rate": 1.1606118717249117e-06,
      "loss": 0.80249321,
      "num_input_tokens_seen": 116090405,
      "step": 5395,
      "time_per_iteration": 3.601853132247925
    },
    {
      "auxiliary_loss_clip": 0.01182433,
      "auxiliary_loss_mlp": 0.01026461,
      "balance_loss_clip": 1.05101943,
      "balance_loss_mlp": 1.0188849,
      "epoch": 0.6488306378885348,
      "flos": 22930400010240.0,
      "grad_norm": 1.9683529651889502,
      "language_loss": 0.67961425,
      "learning_rate": 1.1599048896204787e-06,
      "loss": 0.70170319,
      "num_input_tokens_seen": 116110285,
      "step": 5396,
      "time_per_iteration": 2.5810952186584473
    },
    {
      "auxiliary_loss_clip": 0.0133643,
      "auxiliary_loss_mlp": 0.01028174,
      "balance_loss_clip": 1.0488646,
      "balance_loss_mlp": 1.0209738,
      "epoch": 0.648950880779174,
      "flos": 20376110010240.0,
      "grad_norm": 2.952466828239897,
      "language_loss": 0.81185448,
      "learning_rate": 1.1591980349578061e-06,
      "loss": 0.83550048,
      "num_input_tokens_seen": 116128955,
      "step": 5397,
      "time_per_iteration": 2.733032464981079
    },
    {
      "auxiliary_loss_clip": 0.01224289,
      "auxiliary_loss_mlp": 0.01000193,
      "balance_loss_clip": 1.009619,
      "balance_loss_mlp": 0.99914378,
      "epoch": 0.649071123669813,
      "flos": 59930889310080.0,
      "grad_norm": 0.7276237160985991,
      "language_loss": 0.5426712,
      "learning_rate": 1.158491307844123e-06,
      "loss": 0.56491601,
      "num_input_tokens_seen": 116188875,
      "step": 5398,
      "time_per_iteration": 3.208956718444824
    },
    {
      "auxiliary_loss_clip": 0.01275602,
      "auxiliary_loss_mlp": 0.01027188,
      "balance_loss_clip": 1.04791427,
      "balance_loss_mlp": 1.02047694,
      "epoch": 0.6491913665604521,
      "flos": 20446566537600.0,
      "grad_norm": 1.594479712874919,
      "language_loss": 0.84181064,
      "learning_rate": 1.1577847083866387e-06,
      "loss": 0.86483854,
      "num_input_tokens_seen": 116207910,
      "step": 5399,
      "time_per_iteration": 2.755687713623047
    },
    {
      "auxiliary_loss_clip": 0.01270548,
      "auxiliary_loss_mlp": 0.01025655,
      "balance_loss_clip": 1.04645693,
      "balance_loss_mlp": 1.01830542,
      "epoch": 0.6493116094510912,
      "flos": 16946820702720.0,
      "grad_norm": 2.3257036348853446,
      "language_loss": 0.72403407,
      "learning_rate": 1.1570782366925453e-06,
      "loss": 0.74699616,
      "num_input_tokens_seen": 116226425,
      "step": 5400,
      "time_per_iteration": 2.64321231842041
    },
    {
      "auxiliary_loss_clip": 0.0127907,
      "auxiliary_loss_mlp": 0.01026415,
      "balance_loss_clip": 1.04522979,
      "balance_loss_mlp": 1.01923871,
      "epoch": 0.6494318523417303,
      "flos": 18802935072000.0,
      "grad_norm": 2.068440738899054,
      "language_loss": 0.75223321,
      "learning_rate": 1.1563718928690132e-06,
      "loss": 0.77528805,
      "num_input_tokens_seen": 116243860,
      "step": 5401,
      "time_per_iteration": 2.6680045127868652
    },
    {
      "auxiliary_loss_clip": 0.01323928,
      "auxiliary_loss_mlp": 0.01023995,
      "balance_loss_clip": 1.04893732,
      "balance_loss_mlp": 1.01659226,
      "epoch": 0.6495520952323693,
      "flos": 18982847318400.0,
      "grad_norm": 2.0755369378996487,
      "language_loss": 0.71829271,
      "learning_rate": 1.1556656770231942e-06,
      "loss": 0.74177194,
      "num_input_tokens_seen": 116260055,
      "step": 5402,
      "time_per_iteration": 2.7182188034057617
    },
    {
      "auxiliary_loss_clip": 0.01226846,
      "auxiliary_loss_mlp": 0.01028273,
      "balance_loss_clip": 1.04714775,
      "balance_loss_mlp": 1.02162766,
      "epoch": 0.6496723381230085,
      "flos": 22745388032640.0,
      "grad_norm": 1.4878543450274582,
      "language_loss": 0.75991094,
      "learning_rate": 1.1549595892622207e-06,
      "loss": 0.78246212,
      "num_input_tokens_seen": 116278825,
      "step": 5403,
      "time_per_iteration": 2.6820600032806396
    },
    {
      "auxiliary_loss_clip": 0.01270153,
      "auxiliary_loss_mlp": 0.01002272,
      "balance_loss_clip": 1.01197076,
      "balance_loss_mlp": 1.00128865,
      "epoch": 0.6497925810136476,
      "flos": 62145283887360.0,
      "grad_norm": 0.8843696729932652,
      "language_loss": 0.58970755,
      "learning_rate": 1.1542536296932047e-06,
      "loss": 0.61243176,
      "num_input_tokens_seen": 116342360,
      "step": 5404,
      "time_per_iteration": 3.232889413833618
    },
    {
      "auxiliary_loss_clip": 0.01332371,
      "auxiliary_loss_mlp": 0.01028273,
      "balance_loss_clip": 1.04594159,
      "balance_loss_mlp": 1.02045298,
      "epoch": 0.6499128239042866,
      "flos": 20156731695360.0,
      "grad_norm": 1.9754430256653954,
      "language_loss": 0.69986326,
      "learning_rate": 1.1535477984232414e-06,
      "loss": 0.72346967,
      "num_input_tokens_seen": 116362235,
      "step": 5405,
      "time_per_iteration": 2.721411943435669
    },
    {
      "auxiliary_loss_clip": 0.01275345,
      "auxiliary_loss_mlp": 0.01026069,
      "balance_loss_clip": 1.04001045,
      "balance_loss_mlp": 1.01903284,
      "epoch": 0.6500330667949258,
      "flos": 24462420940800.0,
      "grad_norm": 1.9615277044579729,
      "language_loss": 0.76846433,
      "learning_rate": 1.152842095559404e-06,
      "loss": 0.79147851,
      "num_input_tokens_seen": 116382895,
      "step": 5406,
      "time_per_iteration": 2.8004376888275146
    },
    {
      "auxiliary_loss_clip": 0.01278252,
      "auxiliary_loss_mlp": 0.01032687,
      "balance_loss_clip": 1.04437375,
      "balance_loss_mlp": 1.02621651,
      "epoch": 0.6501533096855648,
      "flos": 25477399549440.0,
      "grad_norm": 4.636066553665009,
      "language_loss": 0.76876181,
      "learning_rate": 1.1521365212087474e-06,
      "loss": 0.79187119,
      "num_input_tokens_seen": 116402880,
      "step": 5407,
      "time_per_iteration": 2.69944429397583
    },
    {
      "auxiliary_loss_clip": 0.0122621,
      "auxiliary_loss_mlp": 0.01026341,
      "balance_loss_clip": 1.04747748,
      "balance_loss_mlp": 1.01921797,
      "epoch": 0.6502735525762039,
      "flos": 44819245347840.0,
      "grad_norm": 1.6003861585554322,
      "language_loss": 0.70790243,
      "learning_rate": 1.1514310754783062e-06,
      "loss": 0.73042792,
      "num_input_tokens_seen": 116425830,
      "step": 5408,
      "time_per_iteration": 2.8331291675567627
    },
    {
      "auxiliary_loss_clip": 0.01279082,
      "auxiliary_loss_mlp": 0.01031352,
      "balance_loss_clip": 1.0480547,
      "balance_loss_mlp": 1.02453601,
      "epoch": 0.6503937954668431,
      "flos": 28658546726400.0,
      "grad_norm": 2.3551354867753505,
      "language_loss": 0.73427004,
      "learning_rate": 1.1507257584750964e-06,
      "loss": 0.75737441,
      "num_input_tokens_seen": 116446010,
      "step": 5409,
      "time_per_iteration": 2.690790891647339
    },
    {
      "auxiliary_loss_clip": 0.01179674,
      "auxiliary_loss_mlp": 0.01032601,
      "balance_loss_clip": 1.05183864,
      "balance_loss_mlp": 1.02451324,
      "epoch": 0.6505140383574821,
      "flos": 20922562592640.0,
      "grad_norm": 1.9458585331150509,
      "language_loss": 0.7731567,
      "learning_rate": 1.150020570306113e-06,
      "loss": 0.7952795,
      "num_input_tokens_seen": 116465150,
      "step": 5410,
      "time_per_iteration": 2.6193864345550537
    },
    {
      "auxiliary_loss_clip": 0.01272256,
      "auxiliary_loss_mlp": 0.01029034,
      "balance_loss_clip": 1.04225099,
      "balance_loss_mlp": 1.02168202,
      "epoch": 0.6506342812481212,
      "flos": 20595236929920.0,
      "grad_norm": 1.9165760361507775,
      "language_loss": 0.74948072,
      "learning_rate": 1.1493155110783338e-06,
      "loss": 0.77249366,
      "num_input_tokens_seen": 116483675,
      "step": 5411,
      "time_per_iteration": 2.650876045227051
    },
    {
      "auxiliary_loss_clip": 0.0122683,
      "auxiliary_loss_mlp": 0.01028527,
      "balance_loss_clip": 1.04938745,
      "balance_loss_mlp": 1.02172649,
      "epoch": 0.6507545241387603,
      "flos": 30226478279040.0,
      "grad_norm": 2.5517627452400564,
      "language_loss": 0.70543474,
      "learning_rate": 1.1486105808987155e-06,
      "loss": 0.7279883,
      "num_input_tokens_seen": 116505165,
      "step": 5412,
      "time_per_iteration": 2.6841177940368652
    },
    {
      "auxiliary_loss_clip": 0.01228533,
      "auxiliary_loss_mlp": 0.01028387,
      "balance_loss_clip": 1.05018401,
      "balance_loss_mlp": 1.02093041,
      "epoch": 0.6508747670293994,
      "flos": 17128241320320.0,
      "grad_norm": 1.928626003285675,
      "language_loss": 0.81366372,
      "learning_rate": 1.1479057798741947e-06,
      "loss": 0.8362329,
      "num_input_tokens_seen": 116523220,
      "step": 5413,
      "time_per_iteration": 3.6132447719573975
    },
    {
      "auxiliary_loss_clip": 0.01187007,
      "auxiliary_loss_mlp": 0.01001746,
      "balance_loss_clip": 1.02009058,
      "balance_loss_mlp": 1.00075102,
      "epoch": 0.6509950099200384,
      "flos": 68559826573440.0,
      "grad_norm": 0.7806073404423542,
      "language_loss": 0.53261048,
      "learning_rate": 1.14720110811169e-06,
      "loss": 0.55449796,
      "num_input_tokens_seen": 116580450,
      "step": 5414,
      "time_per_iteration": 4.147408723831177
    },
    {
      "auxiliary_loss_clip": 0.01229998,
      "auxiliary_loss_mlp": 0.01028585,
      "balance_loss_clip": 1.05064404,
      "balance_loss_mlp": 1.02112818,
      "epoch": 0.6511152528106776,
      "flos": 22347462188160.0,
      "grad_norm": 4.955999817011386,
      "language_loss": 0.76824355,
      "learning_rate": 1.146496565718098e-06,
      "loss": 0.79082942,
      "num_input_tokens_seen": 116601020,
      "step": 5415,
      "time_per_iteration": 2.655353546142578
    },
    {
      "auxiliary_loss_clip": 0.01285383,
      "auxiliary_loss_mlp": 0.01031146,
      "balance_loss_clip": 1.05400395,
      "balance_loss_mlp": 1.0241785,
      "epoch": 0.6512354957013167,
      "flos": 20522158709760.0,
      "grad_norm": 2.120412173402629,
      "language_loss": 0.75945699,
      "learning_rate": 1.1457921528002996e-06,
      "loss": 0.78262222,
      "num_input_tokens_seen": 116619455,
      "step": 5416,
      "time_per_iteration": 3.587359666824341
    },
    {
      "auxiliary_loss_clip": 0.01179805,
      "auxiliary_loss_mlp": 0.02566258,
      "balance_loss_clip": 1.05088329,
      "balance_loss_mlp": 0.99996865,
      "epoch": 0.6513557385919557,
      "flos": 32337342881280.0,
      "grad_norm": 2.6100917975246962,
      "language_loss": 0.72412491,
      "learning_rate": 1.1450878694651522e-06,
      "loss": 0.76158559,
      "num_input_tokens_seen": 116640020,
      "step": 5417,
      "time_per_iteration": 2.679577112197876
    },
    {
      "auxiliary_loss_clip": 0.01374735,
      "auxiliary_loss_mlp": 0.01025889,
      "balance_loss_clip": 1.04128397,
      "balance_loss_mlp": 1.01844466,
      "epoch": 0.6514759814825949,
      "flos": 12093206417280.0,
      "grad_norm": 2.4622043516333028,
      "language_loss": 0.63214946,
      "learning_rate": 1.1443837158194954e-06,
      "loss": 0.65615571,
      "num_input_tokens_seen": 116655165,
      "step": 5418,
      "time_per_iteration": 2.7450952529907227
    },
    {
      "auxiliary_loss_clip": 0.01332207,
      "auxiliary_loss_mlp": 0.01034439,
      "balance_loss_clip": 1.05174291,
      "balance_loss_mlp": 1.02708364,
      "epoch": 0.651596224373234,
      "flos": 22526907557760.0,
      "grad_norm": 1.7840738608253814,
      "language_loss": 0.74496222,
      "learning_rate": 1.1436796919701484e-06,
      "loss": 0.76862872,
      "num_input_tokens_seen": 116673880,
      "step": 5419,
      "time_per_iteration": 2.688143730163574
    },
    {
      "auxiliary_loss_clip": 0.01285005,
      "auxiliary_loss_mlp": 0.01032247,
      "balance_loss_clip": 1.05306101,
      "balance_loss_mlp": 1.0252049,
      "epoch": 0.651716467263873,
      "flos": 27818955250560.0,
      "grad_norm": 1.903405810962586,
      "language_loss": 0.62004596,
      "learning_rate": 1.1429757980239115e-06,
      "loss": 0.64321852,
      "num_input_tokens_seen": 116694305,
      "step": 5420,
      "time_per_iteration": 2.732534885406494
    },
    {
      "auxiliary_loss_clip": 0.01182413,
      "auxiliary_loss_mlp": 0.01028814,
      "balance_loss_clip": 1.05148578,
      "balance_loss_mlp": 1.02125621,
      "epoch": 0.6518367101545122,
      "flos": 24316300414080.0,
      "grad_norm": 6.480381621836686,
      "language_loss": 0.81825805,
      "learning_rate": 1.1422720340875636e-06,
      "loss": 0.8403703,
      "num_input_tokens_seen": 116713055,
      "step": 5421,
      "time_per_iteration": 3.5740420818328857
    },
    {
      "auxiliary_loss_clip": 0.0123635,
      "auxiliary_loss_mlp": 0.01027464,
      "balance_loss_clip": 1.05239463,
      "balance_loss_mlp": 1.0198878,
      "epoch": 0.6519569530451512,
      "flos": 20011939971840.0,
      "grad_norm": 14.699523670367348,
      "language_loss": 0.79339868,
      "learning_rate": 1.1415684002678671e-06,
      "loss": 0.81603682,
      "num_input_tokens_seen": 116731815,
      "step": 5422,
      "time_per_iteration": 2.6206202507019043
    },
    {
      "auxiliary_loss_clip": 0.01280842,
      "auxiliary_loss_mlp": 0.01024019,
      "balance_loss_clip": 1.04669034,
      "balance_loss_mlp": 1.0166043,
      "epoch": 0.6520771959357903,
      "flos": 21576064682880.0,
      "grad_norm": 2.326401052622643,
      "language_loss": 0.77597362,
      "learning_rate": 1.1408648966715617e-06,
      "loss": 0.7990222,
      "num_input_tokens_seen": 116749335,
      "step": 5423,
      "time_per_iteration": 2.6803457736968994
    },
    {
      "auxiliary_loss_clip": 0.01276693,
      "auxiliary_loss_mlp": 0.01029872,
      "balance_loss_clip": 1.04242778,
      "balance_loss_mlp": 1.02232325,
      "epoch": 0.6521974388264293,
      "flos": 22711021695360.0,
      "grad_norm": 2.854982404129945,
      "language_loss": 0.72517246,
      "learning_rate": 1.1401615234053683e-06,
      "loss": 0.74823809,
      "num_input_tokens_seen": 116768155,
      "step": 5424,
      "time_per_iteration": 2.6611742973327637
    },
    {
      "auxiliary_loss_clip": 0.01279635,
      "auxiliary_loss_mlp": 0.01023218,
      "balance_loss_clip": 1.04766488,
      "balance_loss_mlp": 1.01577938,
      "epoch": 0.6523176817170685,
      "flos": 23002939526400.0,
      "grad_norm": 1.8549520358833953,
      "language_loss": 0.75670898,
      "learning_rate": 1.1394582805759885e-06,
      "loss": 0.77973747,
      "num_input_tokens_seen": 116787435,
      "step": 5425,
      "time_per_iteration": 2.75006103515625
    },
    {
      "auxiliary_loss_clip": 0.01226582,
      "auxiliary_loss_mlp": 0.01027954,
      "balance_loss_clip": 1.05147076,
      "balance_loss_mlp": 1.02059639,
      "epoch": 0.6524379246077076,
      "flos": 21688249835520.0,
      "grad_norm": 1.7209484559770951,
      "language_loss": 0.75823462,
      "learning_rate": 1.1387551682901022e-06,
      "loss": 0.78078002,
      "num_input_tokens_seen": 116808040,
      "step": 5426,
      "time_per_iteration": 2.6242623329162598
    },
    {
      "auxiliary_loss_clip": 0.01323016,
      "auxiliary_loss_mlp": 0.01026447,
      "balance_loss_clip": 1.04593408,
      "balance_loss_mlp": 1.01950932,
      "epoch": 0.6525581674983466,
      "flos": 19390936711680.0,
      "grad_norm": 2.5557549654219467,
      "language_loss": 0.70803583,
      "learning_rate": 1.138052186654373e-06,
      "loss": 0.73153055,
      "num_input_tokens_seen": 116825510,
      "step": 5427,
      "time_per_iteration": 2.7227444648742676
    },
    {
      "auxiliary_loss_clip": 0.01281156,
      "auxiliary_loss_mlp": 0.01027443,
      "balance_loss_clip": 1.04778504,
      "balance_loss_mlp": 1.0198853,
      "epoch": 0.6526784103889858,
      "flos": 17165444832000.0,
      "grad_norm": 2.444798574118974,
      "language_loss": 0.87969351,
      "learning_rate": 1.1373493357754417e-06,
      "loss": 0.90277946,
      "num_input_tokens_seen": 116844415,
      "step": 5428,
      "time_per_iteration": 2.643268346786499
    },
    {
      "auxiliary_loss_clip": 0.01177988,
      "auxiliary_loss_mlp": 0.01025419,
      "balance_loss_clip": 1.04971027,
      "balance_loss_mlp": 1.01830816,
      "epoch": 0.6527986532796248,
      "flos": 18989168112000.0,
      "grad_norm": 1.8800330360825315,
      "language_loss": 0.77596676,
      "learning_rate": 1.1366466157599303e-06,
      "loss": 0.79800081,
      "num_input_tokens_seen": 116863690,
      "step": 5429,
      "time_per_iteration": 2.6175243854522705
    },
    {
      "auxiliary_loss_clip": 0.01373468,
      "auxiliary_loss_mlp": 0.02567064,
      "balance_loss_clip": 1.04483056,
      "balance_loss_mlp": 0.99996245,
      "epoch": 0.6529188961702639,
      "flos": 14238581011200.0,
      "grad_norm": 3.170309959805079,
      "language_loss": 0.7646805,
      "learning_rate": 1.1359440267144412e-06,
      "loss": 0.80408579,
      "num_input_tokens_seen": 116881145,
      "step": 5430,
      "time_per_iteration": 2.7027971744537354
    },
    {
      "auxiliary_loss_clip": 0.01230383,
      "auxiliary_loss_mlp": 0.0102524,
      "balance_loss_clip": 1.04906821,
      "balance_loss_mlp": 1.01809955,
      "epoch": 0.653039139060903,
      "flos": 36682929158400.0,
      "grad_norm": 3.0913279831993696,
      "language_loss": 0.74007869,
      "learning_rate": 1.1352415687455556e-06,
      "loss": 0.76263499,
      "num_input_tokens_seen": 116902405,
      "step": 5431,
      "time_per_iteration": 2.7686328887939453
    },
    {
      "auxiliary_loss_clip": 0.0122882,
      "auxiliary_loss_mlp": 0.01028993,
      "balance_loss_clip": 1.05229878,
      "balance_loss_mlp": 1.02137852,
      "epoch": 0.6531593819515421,
      "flos": 25376275785600.0,
      "grad_norm": 2.751591341834461,
      "language_loss": 0.63648117,
      "learning_rate": 1.1345392419598362e-06,
      "loss": 0.65905929,
      "num_input_tokens_seen": 116921285,
      "step": 5432,
      "time_per_iteration": 2.6842434406280518
    },
    {
      "auxiliary_loss_clip": 0.01225376,
      "auxiliary_loss_mlp": 0.0102624,
      "balance_loss_clip": 1.04724002,
      "balance_loss_mlp": 1.0192157,
      "epoch": 0.6532796248421812,
      "flos": 21178533888000.0,
      "grad_norm": 2.151936229129589,
      "language_loss": 0.71875679,
      "learning_rate": 1.1338370464638263e-06,
      "loss": 0.74127293,
      "num_input_tokens_seen": 116940685,
      "step": 5433,
      "time_per_iteration": 2.607154130935669
    },
    {
      "auxiliary_loss_clip": 0.01177789,
      "auxiliary_loss_mlp": 0.01023164,
      "balance_loss_clip": 1.0500828,
      "balance_loss_mlp": 1.01604104,
      "epoch": 0.6533998677328203,
      "flos": 17675950878720.0,
      "grad_norm": 2.502683706857755,
      "language_loss": 0.63423592,
      "learning_rate": 1.1331349823640474e-06,
      "loss": 0.65624541,
      "num_input_tokens_seen": 116958115,
      "step": 5434,
      "time_per_iteration": 2.5923633575439453
    },
    {
      "auxiliary_loss_clip": 0.01230486,
      "auxiliary_loss_mlp": 0.025613,
      "balance_loss_clip": 1.05113053,
      "balance_loss_mlp": 0.99998116,
      "epoch": 0.6535201106234594,
      "flos": 28400384701440.0,
      "grad_norm": 2.306020775860141,
      "language_loss": 0.78107834,
      "learning_rate": 1.132433049767003e-06,
      "loss": 0.81899619,
      "num_input_tokens_seen": 116976030,
      "step": 5435,
      "time_per_iteration": 2.7102584838867188
    },
    {
      "auxiliary_loss_clip": 0.01273893,
      "auxiliary_loss_mlp": 0.01027516,
      "balance_loss_clip": 1.0480907,
      "balance_loss_mlp": 1.02074742,
      "epoch": 0.6536403535140984,
      "flos": 23586667447680.0,
      "grad_norm": 1.5403289652308572,
      "language_loss": 0.81058347,
      "learning_rate": 1.1317312487791748e-06,
      "loss": 0.83359754,
      "num_input_tokens_seen": 116997680,
      "step": 5436,
      "time_per_iteration": 2.6879794597625732
    },
    {
      "auxiliary_loss_clip": 0.01225619,
      "auxiliary_loss_mlp": 0.01023324,
      "balance_loss_clip": 1.04639959,
      "balance_loss_mlp": 1.01612377,
      "epoch": 0.6537605964047376,
      "flos": 21579476474880.0,
      "grad_norm": 1.851989820708844,
      "language_loss": 0.73142278,
      "learning_rate": 1.1310295795070253e-06,
      "loss": 0.75391221,
      "num_input_tokens_seen": 117017620,
      "step": 5437,
      "time_per_iteration": 2.5850019454956055
    },
    {
      "auxiliary_loss_clip": 0.01281326,
      "auxiliary_loss_mlp": 0.01022766,
      "balance_loss_clip": 1.04324019,
      "balance_loss_mlp": 1.01566982,
      "epoch": 0.6538808392953767,
      "flos": 26833997433600.0,
      "grad_norm": 1.7869915383695245,
      "language_loss": 0.81111264,
      "learning_rate": 1.1303280420569982e-06,
      "loss": 0.83415353,
      "num_input_tokens_seen": 117039505,
      "step": 5438,
      "time_per_iteration": 2.7777106761932373
    },
    {
      "auxiliary_loss_clip": 0.01225659,
      "auxiliary_loss_mlp": 0.01023986,
      "balance_loss_clip": 1.04874146,
      "balance_loss_mlp": 1.01674366,
      "epoch": 0.6540010821860157,
      "flos": 30738241301760.0,
      "grad_norm": 1.885830325799197,
      "language_loss": 0.77422762,
      "learning_rate": 1.1296266365355158e-06,
      "loss": 0.79672402,
      "num_input_tokens_seen": 117062890,
      "step": 5439,
      "time_per_iteration": 3.52317214012146
    },
    {
      "auxiliary_loss_clip": 0.01330105,
      "auxiliary_loss_mlp": 0.01030467,
      "balance_loss_clip": 1.0482465,
      "balance_loss_mlp": 1.02273607,
      "epoch": 0.6541213250766549,
      "flos": 26907147480960.0,
      "grad_norm": 2.0073581796269155,
      "language_loss": 0.73465717,
      "learning_rate": 1.1289253630489806e-06,
      "loss": 0.75826287,
      "num_input_tokens_seen": 117083940,
      "step": 5440,
      "time_per_iteration": 3.6282200813293457
    },
    {
      "auxiliary_loss_clip": 0.01233877,
      "auxiliary_loss_mlp": 0.01032998,
      "balance_loss_clip": 1.04880667,
      "balance_loss_mlp": 1.0247848,
      "epoch": 0.6542415679672939,
      "flos": 19172384409600.0,
      "grad_norm": 10.493789351798704,
      "language_loss": 0.72855526,
      "learning_rate": 1.1282242217037753e-06,
      "loss": 0.75122404,
      "num_input_tokens_seen": 117101440,
      "step": 5441,
      "time_per_iteration": 2.6482295989990234
    },
    {
      "auxiliary_loss_clip": 0.01374916,
      "auxiliary_loss_mlp": 0.01030344,
      "balance_loss_clip": 1.04128003,
      "balance_loss_mlp": 1.02286947,
      "epoch": 0.654361810857933,
      "flos": 48173517100800.0,
      "grad_norm": 1.7070326702621832,
      "language_loss": 0.62038243,
      "learning_rate": 1.127523212606262e-06,
      "loss": 0.64443505,
      "num_input_tokens_seen": 117124265,
      "step": 5442,
      "time_per_iteration": 3.8822784423828125
    },
    {
      "auxiliary_loss_clip": 0.01223403,
      "auxiliary_loss_mlp": 0.01029346,
      "balance_loss_clip": 1.04732311,
      "balance_loss_mlp": 1.02217007,
      "epoch": 0.6544820537485722,
      "flos": 26943165843840.0,
      "grad_norm": 2.2415187869201776,
      "language_loss": 0.73452568,
      "learning_rate": 1.1268223358627835e-06,
      "loss": 0.75705314,
      "num_input_tokens_seen": 117146755,
      "step": 5443,
      "time_per_iteration": 2.6701555252075195
    },
    {
      "auxiliary_loss_clip": 0.01177097,
      "auxiliary_loss_mlp": 0.01023173,
      "balance_loss_clip": 1.04918349,
      "balance_loss_mlp": 1.01571047,
      "epoch": 0.6546022966392112,
      "flos": 20886328748160.0,
      "grad_norm": 2.043749796469184,
      "language_loss": 0.7232517,
      "learning_rate": 1.126121591579663e-06,
      "loss": 0.7452544,
      "num_input_tokens_seen": 117165960,
      "step": 5444,
      "time_per_iteration": 2.587313652038574
    },
    {
      "auxiliary_loss_clip": 0.01224303,
      "auxiliary_loss_mlp": 0.01028213,
      "balance_loss_clip": 1.05015612,
      "balance_loss_mlp": 1.02135575,
      "epoch": 0.6547225395298503,
      "flos": 24936693143040.0,
      "grad_norm": 1.8017052151299981,
      "language_loss": 0.6930272,
      "learning_rate": 1.1254209798632018e-06,
      "loss": 0.71555233,
      "num_input_tokens_seen": 117186980,
      "step": 5445,
      "time_per_iteration": 2.7308852672576904
    },
    {
      "auxiliary_loss_clip": 0.01417698,
      "auxiliary_loss_mlp": 0.0102673,
      "balance_loss_clip": 1.04107296,
      "balance_loss_mlp": 1.01925516,
      "epoch": 0.6548427824204894,
      "flos": 22565942663040.0,
      "grad_norm": 2.0190456340500558,
      "language_loss": 0.84617686,
      "learning_rate": 1.124720500819683e-06,
      "loss": 0.87062114,
      "num_input_tokens_seen": 117205135,
      "step": 5446,
      "time_per_iteration": 2.887286901473999
    },
    {
      "auxiliary_loss_clip": 0.01180559,
      "auxiliary_loss_mlp": 0.01028793,
      "balance_loss_clip": 1.05230165,
      "balance_loss_mlp": 1.02150881,
      "epoch": 0.6549630253111285,
      "flos": 18442500048000.0,
      "grad_norm": 1.8159591016737833,
      "language_loss": 0.82501167,
      "learning_rate": 1.1240201545553682e-06,
      "loss": 0.84710521,
      "num_input_tokens_seen": 117222935,
      "step": 5447,
      "time_per_iteration": 3.6231441497802734
    },
    {
      "auxiliary_loss_clip": 0.01325051,
      "auxiliary_loss_mlp": 0.01032113,
      "balance_loss_clip": 1.04756856,
      "balance_loss_mlp": 1.02486467,
      "epoch": 0.6550832682017675,
      "flos": 25187313312000.0,
      "grad_norm": 1.7698258579747173,
      "language_loss": 0.7321533,
      "learning_rate": 1.1233199411764987e-06,
      "loss": 0.75572491,
      "num_input_tokens_seen": 117242370,
      "step": 5448,
      "time_per_iteration": 2.684396505355835
    },
    {
      "auxiliary_loss_clip": 0.01318192,
      "auxiliary_loss_mlp": 0.01023339,
      "balance_loss_clip": 1.04289639,
      "balance_loss_mlp": 1.0164963,
      "epoch": 0.6552035110924067,
      "flos": 22748153379840.0,
      "grad_norm": 2.2471085273305813,
      "language_loss": 0.68732458,
      "learning_rate": 1.1226198607892978e-06,
      "loss": 0.71073991,
      "num_input_tokens_seen": 117262930,
      "step": 5449,
      "time_per_iteration": 2.666257619857788
    },
    {
      "auxiliary_loss_clip": 0.01374545,
      "auxiliary_loss_mlp": 0.01023908,
      "balance_loss_clip": 1.04552627,
      "balance_loss_mlp": 1.01763463,
      "epoch": 0.6553237539830458,
      "flos": 21799178012160.0,
      "grad_norm": 1.8457519126083552,
      "language_loss": 0.80269736,
      "learning_rate": 1.1219199134999664e-06,
      "loss": 0.82668191,
      "num_input_tokens_seen": 117281430,
      "step": 5450,
      "time_per_iteration": 2.742067337036133
    },
    {
      "auxiliary_loss_clip": 0.01180523,
      "auxiliary_loss_mlp": 0.0103023,
      "balance_loss_clip": 1.0485214,
      "balance_loss_mlp": 1.02252293,
      "epoch": 0.6554439968736848,
      "flos": 20887226588160.0,
      "grad_norm": 5.357322027637243,
      "language_loss": 0.78694856,
      "learning_rate": 1.1212200994146863e-06,
      "loss": 0.80905604,
      "num_input_tokens_seen": 117299185,
      "step": 5451,
      "time_per_iteration": 2.6347031593322754
    },
    {
      "auxiliary_loss_clip": 0.0132486,
      "auxiliary_loss_mlp": 0.01026742,
      "balance_loss_clip": 1.0401001,
      "balance_loss_mlp": 1.01977706,
      "epoch": 0.655564239764324,
      "flos": 16139045698560.0,
      "grad_norm": 1.7916631768620825,
      "language_loss": 0.75622189,
      "learning_rate": 1.120520418639618e-06,
      "loss": 0.77973795,
      "num_input_tokens_seen": 117317720,
      "step": 5452,
      "time_per_iteration": 2.6369264125823975
    },
    {
      "auxiliary_loss_clip": 0.01227682,
      "auxiliary_loss_mlp": 0.01025483,
      "balance_loss_clip": 1.05075884,
      "balance_loss_mlp": 1.01843476,
      "epoch": 0.655684482654963,
      "flos": 29570354496000.0,
      "grad_norm": 2.365254205156692,
      "language_loss": 0.8375107,
      "learning_rate": 1.119820871280903e-06,
      "loss": 0.86004245,
      "num_input_tokens_seen": 117338795,
      "step": 5453,
      "time_per_iteration": 2.6866512298583984
    },
    {
      "auxiliary_loss_clip": 0.01226729,
      "auxiliary_loss_mlp": 0.01027573,
      "balance_loss_clip": 1.04771328,
      "balance_loss_mlp": 1.02041721,
      "epoch": 0.6558047255456021,
      "flos": 29789409588480.0,
      "grad_norm": 2.7070796565877036,
      "language_loss": 0.73645258,
      "learning_rate": 1.1191214574446614e-06,
      "loss": 0.75899559,
      "num_input_tokens_seen": 117359040,
      "step": 5454,
      "time_per_iteration": 2.633678674697876
    },
    {
      "auxiliary_loss_clip": 0.01270436,
      "auxiliary_loss_mlp": 0.01026973,
      "balance_loss_clip": 1.04465902,
      "balance_loss_mlp": 1.01957035,
      "epoch": 0.6559249684362413,
      "flos": 29059166090880.0,
      "grad_norm": 2.774114463800792,
      "language_loss": 0.80145884,
      "learning_rate": 1.118422177236995e-06,
      "loss": 0.82443297,
      "num_input_tokens_seen": 117380865,
      "step": 5455,
      "time_per_iteration": 2.719125270843506
    },
    {
      "auxiliary_loss_clip": 0.01276532,
      "auxiliary_loss_mlp": 0.01026127,
      "balance_loss_clip": 1.04644585,
      "balance_loss_mlp": 1.01869476,
      "epoch": 0.6560452113268803,
      "flos": 20225464369920.0,
      "grad_norm": 26.609557876332538,
      "language_loss": 0.85800976,
      "learning_rate": 1.1177230307639835e-06,
      "loss": 0.8810364,
      "num_input_tokens_seen": 117398405,
      "step": 5456,
      "time_per_iteration": 2.6332247257232666
    },
    {
      "auxiliary_loss_clip": 0.01321591,
      "auxiliary_loss_mlp": 0.01033101,
      "balance_loss_clip": 1.04380322,
      "balance_loss_mlp": 1.0264256,
      "epoch": 0.6561654542175194,
      "flos": 25045538330880.0,
      "grad_norm": 1.6583573679959318,
      "language_loss": 0.78998101,
      "learning_rate": 1.1170240181316865e-06,
      "loss": 0.81352794,
      "num_input_tokens_seen": 117419850,
      "step": 5457,
      "time_per_iteration": 2.744224786758423
    },
    {
      "auxiliary_loss_clip": 0.01322232,
      "auxiliary_loss_mlp": 0.01026955,
      "balance_loss_clip": 1.04162991,
      "balance_loss_mlp": 1.01956964,
      "epoch": 0.6562856971081584,
      "flos": 22856711258880.0,
      "grad_norm": 2.1062334415376616,
      "language_loss": 0.78848195,
      "learning_rate": 1.1163251394461442e-06,
      "loss": 0.81197387,
      "num_input_tokens_seen": 117438330,
      "step": 5458,
      "time_per_iteration": 2.6836535930633545
    },
    {
      "auxiliary_loss_clip": 0.01224048,
      "auxiliary_loss_mlp": 0.01028055,
      "balance_loss_clip": 1.04674816,
      "balance_loss_mlp": 1.02003241,
      "epoch": 0.6564059399987976,
      "flos": 18872565586560.0,
      "grad_norm": 1.9093226040754003,
      "language_loss": 0.82462692,
      "learning_rate": 1.1156263948133746e-06,
      "loss": 0.84714794,
      "num_input_tokens_seen": 117454985,
      "step": 5459,
      "time_per_iteration": 2.6298000812530518
    },
    {
      "auxiliary_loss_clip": 0.01367695,
      "auxiliary_loss_mlp": 0.02566053,
      "balance_loss_clip": 1.04330742,
      "balance_loss_mlp": 0.99995399,
      "epoch": 0.6565261828894366,
      "flos": 25484187219840.0,
      "grad_norm": 1.8022589529770474,
      "language_loss": 0.77475584,
      "learning_rate": 1.1149277843393787e-06,
      "loss": 0.81409329,
      "num_input_tokens_seen": 117476145,
      "step": 5460,
      "time_per_iteration": 2.7474470138549805
    },
    {
      "auxiliary_loss_clip": 0.01326672,
      "auxiliary_loss_mlp": 0.02565902,
      "balance_loss_clip": 1.03876173,
      "balance_loss_mlp": 0.99999362,
      "epoch": 0.6566464257800757,
      "flos": 19683500987520.0,
      "grad_norm": 2.576168140481698,
      "language_loss": 0.63363242,
      "learning_rate": 1.1142293081301342e-06,
      "loss": 0.67255813,
      "num_input_tokens_seen": 117494025,
      "step": 5461,
      "time_per_iteration": 2.8075101375579834
    },
    {
      "auxiliary_loss_clip": 0.0127173,
      "auxiliary_loss_mlp": 0.01023152,
      "balance_loss_clip": 1.045928,
      "balance_loss_mlp": 1.01616311,
      "epoch": 0.6567666686707149,
      "flos": 23514127931520.0,
      "grad_norm": 1.9498911895038764,
      "language_loss": 0.6799826,
      "learning_rate": 1.1135309662915995e-06,
      "loss": 0.7029314,
      "num_input_tokens_seen": 117514190,
      "step": 5462,
      "time_per_iteration": 2.6712820529937744
    },
    {
      "auxiliary_loss_clip": 0.01276379,
      "auxiliary_loss_mlp": 0.0102691,
      "balance_loss_clip": 1.04165447,
      "balance_loss_mlp": 1.01971591,
      "epoch": 0.6568869115613539,
      "flos": 32781342896640.0,
      "grad_norm": 2.986990336406726,
      "language_loss": 0.60419726,
      "learning_rate": 1.112832758929712e-06,
      "loss": 0.62723017,
      "num_input_tokens_seen": 117536800,
      "step": 5463,
      "time_per_iteration": 2.812462568283081
    },
    {
      "auxiliary_loss_clip": 0.01224842,
      "auxiliary_loss_mlp": 0.01030544,
      "balance_loss_clip": 1.04855466,
      "balance_loss_mlp": 1.02333212,
      "epoch": 0.657007154451993,
      "flos": 18442428220800.0,
      "grad_norm": 1.9068719320345653,
      "language_loss": 0.74978077,
      "learning_rate": 1.11213468615039e-06,
      "loss": 0.77233469,
      "num_input_tokens_seen": 117556230,
      "step": 5464,
      "time_per_iteration": 2.6571929454803467
    },
    {
      "auxiliary_loss_clip": 0.01322208,
      "auxiliary_loss_mlp": 0.01026333,
      "balance_loss_clip": 1.04054785,
      "balance_loss_mlp": 1.01951385,
      "epoch": 0.6571273973426321,
      "flos": 25156717902720.0,
      "grad_norm": 2.1302425531170845,
      "language_loss": 0.75519741,
      "learning_rate": 1.1114367480595292e-06,
      "loss": 0.77868283,
      "num_input_tokens_seen": 117577310,
      "step": 5465,
      "time_per_iteration": 3.842916250228882
    },
    {
      "auxiliary_loss_clip": 0.01426748,
      "auxiliary_loss_mlp": 0.01028197,
      "balance_loss_clip": 1.04470587,
      "balance_loss_mlp": 1.02020454,
      "epoch": 0.6572476402332712,
      "flos": 17529830352000.0,
      "grad_norm": 2.91766769542174,
      "language_loss": 0.81288111,
      "learning_rate": 1.1107389447630086e-06,
      "loss": 0.8374306,
      "num_input_tokens_seen": 117596010,
      "step": 5466,
      "time_per_iteration": 2.8885416984558105
    },
    {
      "auxiliary_loss_clip": 0.01276524,
      "auxiliary_loss_mlp": 0.02562854,
      "balance_loss_clip": 1.04521227,
      "balance_loss_mlp": 1.00003326,
      "epoch": 0.6573678831239103,
      "flos": 17014260487680.0,
      "grad_norm": 2.601936796310565,
      "language_loss": 0.78409708,
      "learning_rate": 1.1100412763666818e-06,
      "loss": 0.82249081,
      "num_input_tokens_seen": 117611270,
      "step": 5467,
      "time_per_iteration": 2.6148431301116943
    },
    {
      "auxiliary_loss_clip": 0.01278702,
      "auxiliary_loss_mlp": 0.01025205,
      "balance_loss_clip": 1.0487386,
      "balance_loss_mlp": 1.01793957,
      "epoch": 0.6574881260145494,
      "flos": 23910078528000.0,
      "grad_norm": 1.4414680272125278,
      "language_loss": 0.8012858,
      "learning_rate": 1.1093437429763865e-06,
      "loss": 0.82432485,
      "num_input_tokens_seen": 117631535,
      "step": 5468,
      "time_per_iteration": 3.6090495586395264
    },
    {
      "auxiliary_loss_clip": 0.01223864,
      "auxiliary_loss_mlp": 0.01028132,
      "balance_loss_clip": 1.04882431,
      "balance_loss_mlp": 1.02141118,
      "epoch": 0.6576083689051885,
      "flos": 11218458504960.0,
      "grad_norm": 18.990648333320188,
      "language_loss": 0.73066866,
      "learning_rate": 1.1086463446979361e-06,
      "loss": 0.75318861,
      "num_input_tokens_seen": 117649885,
      "step": 5469,
      "time_per_iteration": 2.6246602535247803
    },
    {
      "auxiliary_loss_clip": 0.01232251,
      "auxiliary_loss_mlp": 0.01028751,
      "balance_loss_clip": 1.05377316,
      "balance_loss_mlp": 1.020818,
      "epoch": 0.6577286117958275,
      "flos": 22455553190400.0,
      "grad_norm": 6.855671173346874,
      "language_loss": 0.77268797,
      "learning_rate": 1.1079490816371277e-06,
      "loss": 0.79529798,
      "num_input_tokens_seen": 117669650,
      "step": 5470,
      "time_per_iteration": 2.589524507522583
    },
    {
      "auxiliary_loss_clip": 0.01227835,
      "auxiliary_loss_mlp": 0.02565632,
      "balance_loss_clip": 1.04872978,
      "balance_loss_mlp": 1.00001144,
      "epoch": 0.6578488546864667,
      "flos": 21872184405120.0,
      "grad_norm": 1.961221999880228,
      "language_loss": 0.74683893,
      "learning_rate": 1.1072519538997352e-06,
      "loss": 0.78477353,
      "num_input_tokens_seen": 117688790,
      "step": 5471,
      "time_per_iteration": 2.620088577270508
    },
    {
      "auxiliary_loss_clip": 0.01275959,
      "auxiliary_loss_mlp": 0.0102178,
      "balance_loss_clip": 1.04392087,
      "balance_loss_mlp": 1.01488721,
      "epoch": 0.6579690975771058,
      "flos": 23543753673600.0,
      "grad_norm": 1.9026035522808549,
      "language_loss": 0.82588035,
      "learning_rate": 1.1065549615915095e-06,
      "loss": 0.84885776,
      "num_input_tokens_seen": 117708620,
      "step": 5472,
      "time_per_iteration": 2.7519476413726807
    },
    {
      "auxiliary_loss_clip": 0.01228487,
      "auxiliary_loss_mlp": 0.01026018,
      "balance_loss_clip": 1.0528034,
      "balance_loss_mlp": 1.01852584,
      "epoch": 0.6580893404677448,
      "flos": 32743995730560.0,
      "grad_norm": 2.6792391778299716,
      "language_loss": 0.78506422,
      "learning_rate": 1.105858104818187e-06,
      "loss": 0.80760932,
      "num_input_tokens_seen": 117729775,
      "step": 5473,
      "time_per_iteration": 3.55314302444458
    },
    {
      "auxiliary_loss_clip": 0.01232849,
      "auxiliary_loss_mlp": 0.01028539,
      "balance_loss_clip": 1.05039191,
      "balance_loss_mlp": 1.02083206,
      "epoch": 0.658209583358384,
      "flos": 15888138220800.0,
      "grad_norm": 2.6030154700700203,
      "language_loss": 0.75067854,
      "learning_rate": 1.105161383685478e-06,
      "loss": 0.77329242,
      "num_input_tokens_seen": 117746160,
      "step": 5474,
      "time_per_iteration": 2.6238620281219482
    },
    {
      "auxiliary_loss_clip": 0.01218791,
      "auxiliary_loss_mlp": 0.00999148,
      "balance_loss_clip": 1.00924635,
      "balance_loss_mlp": 0.9982416,
      "epoch": 0.658329826249023,
      "flos": 62695902447360.0,
      "grad_norm": 0.7256989419024298,
      "language_loss": 0.56357932,
      "learning_rate": 1.1044647982990771e-06,
      "loss": 0.58575869,
      "num_input_tokens_seen": 117808045,
      "step": 5475,
      "time_per_iteration": 3.1921029090881348
    },
    {
      "auxiliary_loss_clip": 0.01274808,
      "auxiliary_loss_mlp": 0.01027845,
      "balance_loss_clip": 1.04844105,
      "balance_loss_mlp": 1.02065921,
      "epoch": 0.6584500691396621,
      "flos": 31722624501120.0,
      "grad_norm": 2.4250154459510114,
      "language_loss": 0.64347148,
      "learning_rate": 1.1037683487646536e-06,
      "loss": 0.66649806,
      "num_input_tokens_seen": 117828330,
      "step": 5476,
      "time_per_iteration": 2.759403705596924
    },
    {
      "auxiliary_loss_clip": 0.01273766,
      "auxiliary_loss_mlp": 0.02564117,
      "balance_loss_clip": 1.04998446,
      "balance_loss_mlp": 0.99995828,
      "epoch": 0.6585703120303013,
      "flos": 18406086635520.0,
      "grad_norm": 2.1250114595909015,
      "language_loss": 0.77019048,
      "learning_rate": 1.1030720351878583e-06,
      "loss": 0.80856931,
      "num_input_tokens_seen": 117846450,
      "step": 5477,
      "time_per_iteration": 2.625041961669922
    },
    {
      "auxiliary_loss_clip": 0.01173574,
      "auxiliary_loss_mlp": 0.01001554,
      "balance_loss_clip": 1.00968719,
      "balance_loss_mlp": 1.00053501,
      "epoch": 0.6586905549209403,
      "flos": 58309880434560.0,
      "grad_norm": 0.8066425328485106,
      "language_loss": 0.57590562,
      "learning_rate": 1.102375857674323e-06,
      "loss": 0.59765691,
      "num_input_tokens_seen": 117908365,
      "step": 5478,
      "time_per_iteration": 3.2424581050872803
    },
    {
      "auxiliary_loss_clip": 0.01276464,
      "auxiliary_loss_mlp": 0.01027496,
      "balance_loss_clip": 1.04453194,
      "balance_loss_mlp": 1.02042687,
      "epoch": 0.6588107978115794,
      "flos": 22782627457920.0,
      "grad_norm": 1.770267910107722,
      "language_loss": 0.89952385,
      "learning_rate": 1.1016798163296561e-06,
      "loss": 0.92256343,
      "num_input_tokens_seen": 117927565,
      "step": 5479,
      "time_per_iteration": 2.643224000930786
    },
    {
      "auxiliary_loss_clip": 0.01129812,
      "auxiliary_loss_mlp": 0.01031,
      "balance_loss_clip": 1.05000925,
      "balance_loss_mlp": 1.02401137,
      "epoch": 0.6589310407022185,
      "flos": 20667525050880.0,
      "grad_norm": 5.388688971263541,
      "language_loss": 0.66506004,
      "learning_rate": 1.1009839112594471e-06,
      "loss": 0.68666822,
      "num_input_tokens_seen": 117945590,
      "step": 5480,
      "time_per_iteration": 2.630459785461426
    },
    {
      "auxiliary_loss_clip": 0.01229078,
      "auxiliary_loss_mlp": 0.01023776,
      "balance_loss_clip": 1.0497328,
      "balance_loss_mlp": 1.01674581,
      "epoch": 0.6590512835928576,
      "flos": 25630595055360.0,
      "grad_norm": 2.296207521511213,
      "language_loss": 0.72355533,
      "learning_rate": 1.1002881425692638e-06,
      "loss": 0.74608386,
      "num_input_tokens_seen": 117966020,
      "step": 5481,
      "time_per_iteration": 2.6454477310180664
    },
    {
      "auxiliary_loss_clip": 0.0122495,
      "auxiliary_loss_mlp": 0.01028061,
      "balance_loss_clip": 1.04627693,
      "balance_loss_mlp": 1.02073002,
      "epoch": 0.6591715264834966,
      "flos": 23726108044800.0,
      "grad_norm": 1.7255171801342932,
      "language_loss": 0.7519381,
      "learning_rate": 1.0995925103646532e-06,
      "loss": 0.7744683,
      "num_input_tokens_seen": 117984620,
      "step": 5482,
      "time_per_iteration": 2.635377883911133
    },
    {
      "auxiliary_loss_clip": 0.01320378,
      "auxiliary_loss_mlp": 0.01026638,
      "balance_loss_clip": 1.04604435,
      "balance_loss_mlp": 1.01972997,
      "epoch": 0.6592917693741358,
      "flos": 35773850822400.0,
      "grad_norm": 1.488041581069915,
      "language_loss": 0.67109591,
      "learning_rate": 1.0988970147511437e-06,
      "loss": 0.69456607,
      "num_input_tokens_seen": 118006500,
      "step": 5483,
      "time_per_iteration": 2.8348073959350586
    },
    {
      "auxiliary_loss_clip": 0.01274273,
      "auxiliary_loss_mlp": 0.01023971,
      "balance_loss_clip": 1.0486474,
      "balance_loss_mlp": 1.01665187,
      "epoch": 0.6594120122647749,
      "flos": 21396834794880.0,
      "grad_norm": 2.7070497665171604,
      "language_loss": 0.80232364,
      "learning_rate": 1.0982016558342405e-06,
      "loss": 0.82530606,
      "num_input_tokens_seen": 118025470,
      "step": 5484,
      "time_per_iteration": 2.6559014320373535
    },
    {
      "auxiliary_loss_clip": 0.01179223,
      "auxiliary_loss_mlp": 0.01027246,
      "balance_loss_clip": 1.0526526,
      "balance_loss_mlp": 1.02045727,
      "epoch": 0.6595322551554139,
      "flos": 19351829779200.0,
      "grad_norm": 7.648347431674363,
      "language_loss": 0.71658325,
      "learning_rate": 1.0975064337194291e-06,
      "loss": 0.73864794,
      "num_input_tokens_seen": 118043515,
      "step": 5485,
      "time_per_iteration": 2.6056840419769287
    },
    {
      "auxiliary_loss_clip": 0.01329776,
      "auxiliary_loss_mlp": 0.01027732,
      "balance_loss_clip": 1.04834938,
      "balance_loss_mlp": 1.02073455,
      "epoch": 0.6596524980460531,
      "flos": 16837113588480.0,
      "grad_norm": 2.185468450648023,
      "language_loss": 0.7033155,
      "learning_rate": 1.0968113485121743e-06,
      "loss": 0.72689056,
      "num_input_tokens_seen": 118063105,
      "step": 5486,
      "time_per_iteration": 2.7169063091278076
    },
    {
      "auxiliary_loss_clip": 0.01229027,
      "auxiliary_loss_mlp": 0.02567182,
      "balance_loss_clip": 1.04760838,
      "balance_loss_mlp": 0.99997717,
      "epoch": 0.6597727409366921,
      "flos": 21798567480960.0,
      "grad_norm": 2.7532837228679226,
      "language_loss": 0.80023277,
      "learning_rate": 1.0961164003179185e-06,
      "loss": 0.83819485,
      "num_input_tokens_seen": 118081615,
      "step": 5487,
      "time_per_iteration": 2.6531145572662354
    },
    {
      "auxiliary_loss_clip": 0.01328394,
      "auxiliary_loss_mlp": 0.01024856,
      "balance_loss_clip": 1.04504502,
      "balance_loss_mlp": 1.01790035,
      "epoch": 0.6598929838273312,
      "flos": 23730704985600.0,
      "grad_norm": 2.357185210308578,
      "language_loss": 0.84051317,
      "learning_rate": 1.0954215892420884e-06,
      "loss": 0.86404562,
      "num_input_tokens_seen": 118102315,
      "step": 5488,
      "time_per_iteration": 2.698723793029785
    },
    {
      "auxiliary_loss_clip": 0.01329301,
      "auxiliary_loss_mlp": 0.01024688,
      "balance_loss_clip": 1.04846406,
      "balance_loss_mlp": 1.01740456,
      "epoch": 0.6600132267179702,
      "flos": 19974520978560.0,
      "grad_norm": 1.6778929052993212,
      "language_loss": 0.70393747,
      "learning_rate": 1.094726915390082e-06,
      "loss": 0.72747737,
      "num_input_tokens_seen": 118120650,
      "step": 5489,
      "time_per_iteration": 2.712451934814453
    },
    {
      "auxiliary_loss_clip": 0.01229477,
      "auxiliary_loss_mlp": 0.01021589,
      "balance_loss_clip": 1.05102742,
      "balance_loss_mlp": 1.01424527,
      "epoch": 0.6601334696086094,
      "flos": 22342649765760.0,
      "grad_norm": 2.135844974541871,
      "language_loss": 0.69788098,
      "learning_rate": 1.0940323788672836e-06,
      "loss": 0.72039163,
      "num_input_tokens_seen": 118139825,
      "step": 5490,
      "time_per_iteration": 2.5877645015716553
    },
    {
      "auxiliary_loss_clip": 0.01222006,
      "auxiliary_loss_mlp": 0.01025963,
      "balance_loss_clip": 1.04878855,
      "balance_loss_mlp": 1.01886737,
      "epoch": 0.6602537124992485,
      "flos": 25703098657920.0,
      "grad_norm": 1.9826925355280898,
      "language_loss": 0.74273932,
      "learning_rate": 1.0933379797790522e-06,
      "loss": 0.76521903,
      "num_input_tokens_seen": 118159240,
      "step": 5491,
      "time_per_iteration": 3.5598385334014893
    },
    {
      "auxiliary_loss_clip": 0.01179774,
      "auxiliary_loss_mlp": 0.01029124,
      "balance_loss_clip": 1.05208576,
      "balance_loss_mlp": 1.02213883,
      "epoch": 0.6603739553898875,
      "flos": 25848572739840.0,
      "grad_norm": 2.6703478945513166,
      "language_loss": 0.7169143,
      "learning_rate": 1.0926437182307293e-06,
      "loss": 0.7390033,
      "num_input_tokens_seen": 118178050,
      "step": 5492,
      "time_per_iteration": 3.4581408500671387
    },
    {
      "auxiliary_loss_clip": 0.01281549,
      "auxiliary_loss_mlp": 0.01023135,
      "balance_loss_clip": 1.04670238,
      "balance_loss_mlp": 1.01582122,
      "epoch": 0.6604941982805267,
      "flos": 24570296461440.0,
      "grad_norm": 2.8959503496679098,
      "language_loss": 0.78372419,
      "learning_rate": 1.0919495943276338e-06,
      "loss": 0.80677098,
      "num_input_tokens_seen": 118199070,
      "step": 5493,
      "time_per_iteration": 2.670224666595459
    },
    {
      "auxiliary_loss_clip": 0.01333953,
      "auxiliary_loss_mlp": 0.0102414,
      "balance_loss_clip": 1.04446101,
      "balance_loss_mlp": 1.01663613,
      "epoch": 0.6606144411711657,
      "flos": 13261775581440.0,
      "grad_norm": 2.2995361233913547,
      "language_loss": 0.76168799,
      "learning_rate": 1.0912556081750611e-06,
      "loss": 0.78526896,
      "num_input_tokens_seen": 118217000,
      "step": 5494,
      "time_per_iteration": 3.584895610809326
    },
    {
      "auxiliary_loss_clip": 0.01272862,
      "auxiliary_loss_mlp": 0.01023395,
      "balance_loss_clip": 1.05008221,
      "balance_loss_mlp": 1.01638544,
      "epoch": 0.6607346840618048,
      "flos": 25155281358720.0,
      "grad_norm": 1.9931413128580504,
      "language_loss": 0.76469827,
      "learning_rate": 1.0905617598782909e-06,
      "loss": 0.78766078,
      "num_input_tokens_seen": 118237205,
      "step": 5495,
      "time_per_iteration": 2.673684597015381
    },
    {
      "auxiliary_loss_clip": 0.01370342,
      "auxiliary_loss_mlp": 0.01025245,
      "balance_loss_clip": 1.04328668,
      "balance_loss_mlp": 1.01815832,
      "epoch": 0.660854926952444,
      "flos": 17638029095040.0,
      "grad_norm": 2.3076713966539475,
      "language_loss": 0.81521726,
      "learning_rate": 1.0898680495425775e-06,
      "loss": 0.83917308,
      "num_input_tokens_seen": 118255495,
      "step": 5496,
      "time_per_iteration": 2.740471839904785
    },
    {
      "auxiliary_loss_clip": 0.012788,
      "auxiliary_loss_mlp": 0.01029848,
      "balance_loss_clip": 1.04889059,
      "balance_loss_mlp": 1.02271962,
      "epoch": 0.660975169843083,
      "flos": 16836000266880.0,
      "grad_norm": 1.9092998098466674,
      "language_loss": 0.80177587,
      "learning_rate": 1.0891744772731594e-06,
      "loss": 0.82486242,
      "num_input_tokens_seen": 118273310,
      "step": 5497,
      "time_per_iteration": 2.6357572078704834
    },
    {
      "auxiliary_loss_clip": 0.01227825,
      "auxiliary_loss_mlp": 0.01024599,
      "balance_loss_clip": 1.04784918,
      "balance_loss_mlp": 1.0174973,
      "epoch": 0.6610954127337221,
      "flos": 26870410846080.0,
      "grad_norm": 1.8444955740469628,
      "language_loss": 0.66068739,
      "learning_rate": 1.088481043175248e-06,
      "loss": 0.68321162,
      "num_input_tokens_seen": 118293880,
      "step": 5498,
      "time_per_iteration": 2.638779640197754
    },
    {
      "auxiliary_loss_clip": 0.01270777,
      "auxiliary_loss_mlp": 0.01024819,
      "balance_loss_clip": 1.04412353,
      "balance_loss_mlp": 1.01800358,
      "epoch": 0.6612156556243612,
      "flos": 26465697331200.0,
      "grad_norm": 1.6515689148666428,
      "language_loss": 0.75555182,
      "learning_rate": 1.0877877473540368e-06,
      "loss": 0.77850771,
      "num_input_tokens_seen": 118314465,
      "step": 5499,
      "time_per_iteration": 3.691736936569214
    },
    {
      "auxiliary_loss_clip": 0.01179351,
      "auxiliary_loss_mlp": 0.01029363,
      "balance_loss_clip": 1.05028105,
      "balance_loss_mlp": 1.02136374,
      "epoch": 0.6613358985150003,
      "flos": 19791915212160.0,
      "grad_norm": 2.255625154525312,
      "language_loss": 0.72829193,
      "learning_rate": 1.0870945899147002e-06,
      "loss": 0.75037909,
      "num_input_tokens_seen": 118331110,
      "step": 5500,
      "time_per_iteration": 2.580784797668457
    },
    {
      "auxiliary_loss_clip": 0.01225428,
      "auxiliary_loss_mlp": 0.01024548,
      "balance_loss_clip": 1.0504142,
      "balance_loss_mlp": 1.01780689,
      "epoch": 0.6614561414056394,
      "flos": 26831627136000.0,
      "grad_norm": 2.2863270141282968,
      "language_loss": 0.76088089,
      "learning_rate": 1.0864015709623879e-06,
      "loss": 0.78338063,
      "num_input_tokens_seen": 118351980,
      "step": 5501,
      "time_per_iteration": 2.7074501514434814
    },
    {
      "auxiliary_loss_clip": 0.0123166,
      "auxiliary_loss_mlp": 0.01026342,
      "balance_loss_clip": 1.04933703,
      "balance_loss_mlp": 1.01963377,
      "epoch": 0.6615763842962785,
      "flos": 22894597128960.0,
      "grad_norm": 2.8381154113330678,
      "language_loss": 0.80326664,
      "learning_rate": 1.0857086906022313e-06,
      "loss": 0.82584667,
      "num_input_tokens_seen": 118370315,
      "step": 5502,
      "time_per_iteration": 2.5841357707977295
    },
    {
      "auxiliary_loss_clip": 0.01417241,
      "auxiliary_loss_mlp": 0.01029806,
      "balance_loss_clip": 1.04485607,
      "balance_loss_mlp": 1.02254653,
      "epoch": 0.6616966271869176,
      "flos": 24790321221120.0,
      "grad_norm": 2.347379000619189,
      "language_loss": 0.72956133,
      "learning_rate": 1.0850159489393388e-06,
      "loss": 0.75403178,
      "num_input_tokens_seen": 118389575,
      "step": 5503,
      "time_per_iteration": 2.7797205448150635
    },
    {
      "auxiliary_loss_clip": 0.01320392,
      "auxiliary_loss_mlp": 0.01027647,
      "balance_loss_clip": 1.04032254,
      "balance_loss_mlp": 1.02050626,
      "epoch": 0.6618168700775566,
      "flos": 17202109639680.0,
      "grad_norm": 2.006315264312654,
      "language_loss": 0.82087106,
      "learning_rate": 1.0843233460787992e-06,
      "loss": 0.84435141,
      "num_input_tokens_seen": 118406790,
      "step": 5504,
      "time_per_iteration": 2.6603031158447266
    },
    {
      "auxiliary_loss_clip": 0.01321734,
      "auxiliary_loss_mlp": 0.01027714,
      "balance_loss_clip": 1.04639351,
      "balance_loss_mlp": 1.02053118,
      "epoch": 0.6619371129681958,
      "flos": 25447091448960.0,
      "grad_norm": 2.4510862808091605,
      "language_loss": 0.78084755,
      "learning_rate": 1.0836308821256805e-06,
      "loss": 0.80434203,
      "num_input_tokens_seen": 118427590,
      "step": 5505,
      "time_per_iteration": 2.700047492980957
    },
    {
      "auxiliary_loss_clip": 0.01220502,
      "auxiliary_loss_mlp": 0.01026015,
      "balance_loss_clip": 1.04764557,
      "balance_loss_mlp": 1.01966715,
      "epoch": 0.6620573558588349,
      "flos": 18040444139520.0,
      "grad_norm": 2.0593991503372617,
      "language_loss": 0.78070229,
      "learning_rate": 1.0829385571850282e-06,
      "loss": 0.80316746,
      "num_input_tokens_seen": 118444570,
      "step": 5506,
      "time_per_iteration": 2.656554698944092
    },
    {
      "auxiliary_loss_clip": 0.0118317,
      "auxiliary_loss_mlp": 0.0103498,
      "balance_loss_clip": 1.05109811,
      "balance_loss_mlp": 1.0268681,
      "epoch": 0.6621775987494739,
      "flos": 17785586165760.0,
      "grad_norm": 2.612083129248404,
      "language_loss": 0.83797657,
      "learning_rate": 1.0822463713618679e-06,
      "loss": 0.86015803,
      "num_input_tokens_seen": 118461425,
      "step": 5507,
      "time_per_iteration": 2.5681161880493164
    },
    {
      "auxiliary_loss_clip": 0.01325236,
      "auxiliary_loss_mlp": 0.01024572,
      "balance_loss_clip": 1.04393506,
      "balance_loss_mlp": 1.01727319,
      "epoch": 0.6622978416401131,
      "flos": 17492590926720.0,
      "grad_norm": 2.49346246405378,
      "language_loss": 0.85181266,
      "learning_rate": 1.0815543247612034e-06,
      "loss": 0.87531078,
      "num_input_tokens_seen": 118478495,
      "step": 5508,
      "time_per_iteration": 2.688002109527588
    },
    {
      "auxiliary_loss_clip": 0.01275377,
      "auxiliary_loss_mlp": 0.01025928,
      "balance_loss_clip": 1.04253531,
      "balance_loss_mlp": 1.01900148,
      "epoch": 0.6624180845307521,
      "flos": 21648352803840.0,
      "grad_norm": 1.5044380476087564,
      "language_loss": 0.82883453,
      "learning_rate": 1.0808624174880168e-06,
      "loss": 0.85184753,
      "num_input_tokens_seen": 118499145,
      "step": 5509,
      "time_per_iteration": 2.6334853172302246
    },
    {
      "auxiliary_loss_clip": 0.01174651,
      "auxiliary_loss_mlp": 0.01024821,
      "balance_loss_clip": 1.050318,
      "balance_loss_mlp": 1.0181812,
      "epoch": 0.6625383274213912,
      "flos": 23805902108160.0,
      "grad_norm": 2.0399557502494265,
      "language_loss": 0.79930413,
      "learning_rate": 1.080170649647272e-06,
      "loss": 0.8212989,
      "num_input_tokens_seen": 118518950,
      "step": 5510,
      "time_per_iteration": 2.6194918155670166
    },
    {
      "auxiliary_loss_clip": 0.01176297,
      "auxiliary_loss_mlp": 0.01025117,
      "balance_loss_clip": 1.04893792,
      "balance_loss_mlp": 1.01804471,
      "epoch": 0.6626585703120303,
      "flos": 33262941473280.0,
      "grad_norm": 1.9981682594140222,
      "language_loss": 0.6725536,
      "learning_rate": 1.0794790213439068e-06,
      "loss": 0.69456774,
      "num_input_tokens_seen": 118545850,
      "step": 5511,
      "time_per_iteration": 2.726300001144409
    },
    {
      "auxiliary_loss_clip": 0.01376859,
      "auxiliary_loss_mlp": 0.01024984,
      "balance_loss_clip": 1.04499555,
      "balance_loss_mlp": 1.01737821,
      "epoch": 0.6627788132026694,
      "flos": 22085780630400.0,
      "grad_norm": 1.7662786169204687,
      "language_loss": 0.78302264,
      "learning_rate": 1.078787532682843e-06,
      "loss": 0.80704105,
      "num_input_tokens_seen": 118563325,
      "step": 5512,
      "time_per_iteration": 2.7515039443969727
    },
    {
      "auxiliary_loss_clip": 0.0122491,
      "auxiliary_loss_mlp": 0.01027203,
      "balance_loss_clip": 1.04914057,
      "balance_loss_mlp": 1.01969922,
      "epoch": 0.6628990560933085,
      "flos": 36173608260480.0,
      "grad_norm": 3.219044728636975,
      "language_loss": 0.75697464,
      "learning_rate": 1.0780961837689773e-06,
      "loss": 0.77949584,
      "num_input_tokens_seen": 118582835,
      "step": 5513,
      "time_per_iteration": 2.7394988536834717
    },
    {
      "auxiliary_loss_clip": 0.01272368,
      "auxiliary_loss_mlp": 0.01024523,
      "balance_loss_clip": 1.04762757,
      "balance_loss_mlp": 1.017555,
      "epoch": 0.6630192989839476,
      "flos": 18513567106560.0,
      "grad_norm": 2.2799869220126636,
      "language_loss": 0.70258355,
      "learning_rate": 1.0774049747071883e-06,
      "loss": 0.72555244,
      "num_input_tokens_seen": 118600715,
      "step": 5514,
      "time_per_iteration": 2.60937237739563
    },
    {
      "auxiliary_loss_clip": 0.01371696,
      "auxiliary_loss_mlp": 0.01027394,
      "balance_loss_clip": 1.04684401,
      "balance_loss_mlp": 1.0208286,
      "epoch": 0.6631395418745867,
      "flos": 35809510049280.0,
      "grad_norm": 1.9064255369633842,
      "language_loss": 0.68207538,
      "learning_rate": 1.076713905602332e-06,
      "loss": 0.70606625,
      "num_input_tokens_seen": 118621290,
      "step": 5515,
      "time_per_iteration": 2.8440396785736084
    },
    {
      "auxiliary_loss_clip": 0.01234325,
      "auxiliary_loss_mlp": 0.01028294,
      "balance_loss_clip": 1.0533545,
      "balance_loss_mlp": 1.02090907,
      "epoch": 0.6632597847652257,
      "flos": 20047742853120.0,
      "grad_norm": 3.0687568022116025,
      "language_loss": 0.81071752,
      "learning_rate": 1.07602297655924e-06,
      "loss": 0.83334374,
      "num_input_tokens_seen": 118639610,
      "step": 5516,
      "time_per_iteration": 2.6223411560058594
    },
    {
      "auxiliary_loss_clip": 0.01178618,
      "auxiliary_loss_mlp": 0.0102494,
      "balance_loss_clip": 1.05178285,
      "balance_loss_mlp": 1.01811564,
      "epoch": 0.6633800276558649,
      "flos": 21214480423680.0,
      "grad_norm": 2.023862389264744,
      "language_loss": 0.81168312,
      "learning_rate": 1.0753321876827292e-06,
      "loss": 0.83371872,
      "num_input_tokens_seen": 118658895,
      "step": 5517,
      "time_per_iteration": 2.549839973449707
    },
    {
      "auxiliary_loss_clip": 0.01178751,
      "auxiliary_loss_mlp": 0.01025636,
      "balance_loss_clip": 1.05024993,
      "balance_loss_mlp": 1.01843023,
      "epoch": 0.663500270546504,
      "flos": 23987753688960.0,
      "grad_norm": 2.1797808943871977,
      "language_loss": 0.74202359,
      "learning_rate": 1.0746415390775893e-06,
      "loss": 0.76406741,
      "num_input_tokens_seen": 118677025,
      "step": 5518,
      "time_per_iteration": 4.583606958389282
    },
    {
      "auxiliary_loss_clip": 0.01178786,
      "auxiliary_loss_mlp": 0.01027857,
      "balance_loss_clip": 1.05213583,
      "balance_loss_mlp": 1.0211755,
      "epoch": 0.663620513437143,
      "flos": 17932389050880.0,
      "grad_norm": 1.8883854072974364,
      "language_loss": 0.76686215,
      "learning_rate": 1.0739510308485939e-06,
      "loss": 0.78892863,
      "num_input_tokens_seen": 118694240,
      "step": 5519,
      "time_per_iteration": 2.5728931427001953
    },
    {
      "auxiliary_loss_clip": 0.01131292,
      "auxiliary_loss_mlp": 0.0099829,
      "balance_loss_clip": 1.01195109,
      "balance_loss_mlp": 0.99728882,
      "epoch": 0.6637407563277821,
      "flos": 57840241086720.0,
      "grad_norm": 0.8043689979672518,
      "language_loss": 0.62547088,
      "learning_rate": 1.07326066310049e-06,
      "loss": 0.64676666,
      "num_input_tokens_seen": 118758365,
      "step": 5520,
      "time_per_iteration": 4.192986249923706
    },
    {
      "auxiliary_loss_clip": 0.01324277,
      "auxiliary_loss_mlp": 0.01030192,
      "balance_loss_clip": 1.04570401,
      "balance_loss_mlp": 1.02253246,
      "epoch": 0.6638609992184212,
      "flos": 27306007079040.0,
      "grad_norm": 1.903197794320203,
      "language_loss": 0.79663008,
      "learning_rate": 1.0725704359380059e-06,
      "loss": 0.82017475,
      "num_input_tokens_seen": 118778220,
      "step": 5521,
      "time_per_iteration": 2.712439775466919
    },
    {
      "auxiliary_loss_clip": 0.0117782,
      "auxiliary_loss_mlp": 0.01027334,
      "balance_loss_clip": 1.05021977,
      "balance_loss_mlp": 1.02034211,
      "epoch": 0.6639812421090603,
      "flos": 18624854419200.0,
      "grad_norm": 2.2423425979476344,
      "language_loss": 0.72472632,
      "learning_rate": 1.0718803494658497e-06,
      "loss": 0.74677783,
      "num_input_tokens_seen": 118797110,
      "step": 5522,
      "time_per_iteration": 2.5610995292663574
    },
    {
      "auxiliary_loss_clip": 0.01422733,
      "auxiliary_loss_mlp": 0.01029435,
      "balance_loss_clip": 1.03904581,
      "balance_loss_mlp": 1.02210701,
      "epoch": 0.6641014849996993,
      "flos": 15924479806080.0,
      "grad_norm": 8.348746104356044,
      "language_loss": 0.83815414,
      "learning_rate": 1.071190403788707e-06,
      "loss": 0.86267585,
      "num_input_tokens_seen": 118812415,
      "step": 5523,
      "time_per_iteration": 2.932133197784424
    },
    {
      "auxiliary_loss_clip": 0.01331102,
      "auxiliary_loss_mlp": 0.01027954,
      "balance_loss_clip": 1.04895937,
      "balance_loss_mlp": 1.02069676,
      "epoch": 0.6642217278903385,
      "flos": 26505486622080.0,
      "grad_norm": 2.98664321500518,
      "language_loss": 0.75411922,
      "learning_rate": 1.0705005990112415e-06,
      "loss": 0.77770978,
      "num_input_tokens_seen": 118832195,
      "step": 5524,
      "time_per_iteration": 2.880600690841675
    },
    {
      "auxiliary_loss_clip": 0.01365062,
      "auxiliary_loss_mlp": 0.01029394,
      "balance_loss_clip": 1.04501748,
      "balance_loss_mlp": 1.02225053,
      "epoch": 0.6643419707809776,
      "flos": 15377308951680.0,
      "grad_norm": 2.334941119652741,
      "language_loss": 0.74703753,
      "learning_rate": 1.0698109352380957e-06,
      "loss": 0.77098209,
      "num_input_tokens_seen": 118849795,
      "step": 5525,
      "time_per_iteration": 3.6938283443450928
    },
    {
      "auxiliary_loss_clip": 0.01175247,
      "auxiliary_loss_mlp": 0.01028287,
      "balance_loss_clip": 1.04964948,
      "balance_loss_mlp": 1.02113771,
      "epoch": 0.6644622136716166,
      "flos": 25117610970240.0,
      "grad_norm": 2.591865875067558,
      "language_loss": 0.78055155,
      "learning_rate": 1.0691214125738909e-06,
      "loss": 0.80258685,
      "num_input_tokens_seen": 118870000,
      "step": 5526,
      "time_per_iteration": 2.6220219135284424
    },
    {
      "auxiliary_loss_clip": 0.01066642,
      "auxiliary_loss_mlp": 0.01003513,
      "balance_loss_clip": 1.01042819,
      "balance_loss_mlp": 1.00256503,
      "epoch": 0.6645824565622558,
      "flos": 66201717680640.0,
      "grad_norm": 0.8084229792955522,
      "language_loss": 0.57486117,
      "learning_rate": 1.0684320311232287e-06,
      "loss": 0.59556276,
      "num_input_tokens_seen": 118932905,
      "step": 5527,
      "time_per_iteration": 3.2511510848999023
    },
    {
      "auxiliary_loss_clip": 0.01272728,
      "auxiliary_loss_mlp": 0.01027439,
      "balance_loss_clip": 1.04571784,
      "balance_loss_mlp": 1.02007151,
      "epoch": 0.6647026994528948,
      "flos": 25082131311360.0,
      "grad_norm": 2.008985292055761,
      "language_loss": 0.81515557,
      "learning_rate": 1.0677427909906865e-06,
      "loss": 0.83815718,
      "num_input_tokens_seen": 118953355,
      "step": 5528,
      "time_per_iteration": 2.6179802417755127
    },
    {
      "auxiliary_loss_clip": 0.01183645,
      "auxiliary_loss_mlp": 0.01025845,
      "balance_loss_clip": 1.05388629,
      "balance_loss_mlp": 1.01849258,
      "epoch": 0.6648229423435339,
      "flos": 18222187979520.0,
      "grad_norm": 2.398812717152142,
      "language_loss": 0.72352827,
      "learning_rate": 1.0670536922808216e-06,
      "loss": 0.74562323,
      "num_input_tokens_seen": 118973480,
      "step": 5529,
      "time_per_iteration": 2.5919346809387207
    },
    {
      "auxiliary_loss_clip": 0.01276075,
      "auxiliary_loss_mlp": 0.01025917,
      "balance_loss_clip": 1.04831862,
      "balance_loss_mlp": 1.01962852,
      "epoch": 0.6649431852341731,
      "flos": 18296882311680.0,
      "grad_norm": 2.052258998668079,
      "language_loss": 0.72162092,
      "learning_rate": 1.06636473509817e-06,
      "loss": 0.74464083,
      "num_input_tokens_seen": 118989860,
      "step": 5530,
      "time_per_iteration": 2.6255650520324707
    },
    {
      "auxiliary_loss_clip": 0.01274494,
      "auxiliary_loss_mlp": 0.02565725,
      "balance_loss_clip": 1.0466038,
      "balance_loss_mlp": 0.99994272,
      "epoch": 0.6650634281248121,
      "flos": 17019575700480.0,
      "grad_norm": 2.113997359802001,
      "language_loss": 0.80841625,
      "learning_rate": 1.0656759195472447e-06,
      "loss": 0.84681839,
      "num_input_tokens_seen": 119007150,
      "step": 5531,
      "time_per_iteration": 2.6718270778656006
    },
    {
      "auxiliary_loss_clip": 0.01169459,
      "auxiliary_loss_mlp": 0.0099683,
      "balance_loss_clip": 1.0113368,
      "balance_loss_mlp": 0.99582225,
      "epoch": 0.6651836710154512,
      "flos": 69294810666240.0,
      "grad_norm": 0.7883594392411588,
      "language_loss": 0.59722459,
      "learning_rate": 1.0649872457325414e-06,
      "loss": 0.61888748,
      "num_input_tokens_seen": 119068435,
      "step": 5532,
      "time_per_iteration": 3.17917799949646
    },
    {
      "auxiliary_loss_clip": 0.01121179,
      "auxiliary_loss_mlp": 0.01003787,
      "balance_loss_clip": 1.0092597,
      "balance_loss_mlp": 1.00277984,
      "epoch": 0.6653039139060903,
      "flos": 66883444882560.0,
      "grad_norm": 0.8466411745687178,
      "language_loss": 0.55090559,
      "learning_rate": 1.0642987137585278e-06,
      "loss": 0.5721553,
      "num_input_tokens_seen": 119127960,
      "step": 5533,
      "time_per_iteration": 3.1537904739379883
    },
    {
      "auxiliary_loss_clip": 0.01273932,
      "auxiliary_loss_mlp": 0.01021104,
      "balance_loss_clip": 1.04671216,
      "balance_loss_mlp": 1.01356375,
      "epoch": 0.6654241567967294,
      "flos": 21470056669440.0,
      "grad_norm": 1.8683810331111077,
      "language_loss": 0.82866395,
      "learning_rate": 1.0636103237296561e-06,
      "loss": 0.8516143,
      "num_input_tokens_seen": 119146885,
      "step": 5534,
      "time_per_iteration": 2.640620231628418
    },
    {
      "auxiliary_loss_clip": 0.01224165,
      "auxiliary_loss_mlp": 0.0102564,
      "balance_loss_clip": 1.0510869,
      "balance_loss_mlp": 1.01927757,
      "epoch": 0.6655443996873684,
      "flos": 25119514391040.0,
      "grad_norm": 1.7774221543824762,
      "language_loss": 0.8445704,
      "learning_rate": 1.062922075750353e-06,
      "loss": 0.86706847,
      "num_input_tokens_seen": 119166900,
      "step": 5535,
      "time_per_iteration": 2.661778688430786
    },
    {
      "auxiliary_loss_clip": 0.01327753,
      "auxiliary_loss_mlp": 0.0102494,
      "balance_loss_clip": 1.04720831,
      "balance_loss_mlp": 1.01794553,
      "epoch": 0.6656646425780076,
      "flos": 17457326749440.0,
      "grad_norm": 1.9483775966890535,
      "language_loss": 0.72257769,
      "learning_rate": 1.0622339699250267e-06,
      "loss": 0.74610466,
      "num_input_tokens_seen": 119184820,
      "step": 5536,
      "time_per_iteration": 2.723595142364502
    },
    {
      "auxiliary_loss_clip": 0.01324737,
      "auxiliary_loss_mlp": 0.01025561,
      "balance_loss_clip": 1.0451479,
      "balance_loss_mlp": 1.01884055,
      "epoch": 0.6657848854686467,
      "flos": 23434190213760.0,
      "grad_norm": 1.722800401274858,
      "language_loss": 0.79735029,
      "learning_rate": 1.0615460063580624e-06,
      "loss": 0.82085329,
      "num_input_tokens_seen": 119203295,
      "step": 5537,
      "time_per_iteration": 2.7291383743286133
    },
    {
      "auxiliary_loss_clip": 0.01276992,
      "auxiliary_loss_mlp": 0.01025844,
      "balance_loss_clip": 1.04717243,
      "balance_loss_mlp": 1.01932037,
      "epoch": 0.6659051283592857,
      "flos": 11509909459200.0,
      "grad_norm": 2.1523246954035047,
      "language_loss": 0.72678202,
      "learning_rate": 1.060858185153821e-06,
      "loss": 0.74981034,
      "num_input_tokens_seen": 119221395,
      "step": 5538,
      "time_per_iteration": 2.669538974761963
    },
    {
      "auxiliary_loss_clip": 0.01281353,
      "auxiliary_loss_mlp": 0.01030399,
      "balance_loss_clip": 1.04924297,
      "balance_loss_mlp": 1.02296913,
      "epoch": 0.6660253712499249,
      "flos": 20594554571520.0,
      "grad_norm": 2.1864471893660546,
      "language_loss": 0.76109844,
      "learning_rate": 1.0601705064166474e-06,
      "loss": 0.78421599,
      "num_input_tokens_seen": 119239790,
      "step": 5539,
      "time_per_iteration": 2.6237549781799316
    },
    {
      "auxiliary_loss_clip": 0.01276743,
      "auxiliary_loss_mlp": 0.01024175,
      "balance_loss_clip": 1.05150342,
      "balance_loss_mlp": 1.01754379,
      "epoch": 0.666145614140564,
      "flos": 21251504367360.0,
      "grad_norm": 3.130919801878578,
      "language_loss": 0.73383951,
      "learning_rate": 1.0594829702508596e-06,
      "loss": 0.75684869,
      "num_input_tokens_seen": 119257505,
      "step": 5540,
      "time_per_iteration": 2.6599974632263184
    },
    {
      "auxiliary_loss_clip": 0.01327166,
      "auxiliary_loss_mlp": 0.01025753,
      "balance_loss_clip": 1.04713476,
      "balance_loss_mlp": 1.01868415,
      "epoch": 0.666265857031203,
      "flos": 33726188200320.0,
      "grad_norm": 1.8279601845131723,
      "language_loss": 0.55078709,
      "learning_rate": 1.0587955767607592e-06,
      "loss": 0.57431626,
      "num_input_tokens_seen": 119279365,
      "step": 5541,
      "time_per_iteration": 2.8139560222625732
    },
    {
      "auxiliary_loss_clip": 0.01177368,
      "auxiliary_loss_mlp": 0.01027086,
      "balance_loss_clip": 1.04971218,
      "balance_loss_mlp": 1.02031147,
      "epoch": 0.6663860999218422,
      "flos": 17456644391040.0,
      "grad_norm": 2.510514760034638,
      "language_loss": 0.77106804,
      "learning_rate": 1.0581083260506206e-06,
      "loss": 0.79311258,
      "num_input_tokens_seen": 119296150,
      "step": 5542,
      "time_per_iteration": 2.5572571754455566
    },
    {
      "auxiliary_loss_clip": 0.01273635,
      "auxiliary_loss_mlp": 0.01023985,
      "balance_loss_clip": 1.04684281,
      "balance_loss_mlp": 1.01693034,
      "epoch": 0.6665063428124812,
      "flos": 17676740977920.0,
      "grad_norm": 2.6335819576816757,
      "language_loss": 0.7632792,
      "learning_rate": 1.0574212182246993e-06,
      "loss": 0.78625536,
      "num_input_tokens_seen": 119314845,
      "step": 5543,
      "time_per_iteration": 3.7482664585113525
    },
    {
      "auxiliary_loss_clip": 0.01280778,
      "auxiliary_loss_mlp": 0.01025855,
      "balance_loss_clip": 1.04659724,
      "balance_loss_mlp": 1.01864624,
      "epoch": 0.6666265857031203,
      "flos": 27673265687040.0,
      "grad_norm": 3.0837146452480693,
      "language_loss": 0.76162905,
      "learning_rate": 1.0567342533872303e-06,
      "loss": 0.78469539,
      "num_input_tokens_seen": 119334875,
      "step": 5544,
      "time_per_iteration": 3.6421444416046143
    },
    {
      "auxiliary_loss_clip": 0.01275673,
      "auxiliary_loss_mlp": 0.01031027,
      "balance_loss_clip": 1.04853272,
      "balance_loss_mlp": 1.02417862,
      "epoch": 0.6667468285937594,
      "flos": 25046831220480.0,
      "grad_norm": 2.4443624744621304,
      "language_loss": 0.81156909,
      "learning_rate": 1.0560474316424255e-06,
      "loss": 0.83463609,
      "num_input_tokens_seen": 119354635,
      "step": 5545,
      "time_per_iteration": 2.681339979171753
    },
    {
      "auxiliary_loss_clip": 0.0127855,
      "auxiliary_loss_mlp": 0.01029447,
      "balance_loss_clip": 1.04562879,
      "balance_loss_mlp": 1.0217762,
      "epoch": 0.6668670714843985,
      "flos": 22780472641920.0,
      "grad_norm": 2.6227229720593113,
      "language_loss": 0.73554456,
      "learning_rate": 1.0553607530944746e-06,
      "loss": 0.75862455,
      "num_input_tokens_seen": 119372690,
      "step": 5546,
      "time_per_iteration": 3.623143196105957
    },
    {
      "auxiliary_loss_clip": 0.01325814,
      "auxiliary_loss_mlp": 0.01030162,
      "balance_loss_clip": 1.04409099,
      "balance_loss_mlp": 1.02282798,
      "epoch": 0.6669873143750376,
      "flos": 22163886754560.0,
      "grad_norm": 2.0973838001175666,
      "language_loss": 0.8974511,
      "learning_rate": 1.0546742178475463e-06,
      "loss": 0.92101085,
      "num_input_tokens_seen": 119391685,
      "step": 5547,
      "time_per_iteration": 2.674889326095581
    },
    {
      "auxiliary_loss_clip": 0.01277383,
      "auxiliary_loss_mlp": 0.01027517,
      "balance_loss_clip": 1.04486775,
      "balance_loss_mlp": 1.02108574,
      "epoch": 0.6671075572656767,
      "flos": 20514832335360.0,
      "grad_norm": 2.388717979447034,
      "language_loss": 0.86931252,
      "learning_rate": 1.0539878260057868e-06,
      "loss": 0.89236152,
      "num_input_tokens_seen": 119410725,
      "step": 5548,
      "time_per_iteration": 2.722992181777954
    },
    {
      "auxiliary_loss_clip": 0.01238063,
      "auxiliary_loss_mlp": 0.01028942,
      "balance_loss_clip": 1.05536795,
      "balance_loss_mlp": 1.02108622,
      "epoch": 0.6672278001563158,
      "flos": 17931203902080.0,
      "grad_norm": 2.7735877869205994,
      "language_loss": 0.68935311,
      "learning_rate": 1.0533015776733226e-06,
      "loss": 0.71202314,
      "num_input_tokens_seen": 119426875,
      "step": 5549,
      "time_per_iteration": 2.5754356384277344
    },
    {
      "auxiliary_loss_clip": 0.01271621,
      "auxiliary_loss_mlp": 0.01026386,
      "balance_loss_clip": 1.04880214,
      "balance_loss_mlp": 1.01917958,
      "epoch": 0.6673480430469548,
      "flos": 22342146975360.0,
      "grad_norm": 2.351925661412378,
      "language_loss": 0.78394914,
      "learning_rate": 1.0526154729542566e-06,
      "loss": 0.80692923,
      "num_input_tokens_seen": 119446935,
      "step": 5550,
      "time_per_iteration": 2.6515743732452393
    },
    {
      "auxiliary_loss_clip": 0.01327793,
      "auxiliary_loss_mlp": 0.01029967,
      "balance_loss_clip": 1.04934239,
      "balance_loss_mlp": 1.02236724,
      "epoch": 0.6674682859375939,
      "flos": 20703830722560.0,
      "grad_norm": 2.858074486793648,
      "language_loss": 0.80555499,
      "learning_rate": 1.0519295119526699e-06,
      "loss": 0.82913256,
      "num_input_tokens_seen": 119463240,
      "step": 5551,
      "time_per_iteration": 3.6462242603302
    },
    {
      "auxiliary_loss_clip": 0.01279909,
      "auxiliary_loss_mlp": 0.01027592,
      "balance_loss_clip": 1.04876125,
      "balance_loss_mlp": 1.02047837,
      "epoch": 0.667588528828233,
      "flos": 26206673379840.0,
      "grad_norm": 1.6096276633267281,
      "language_loss": 0.83412254,
      "learning_rate": 1.0512436947726227e-06,
      "loss": 0.85719752,
      "num_input_tokens_seen": 119484655,
      "step": 5552,
      "time_per_iteration": 2.6560628414154053
    },
    {
      "auxiliary_loss_clip": 0.01325365,
      "auxiliary_loss_mlp": 0.01023563,
      "balance_loss_clip": 1.04454136,
      "balance_loss_mlp": 1.01667285,
      "epoch": 0.6677087717188721,
      "flos": 23071025756160.0,
      "grad_norm": 2.5964858622827887,
      "language_loss": 0.65482211,
      "learning_rate": 1.0505580215181517e-06,
      "loss": 0.67831135,
      "num_input_tokens_seen": 119502895,
      "step": 5553,
      "time_per_iteration": 2.7225420475006104
    },
    {
      "auxiliary_loss_clip": 0.01212946,
      "auxiliary_loss_mlp": 0.01003318,
      "balance_loss_clip": 1.01132274,
      "balance_loss_mlp": 1.0023644,
      "epoch": 0.6678290146095112,
      "flos": 70941315219840.0,
      "grad_norm": 0.7829112256429517,
      "language_loss": 0.56639367,
      "learning_rate": 1.0498724922932753e-06,
      "loss": 0.58855635,
      "num_input_tokens_seen": 119561010,
      "step": 5554,
      "time_per_iteration": 3.1799941062927246
    },
    {
      "auxiliary_loss_clip": 0.01184596,
      "auxiliary_loss_mlp": 0.01030035,
      "balance_loss_clip": 1.05464649,
      "balance_loss_mlp": 1.02238512,
      "epoch": 0.6679492575001503,
      "flos": 18661088263680.0,
      "grad_norm": 3.6486238724692686,
      "language_loss": 0.86665249,
      "learning_rate": 1.0491871072019851e-06,
      "loss": 0.88879883,
      "num_input_tokens_seen": 119578900,
      "step": 5555,
      "time_per_iteration": 2.597923517227173
    },
    {
      "auxiliary_loss_clip": 0.01331104,
      "auxiliary_loss_mlp": 0.01026313,
      "balance_loss_clip": 1.04425752,
      "balance_loss_mlp": 1.01942253,
      "epoch": 0.6680695003907894,
      "flos": 29711985822720.0,
      "grad_norm": 2.128235629692362,
      "language_loss": 0.63705784,
      "learning_rate": 1.0485018663482555e-06,
      "loss": 0.66063201,
      "num_input_tokens_seen": 119598920,
      "step": 5556,
      "time_per_iteration": 2.7478344440460205
    },
    {
      "auxiliary_loss_clip": 0.01226778,
      "auxiliary_loss_mlp": 0.01027552,
      "balance_loss_clip": 1.04914689,
      "balance_loss_mlp": 1.02032185,
      "epoch": 0.6681897432814284,
      "flos": 28218964083840.0,
      "grad_norm": 2.615700651940876,
      "language_loss": 0.70484126,
      "learning_rate": 1.0478167698360354e-06,
      "loss": 0.72738457,
      "num_input_tokens_seen": 119618220,
      "step": 5557,
      "time_per_iteration": 2.661102533340454
    },
    {
      "auxiliary_loss_clip": 0.01222377,
      "auxiliary_loss_mlp": 0.01023672,
      "balance_loss_clip": 1.04593968,
      "balance_loss_mlp": 1.01639438,
      "epoch": 0.6683099861720676,
      "flos": 25046543911680.0,
      "grad_norm": 1.7652814679903237,
      "language_loss": 0.70249152,
      "learning_rate": 1.0471318177692556e-06,
      "loss": 0.72495198,
      "num_input_tokens_seen": 119638520,
      "step": 5558,
      "time_per_iteration": 2.651160717010498
    },
    {
      "auxiliary_loss_clip": 0.01375629,
      "auxiliary_loss_mlp": 0.01029814,
      "balance_loss_clip": 1.0444783,
      "balance_loss_mlp": 1.02212477,
      "epoch": 0.6684302290627067,
      "flos": 22996977868800.0,
      "grad_norm": 2.1554176479253457,
      "language_loss": 0.75217497,
      "learning_rate": 1.046447010251821e-06,
      "loss": 0.77622938,
      "num_input_tokens_seen": 119655850,
      "step": 5559,
      "time_per_iteration": 2.7954230308532715
    },
    {
      "auxiliary_loss_clip": 0.01275726,
      "auxiliary_loss_mlp": 0.01022536,
      "balance_loss_clip": 1.04981112,
      "balance_loss_mlp": 1.01578867,
      "epoch": 0.6685504719533457,
      "flos": 26573824247040.0,
      "grad_norm": 1.796774397085072,
      "language_loss": 0.75664574,
      "learning_rate": 1.0457623473876157e-06,
      "loss": 0.7796284,
      "num_input_tokens_seen": 119675355,
      "step": 5560,
      "time_per_iteration": 2.7273337841033936
    },
    {
      "auxiliary_loss_clip": 0.0117708,
      "auxiliary_loss_mlp": 0.0102695,
      "balance_loss_clip": 1.05125856,
      "balance_loss_mlp": 1.02017879,
      "epoch": 0.6686707148439849,
      "flos": 28986087870720.0,
      "grad_norm": 2.1517089540393233,
      "language_loss": 0.7120266,
      "learning_rate": 1.0450778292805046e-06,
      "loss": 0.73406696,
      "num_input_tokens_seen": 119695340,
      "step": 5561,
      "time_per_iteration": 2.64261794090271
    },
    {
      "auxiliary_loss_clip": 0.01230049,
      "auxiliary_loss_mlp": 0.0102738,
      "balance_loss_clip": 1.04799962,
      "balance_loss_mlp": 1.02036476,
      "epoch": 0.6687909577346239,
      "flos": 23623152687360.0,
      "grad_norm": 6.008140864238958,
      "language_loss": 0.78930402,
      "learning_rate": 1.0443934560343267e-06,
      "loss": 0.81187832,
      "num_input_tokens_seen": 119716750,
      "step": 5562,
      "time_per_iteration": 2.6808807849884033
    },
    {
      "auxiliary_loss_clip": 0.01315182,
      "auxiliary_loss_mlp": 0.01023942,
      "balance_loss_clip": 1.04464149,
      "balance_loss_mlp": 1.01682234,
      "epoch": 0.668911200625263,
      "flos": 23148593176320.0,
      "grad_norm": 2.064621233969995,
      "language_loss": 0.77729344,
      "learning_rate": 1.0437092277529034e-06,
      "loss": 0.80068469,
      "num_input_tokens_seen": 119736005,
      "step": 5563,
      "time_per_iteration": 2.7106070518493652
    },
    {
      "auxiliary_loss_clip": 0.01272593,
      "auxiliary_loss_mlp": 0.01023047,
      "balance_loss_clip": 1.04684246,
      "balance_loss_mlp": 1.01655638,
      "epoch": 0.6690314435159022,
      "flos": 18551919853440.0,
      "grad_norm": 1.9842519419870406,
      "language_loss": 0.73648596,
      "learning_rate": 1.0430251445400292e-06,
      "loss": 0.75944233,
      "num_input_tokens_seen": 119754050,
      "step": 5564,
      "time_per_iteration": 2.7483930587768555
    },
    {
      "auxiliary_loss_clip": 0.01522951,
      "auxiliary_loss_mlp": 0.0102525,
      "balance_loss_clip": 1.04345012,
      "balance_loss_mlp": 1.01825285,
      "epoch": 0.6691516864065412,
      "flos": 31759540704000.0,
      "grad_norm": 1.9054118322551583,
      "language_loss": 0.62414324,
      "learning_rate": 1.0423412064994787e-06,
      "loss": 0.6496253,
      "num_input_tokens_seen": 119774820,
      "step": 5565,
      "time_per_iteration": 3.014051914215088
    },
    {
      "auxiliary_loss_clip": 0.01325896,
      "auxiliary_loss_mlp": 0.01025314,
      "balance_loss_clip": 1.04370022,
      "balance_loss_mlp": 1.01877856,
      "epoch": 0.6692719292971803,
      "flos": 34933864296960.0,
      "grad_norm": 2.2393843611594098,
      "language_loss": 0.73869038,
      "learning_rate": 1.0416574137350064e-06,
      "loss": 0.7622025,
      "num_input_tokens_seen": 119795525,
      "step": 5566,
      "time_per_iteration": 3.0932059288024902
    },
    {
      "auxiliary_loss_clip": 0.01225872,
      "auxiliary_loss_mlp": 0.01026016,
      "balance_loss_clip": 1.04980755,
      "balance_loss_mlp": 1.0180831,
      "epoch": 0.6693921721878194,
      "flos": 20449188230400.0,
      "grad_norm": 2.296372225022629,
      "language_loss": 0.81224287,
      "learning_rate": 1.0409737663503428e-06,
      "loss": 0.8347618,
      "num_input_tokens_seen": 119813905,
      "step": 5567,
      "time_per_iteration": 2.6090338230133057
    },
    {
      "auxiliary_loss_clip": 0.01225604,
      "auxiliary_loss_mlp": 0.01030428,
      "balance_loss_clip": 1.04541075,
      "balance_loss_mlp": 1.02276254,
      "epoch": 0.6695124150784585,
      "flos": 16614538963200.0,
      "grad_norm": 1.861960652148183,
      "language_loss": 0.82584631,
      "learning_rate": 1.040290264449196e-06,
      "loss": 0.84840667,
      "num_input_tokens_seen": 119832010,
      "step": 5568,
      "time_per_iteration": 2.677114963531494
    },
    {
      "auxiliary_loss_clip": 0.01220377,
      "auxiliary_loss_mlp": 0.01024262,
      "balance_loss_clip": 1.04956675,
      "balance_loss_mlp": 1.01766706,
      "epoch": 0.6696326579690975,
      "flos": 26652145852800.0,
      "grad_norm": 3.1854411183147255,
      "language_loss": 0.64382654,
      "learning_rate": 1.0396069081352532e-06,
      "loss": 0.66627288,
      "num_input_tokens_seen": 119851165,
      "step": 5569,
      "time_per_iteration": 3.8532094955444336
    },
    {
      "auxiliary_loss_clip": 0.01066056,
      "auxiliary_loss_mlp": 0.01003242,
      "balance_loss_clip": 1.00974667,
      "balance_loss_mlp": 1.00232458,
      "epoch": 0.6697529008597367,
      "flos": 66964603662720.0,
      "grad_norm": 0.7734666089265403,
      "language_loss": 0.55993795,
      "learning_rate": 1.0389236975121782e-06,
      "loss": 0.5806309,
      "num_input_tokens_seen": 119906015,
      "step": 5570,
      "time_per_iteration": 4.4609832763671875
    },
    {
      "auxiliary_loss_clip": 0.01179282,
      "auxiliary_loss_mlp": 0.0102583,
      "balance_loss_clip": 1.05053806,
      "balance_loss_mlp": 1.01807499,
      "epoch": 0.6698731437503758,
      "flos": 20886939279360.0,
      "grad_norm": 4.122730893076424,
      "language_loss": 0.70992804,
      "learning_rate": 1.0382406326836147e-06,
      "loss": 0.73197919,
      "num_input_tokens_seen": 119925160,
      "step": 5571,
      "time_per_iteration": 2.596496820449829
    },
    {
      "auxiliary_loss_clip": 0.01235238,
      "auxiliary_loss_mlp": 0.01030334,
      "balance_loss_clip": 1.05233073,
      "balance_loss_mlp": 1.02226305,
      "epoch": 0.6699933866410148,
      "flos": 20409470766720.0,
      "grad_norm": 3.022771400080622,
      "language_loss": 0.76064414,
      "learning_rate": 1.0375577137531828e-06,
      "loss": 0.7832998,
      "num_input_tokens_seen": 119943720,
      "step": 5572,
      "time_per_iteration": 2.6609222888946533
    },
    {
      "auxiliary_loss_clip": 0.01278393,
      "auxiliary_loss_mlp": 0.01027496,
      "balance_loss_clip": 1.04822826,
      "balance_loss_mlp": 1.01980066,
      "epoch": 0.670113629531654,
      "flos": 29023075900800.0,
      "grad_norm": 6.904218405916863,
      "language_loss": 0.72251236,
      "learning_rate": 1.0368749408244802e-06,
      "loss": 0.74557132,
      "num_input_tokens_seen": 119966640,
      "step": 5573,
      "time_per_iteration": 3.5561344623565674
    },
    {
      "auxiliary_loss_clip": 0.01223187,
      "auxiliary_loss_mlp": 0.01026281,
      "balance_loss_clip": 1.04827476,
      "balance_loss_mlp": 1.0196799,
      "epoch": 0.670233872422293,
      "flos": 19791699730560.0,
      "grad_norm": 1.714761703428131,
      "language_loss": 0.78663015,
      "learning_rate": 1.0361923140010836e-06,
      "loss": 0.80912483,
      "num_input_tokens_seen": 119985125,
      "step": 5574,
      "time_per_iteration": 2.6306159496307373
    },
    {
      "auxiliary_loss_clip": 0.01233798,
      "auxiliary_loss_mlp": 0.01021436,
      "balance_loss_clip": 1.05007315,
      "balance_loss_mlp": 1.01463497,
      "epoch": 0.6703541153129321,
      "flos": 24243689070720.0,
      "grad_norm": 2.295365125679275,
      "language_loss": 0.63622963,
      "learning_rate": 1.0355098333865455e-06,
      "loss": 0.65878201,
      "num_input_tokens_seen": 120004355,
      "step": 5575,
      "time_per_iteration": 2.6698997020721436
    },
    {
      "auxiliary_loss_clip": 0.01232107,
      "auxiliary_loss_mlp": 0.01028763,
      "balance_loss_clip": 1.05641091,
      "balance_loss_mlp": 1.022156,
      "epoch": 0.6704743582035713,
      "flos": 26688523351680.0,
      "grad_norm": 1.9906968029082088,
      "language_loss": 0.69355941,
      "learning_rate": 1.0348274990844006e-06,
      "loss": 0.71616817,
      "num_input_tokens_seen": 120027115,
      "step": 5576,
      "time_per_iteration": 2.756074905395508
    },
    {
      "auxiliary_loss_clip": 0.01223182,
      "auxiliary_loss_mlp": 0.01027315,
      "balance_loss_clip": 1.04929638,
      "balance_loss_mlp": 1.01949441,
      "epoch": 0.6705946010942103,
      "flos": 23514379326720.0,
      "grad_norm": 1.8122443430414128,
      "language_loss": 0.72340071,
      "learning_rate": 1.034145311198155e-06,
      "loss": 0.74590576,
      "num_input_tokens_seen": 120047130,
      "step": 5577,
      "time_per_iteration": 3.597355842590332
    },
    {
      "auxiliary_loss_clip": 0.01173612,
      "auxiliary_loss_mlp": 0.01027546,
      "balance_loss_clip": 1.04840899,
      "balance_loss_mlp": 1.0212307,
      "epoch": 0.6707148439848494,
      "flos": 24061011477120.0,
      "grad_norm": 1.8948795596469417,
      "language_loss": 0.6380927,
      "learning_rate": 1.0334632698312989e-06,
      "loss": 0.66010427,
      "num_input_tokens_seen": 120067925,
      "step": 5578,
      "time_per_iteration": 2.6826212406158447
    },
    {
      "auxiliary_loss_clip": 0.01273213,
      "auxiliary_loss_mlp": 0.01030101,
      "balance_loss_clip": 1.04654813,
      "balance_loss_mlp": 1.02271295,
      "epoch": 0.6708350868754885,
      "flos": 22528667324160.0,
      "grad_norm": 1.9051873128306942,
      "language_loss": 0.75533199,
      "learning_rate": 1.032781375087295e-06,
      "loss": 0.77836514,
      "num_input_tokens_seen": 120087825,
      "step": 5579,
      "time_per_iteration": 2.7347004413604736
    },
    {
      "auxiliary_loss_clip": 0.0127781,
      "auxiliary_loss_mlp": 0.0102338,
      "balance_loss_clip": 1.05029786,
      "balance_loss_mlp": 1.01643598,
      "epoch": 0.6709553297661276,
      "flos": 25227749047680.0,
      "grad_norm": 1.5516214510527269,
      "language_loss": 0.67288536,
      "learning_rate": 1.0320996270695891e-06,
      "loss": 0.69589722,
      "num_input_tokens_seen": 120108895,
      "step": 5580,
      "time_per_iteration": 2.7064363956451416
    },
    {
      "auxiliary_loss_clip": 0.01328047,
      "auxiliary_loss_mlp": 0.01032903,
      "balance_loss_clip": 1.04432952,
      "balance_loss_mlp": 1.02541661,
      "epoch": 0.6710755726567667,
      "flos": 20448757267200.0,
      "grad_norm": 2.102990234178898,
      "language_loss": 0.73181462,
      "learning_rate": 1.0314180258815998e-06,
      "loss": 0.75542414,
      "num_input_tokens_seen": 120127535,
      "step": 5581,
      "time_per_iteration": 2.747220516204834
    },
    {
      "auxiliary_loss_clip": 0.01318778,
      "auxiliary_loss_mlp": 0.01022927,
      "balance_loss_clip": 1.04337263,
      "balance_loss_mlp": 1.01619792,
      "epoch": 0.6711958155474057,
      "flos": 25995411538560.0,
      "grad_norm": 2.1947724924685907,
      "language_loss": 0.74419987,
      "learning_rate": 1.0307365716267247e-06,
      "loss": 0.76761693,
      "num_input_tokens_seen": 120147980,
      "step": 5582,
      "time_per_iteration": 2.9099223613739014
    },
    {
      "auxiliary_loss_clip": 0.01226163,
      "auxiliary_loss_mlp": 0.01027972,
      "balance_loss_clip": 1.05068135,
      "balance_loss_mlp": 1.02036619,
      "epoch": 0.6713160584380449,
      "flos": 19937712516480.0,
      "grad_norm": 2.346225919379869,
      "language_loss": 0.78521931,
      "learning_rate": 1.0300552644083423e-06,
      "loss": 0.80776066,
      "num_input_tokens_seen": 120166905,
      "step": 5583,
      "time_per_iteration": 2.5653109550476074
    },
    {
      "auxiliary_loss_clip": 0.01334007,
      "auxiliary_loss_mlp": 0.01025776,
      "balance_loss_clip": 1.05062056,
      "balance_loss_mlp": 1.01811099,
      "epoch": 0.6714363013286839,
      "flos": 18223373128320.0,
      "grad_norm": 15.11653414489998,
      "language_loss": 0.7247566,
      "learning_rate": 1.0293741043298036e-06,
      "loss": 0.74835443,
      "num_input_tokens_seen": 120185255,
      "step": 5584,
      "time_per_iteration": 2.723883867263794
    },
    {
      "auxiliary_loss_clip": 0.01340579,
      "auxiliary_loss_mlp": 0.01027178,
      "balance_loss_clip": 1.0558517,
      "balance_loss_mlp": 1.01941144,
      "epoch": 0.671556544219323,
      "flos": 25812374808960.0,
      "grad_norm": 2.027377042860145,
      "language_loss": 0.71608829,
      "learning_rate": 1.0286930914944436e-06,
      "loss": 0.73976588,
      "num_input_tokens_seen": 120205070,
      "step": 5585,
      "time_per_iteration": 2.7461986541748047
    },
    {
      "auxiliary_loss_clip": 0.0117707,
      "auxiliary_loss_mlp": 0.01024231,
      "balance_loss_clip": 1.04761124,
      "balance_loss_mlp": 1.01716161,
      "epoch": 0.6716767871099621,
      "flos": 15850431918720.0,
      "grad_norm": 2.509918754138467,
      "language_loss": 0.76958847,
      "learning_rate": 1.0280122260055684e-06,
      "loss": 0.79160154,
      "num_input_tokens_seen": 120220780,
      "step": 5586,
      "time_per_iteration": 2.6147093772888184
    },
    {
      "auxiliary_loss_clip": 0.01180183,
      "auxiliary_loss_mlp": 0.01028313,
      "balance_loss_clip": 1.05247164,
      "balance_loss_mlp": 1.02063608,
      "epoch": 0.6717970300006012,
      "flos": 19756112330880.0,
      "grad_norm": 2.004717764097082,
      "language_loss": 0.82126486,
      "learning_rate": 1.0273315079664652e-06,
      "loss": 0.84334981,
      "num_input_tokens_seen": 120238735,
      "step": 5587,
      "time_per_iteration": 2.606456995010376
    },
    {
      "auxiliary_loss_clip": 0.01229864,
      "auxiliary_loss_mlp": 0.01027951,
      "balance_loss_clip": 1.04959869,
      "balance_loss_mlp": 1.02096796,
      "epoch": 0.6719172728912403,
      "flos": 25485049146240.0,
      "grad_norm": 2.6876708252991612,
      "language_loss": 0.74212104,
      "learning_rate": 1.0266509374803992e-06,
      "loss": 0.76469922,
      "num_input_tokens_seen": 120259895,
      "step": 5588,
      "time_per_iteration": 2.6926281452178955
    },
    {
      "auxiliary_loss_clip": 0.01179414,
      "auxiliary_loss_mlp": 0.02567841,
      "balance_loss_clip": 1.05041337,
      "balance_loss_mlp": 0.99993938,
      "epoch": 0.6720375157818794,
      "flos": 15880344969600.0,
      "grad_norm": 2.564801395423503,
      "language_loss": 0.84542668,
      "learning_rate": 1.0259705146506123e-06,
      "loss": 0.88289928,
      "num_input_tokens_seen": 120274790,
      "step": 5589,
      "time_per_iteration": 2.5299551486968994
    },
    {
      "auxiliary_loss_clip": 0.01230123,
      "auxiliary_loss_mlp": 0.01025992,
      "balance_loss_clip": 1.04950929,
      "balance_loss_mlp": 1.01913154,
      "epoch": 0.6721577586725185,
      "flos": 32010843231360.0,
      "grad_norm": 2.1434743146822317,
      "language_loss": 0.7743656,
      "learning_rate": 1.025290239580324e-06,
      "loss": 0.79692674,
      "num_input_tokens_seen": 120295460,
      "step": 5590,
      "time_per_iteration": 2.7838857173919678
    },
    {
      "auxiliary_loss_clip": 0.01375993,
      "auxiliary_loss_mlp": 0.01026652,
      "balance_loss_clip": 1.04323542,
      "balance_loss_mlp": 1.01918411,
      "epoch": 0.6722780015631575,
      "flos": 20737873837440.0,
      "grad_norm": 1.9392342272601666,
      "language_loss": 0.75401914,
      "learning_rate": 1.0246101123727313e-06,
      "loss": 0.77804559,
      "num_input_tokens_seen": 120314440,
      "step": 5591,
      "time_per_iteration": 2.735318660736084
    },
    {
      "auxiliary_loss_clip": 0.01227746,
      "auxiliary_loss_mlp": 0.01029972,
      "balance_loss_clip": 1.0481714,
      "balance_loss_mlp": 1.02301013,
      "epoch": 0.6723982444537967,
      "flos": 16909617191040.0,
      "grad_norm": 2.4746191766339725,
      "language_loss": 0.7863903,
      "learning_rate": 1.0239301331310085e-06,
      "loss": 0.80896747,
      "num_input_tokens_seen": 120332060,
      "step": 5592,
      "time_per_iteration": 2.672780752182007
    },
    {
      "auxiliary_loss_clip": 0.01224586,
      "auxiliary_loss_mlp": 0.01023218,
      "balance_loss_clip": 1.04942572,
      "balance_loss_mlp": 1.01671219,
      "epoch": 0.6725184873444358,
      "flos": 20667812359680.0,
      "grad_norm": 1.7520118090416217,
      "language_loss": 0.88552541,
      "learning_rate": 1.0232503019583088e-06,
      "loss": 0.90800345,
      "num_input_tokens_seen": 120351670,
      "step": 5593,
      "time_per_iteration": 2.62329363822937
    },
    {
      "auxiliary_loss_clip": 0.01223181,
      "auxiliary_loss_mlp": 0.01024288,
      "balance_loss_clip": 1.04909801,
      "balance_loss_mlp": 1.01689446,
      "epoch": 0.6726387302350748,
      "flos": 23727616416000.0,
      "grad_norm": 2.0464782714459666,
      "language_loss": 0.70052189,
      "learning_rate": 1.0225706189577619e-06,
      "loss": 0.72299659,
      "num_input_tokens_seen": 120370195,
      "step": 5594,
      "time_per_iteration": 2.6584572792053223
    },
    {
      "auxiliary_loss_clip": 0.01228497,
      "auxiliary_loss_mlp": 0.01027714,
      "balance_loss_clip": 1.05119705,
      "balance_loss_mlp": 1.02037644,
      "epoch": 0.672758973125714,
      "flos": 15188274650880.0,
      "grad_norm": 2.052416763784095,
      "language_loss": 0.74639338,
      "learning_rate": 1.021891084232475e-06,
      "loss": 0.76895553,
      "num_input_tokens_seen": 120388130,
      "step": 5595,
      "time_per_iteration": 2.5460262298583984
    },
    {
      "auxiliary_loss_clip": 0.01228017,
      "auxiliary_loss_mlp": 0.01025519,
      "balance_loss_clip": 1.04835296,
      "balance_loss_mlp": 1.01822317,
      "epoch": 0.672879216016353,
      "flos": 18077252601600.0,
      "grad_norm": 2.3847622003549027,
      "language_loss": 0.79696798,
      "learning_rate": 1.0212116978855325e-06,
      "loss": 0.81950331,
      "num_input_tokens_seen": 120406145,
      "step": 5596,
      "time_per_iteration": 4.397632837295532
    },
    {
      "auxiliary_loss_clip": 0.01320389,
      "auxiliary_loss_mlp": 0.01024907,
      "balance_loss_clip": 1.04672766,
      "balance_loss_mlp": 1.0175159,
      "epoch": 0.6729994589069921,
      "flos": 23476349802240.0,
      "grad_norm": 2.2502161199821424,
      "language_loss": 0.79025054,
      "learning_rate": 1.020532460019997e-06,
      "loss": 0.81370354,
      "num_input_tokens_seen": 120425395,
      "step": 5597,
      "time_per_iteration": 2.7380754947662354
    },
    {
      "auxiliary_loss_clip": 0.01476237,
      "auxiliary_loss_mlp": 0.01027082,
      "balance_loss_clip": 1.04276514,
      "balance_loss_mlp": 1.01968551,
      "epoch": 0.6731197017976313,
      "flos": 26322018929280.0,
      "grad_norm": 3.031885451138773,
      "language_loss": 0.71204442,
      "learning_rate": 1.0198533707389096e-06,
      "loss": 0.73707759,
      "num_input_tokens_seen": 120446270,
      "step": 5598,
      "time_per_iteration": 3.0769906044006348
    },
    {
      "auxiliary_loss_clip": 0.01225405,
      "auxiliary_loss_mlp": 0.0256745,
      "balance_loss_clip": 1.04951501,
      "balance_loss_mlp": 0.99996507,
      "epoch": 0.6732399446882703,
      "flos": 21616428591360.0,
      "grad_norm": 1.7364839209956702,
      "language_loss": 0.73253977,
      "learning_rate": 1.0191744301452853e-06,
      "loss": 0.77046829,
      "num_input_tokens_seen": 120465570,
      "step": 5599,
      "time_per_iteration": 3.9378957748413086
    },
    {
      "auxiliary_loss_clip": 0.01174603,
      "auxiliary_loss_mlp": 0.0102681,
      "balance_loss_clip": 1.04758406,
      "balance_loss_mlp": 1.01979458,
      "epoch": 0.6733601875789094,
      "flos": 25880173729920.0,
      "grad_norm": 2.0239720144109308,
      "language_loss": 0.70333827,
      "learning_rate": 1.0184956383421208e-06,
      "loss": 0.72535235,
      "num_input_tokens_seen": 120484220,
      "step": 5600,
      "time_per_iteration": 2.7296152114868164
    },
    {
      "auxiliary_loss_clip": 0.01231567,
      "auxiliary_loss_mlp": 0.01024986,
      "balance_loss_clip": 1.0507617,
      "balance_loss_mlp": 1.01793528,
      "epoch": 0.6734804304695485,
      "flos": 22929573997440.0,
      "grad_norm": 2.4059797444957733,
      "language_loss": 0.65669161,
      "learning_rate": 1.017816995432387e-06,
      "loss": 0.67925715,
      "num_input_tokens_seen": 120503320,
      "step": 5601,
      "time_per_iteration": 2.6747119426727295
    },
    {
      "auxiliary_loss_clip": 0.01274238,
      "auxiliary_loss_mlp": 0.01024037,
      "balance_loss_clip": 1.0468328,
      "balance_loss_mlp": 1.01752543,
      "epoch": 0.6736006733601876,
      "flos": 18697968552960.0,
      "grad_norm": 7.723861690522186,
      "language_loss": 0.74046087,
      "learning_rate": 1.0171385015190353e-06,
      "loss": 0.76344359,
      "num_input_tokens_seen": 120523180,
      "step": 5602,
      "time_per_iteration": 2.6880056858062744
    },
    {
      "auxiliary_loss_clip": 0.01276467,
      "auxiliary_loss_mlp": 0.02565142,
      "balance_loss_clip": 1.05141437,
      "balance_loss_mlp": 0.99991649,
      "epoch": 0.6737209162508266,
      "flos": 19427745173760.0,
      "grad_norm": 1.9573512777289763,
      "language_loss": 0.73793697,
      "learning_rate": 1.0164601567049908e-06,
      "loss": 0.77635306,
      "num_input_tokens_seen": 120541710,
      "step": 5603,
      "time_per_iteration": 3.854444742202759
    },
    {
      "auxiliary_loss_clip": 0.01280607,
      "auxiliary_loss_mlp": 0.01029308,
      "balance_loss_clip": 1.04870772,
      "balance_loss_mlp": 1.0218606,
      "epoch": 0.6738411591414658,
      "flos": 20158060498560.0,
      "grad_norm": 1.9243560925337178,
      "language_loss": 0.803868,
      "learning_rate": 1.015781961093158e-06,
      "loss": 0.82696712,
      "num_input_tokens_seen": 120561030,
      "step": 5604,
      "time_per_iteration": 2.6697568893432617
    },
    {
      "auxiliary_loss_clip": 0.01277983,
      "auxiliary_loss_mlp": 0.01027483,
      "balance_loss_clip": 1.04496837,
      "balance_loss_mlp": 1.02022314,
      "epoch": 0.6739614020321049,
      "flos": 21653847584640.0,
      "grad_norm": 1.5910099515217886,
      "language_loss": 0.77140808,
      "learning_rate": 1.0151039147864197e-06,
      "loss": 0.79446274,
      "num_input_tokens_seen": 120581005,
      "step": 5605,
      "time_per_iteration": 2.6987903118133545
    },
    {
      "auxiliary_loss_clip": 0.01474731,
      "auxiliary_loss_mlp": 0.01028423,
      "balance_loss_clip": 1.05039334,
      "balance_loss_mlp": 1.02088857,
      "epoch": 0.6740816449227439,
      "flos": 19171702051200.0,
      "grad_norm": 2.615116744511832,
      "language_loss": 0.66014588,
      "learning_rate": 1.0144260178876336e-06,
      "loss": 0.68517739,
      "num_input_tokens_seen": 120600350,
      "step": 5606,
      "time_per_iteration": 2.930556058883667
    },
    {
      "auxiliary_loss_clip": 0.01283006,
      "auxiliary_loss_mlp": 0.01022493,
      "balance_loss_clip": 1.04785609,
      "balance_loss_mlp": 1.01601434,
      "epoch": 0.6742018878133831,
      "flos": 21097015971840.0,
      "grad_norm": 3.218628284791756,
      "language_loss": 0.67821574,
      "learning_rate": 1.0137482704996388e-06,
      "loss": 0.70127076,
      "num_input_tokens_seen": 120614700,
      "step": 5607,
      "time_per_iteration": 2.769575595855713
    },
    {
      "auxiliary_loss_clip": 0.01329213,
      "auxiliary_loss_mlp": 0.01031838,
      "balance_loss_clip": 1.04827583,
      "balance_loss_mlp": 1.02452493,
      "epoch": 0.6743221307040221,
      "flos": 23549966726400.0,
      "grad_norm": 3.547734277126329,
      "language_loss": 0.78719258,
      "learning_rate": 1.0130706727252461e-06,
      "loss": 0.81080312,
      "num_input_tokens_seen": 120631755,
      "step": 5608,
      "time_per_iteration": 2.696565866470337
    },
    {
      "auxiliary_loss_clip": 0.01326529,
      "auxiliary_loss_mlp": 0.01024883,
      "balance_loss_clip": 1.04661989,
      "balance_loss_mlp": 1.01802552,
      "epoch": 0.6744423735946612,
      "flos": 16249542912000.0,
      "grad_norm": 3.109336959465384,
      "language_loss": 0.67726475,
      "learning_rate": 1.0123932246672468e-06,
      "loss": 0.70077884,
      "num_input_tokens_seen": 120645900,
      "step": 5609,
      "time_per_iteration": 2.7819058895111084
    },
    {
      "auxiliary_loss_clip": 0.0127217,
      "auxiliary_loss_mlp": 0.02507057,
      "balance_loss_clip": 1.00906444,
      "balance_loss_mlp": 0.99983454,
      "epoch": 0.6745626164853004,
      "flos": 57843257829120.0,
      "grad_norm": 0.7448829340269733,
      "language_loss": 0.55764842,
      "learning_rate": 1.0117159264284114e-06,
      "loss": 0.59544063,
      "num_input_tokens_seen": 120709070,
      "step": 5610,
      "time_per_iteration": 3.3621935844421387
    },
    {
      "auxiliary_loss_clip": 0.01277974,
      "auxiliary_loss_mlp": 0.01024892,
      "balance_loss_clip": 1.04860783,
      "balance_loss_mlp": 1.01790357,
      "epoch": 0.6746828593759394,
      "flos": 20485027025280.0,
      "grad_norm": 1.9434949557435146,
      "language_loss": 0.76957762,
      "learning_rate": 1.0110387781114837e-06,
      "loss": 0.79260629,
      "num_input_tokens_seen": 120727685,
      "step": 5611,
      "time_per_iteration": 2.843716621398926
    },
    {
      "auxiliary_loss_clip": 0.01175358,
      "auxiliary_loss_mlp": 0.01023586,
      "balance_loss_clip": 1.04949069,
      "balance_loss_mlp": 1.01631379,
      "epoch": 0.6748031022665785,
      "flos": 19208223204480.0,
      "grad_norm": 2.227642783321111,
      "language_loss": 0.77120113,
      "learning_rate": 1.0103617798191872e-06,
      "loss": 0.7931906,
      "num_input_tokens_seen": 120747160,
      "step": 5612,
      "time_per_iteration": 2.613525629043579
    },
    {
      "auxiliary_loss_clip": 0.01278128,
      "auxiliary_loss_mlp": 0.01023834,
      "balance_loss_clip": 1.05003786,
      "balance_loss_mlp": 1.01656246,
      "epoch": 0.6749233451572175,
      "flos": 15195026407680.0,
      "grad_norm": 2.898200117055019,
      "language_loss": 0.82500875,
      "learning_rate": 1.0096849316542217e-06,
      "loss": 0.84802842,
      "num_input_tokens_seen": 120763710,
      "step": 5613,
      "time_per_iteration": 2.6423797607421875
    },
    {
      "auxiliary_loss_clip": 0.01459834,
      "auxiliary_loss_mlp": 0.01024178,
      "balance_loss_clip": 1.03789258,
      "balance_loss_mlp": 1.01683736,
      "epoch": 0.6750435880478567,
      "flos": 26499489050880.0,
      "grad_norm": 2.0650612665310444,
      "language_loss": 0.753106,
      "learning_rate": 1.0090082337192643e-06,
      "loss": 0.77794605,
      "num_input_tokens_seen": 120783355,
      "step": 5614,
      "time_per_iteration": 2.9286766052246094
    },
    {
      "auxiliary_loss_clip": 0.01417499,
      "auxiliary_loss_mlp": 0.01022234,
      "balance_loss_clip": 1.03688312,
      "balance_loss_mlp": 1.01535594,
      "epoch": 0.6751638309384957,
      "flos": 23404313076480.0,
      "grad_norm": 2.133943121228115,
      "language_loss": 0.78412151,
      "learning_rate": 1.0083316861169705e-06,
      "loss": 0.80851883,
      "num_input_tokens_seen": 120802090,
      "step": 5615,
      "time_per_iteration": 3.498469114303589
    },
    {
      "auxiliary_loss_clip": 0.01332115,
      "auxiliary_loss_mlp": 0.01025963,
      "balance_loss_clip": 1.04570651,
      "balance_loss_mlp": 1.0181365,
      "epoch": 0.6752840738291348,
      "flos": 23441408847360.0,
      "grad_norm": 2.1151483113056484,
      "language_loss": 0.72089994,
      "learning_rate": 1.0076552889499713e-06,
      "loss": 0.74448073,
      "num_input_tokens_seen": 120822855,
      "step": 5616,
      "time_per_iteration": 2.7501678466796875
    },
    {
      "auxiliary_loss_clip": 0.01225192,
      "auxiliary_loss_mlp": 0.01023255,
      "balance_loss_clip": 1.05169594,
      "balance_loss_mlp": 1.01650739,
      "epoch": 0.675404316719774,
      "flos": 30335826257280.0,
      "grad_norm": 2.186101669168665,
      "language_loss": 0.73795855,
      "learning_rate": 1.006979042320876e-06,
      "loss": 0.76044297,
      "num_input_tokens_seen": 120843070,
      "step": 5617,
      "time_per_iteration": 2.6787331104278564
    },
    {
      "auxiliary_loss_clip": 0.01275787,
      "auxiliary_loss_mlp": 0.01024987,
      "balance_loss_clip": 1.04541945,
      "balance_loss_mlp": 1.01772761,
      "epoch": 0.675524559610413,
      "flos": 23622613983360.0,
      "grad_norm": 2.509908302990683,
      "language_loss": 0.63004053,
      "learning_rate": 1.0063029463322702e-06,
      "loss": 0.65304828,
      "num_input_tokens_seen": 120863345,
      "step": 5618,
      "time_per_iteration": 2.682905912399292
    },
    {
      "auxiliary_loss_clip": 0.01369078,
      "auxiliary_loss_mlp": 0.0256628,
      "balance_loss_clip": 1.04132068,
      "balance_loss_mlp": 0.99996185,
      "epoch": 0.6756448025010521,
      "flos": 21248631279360.0,
      "grad_norm": 2.0714597204467617,
      "language_loss": 0.75825465,
      "learning_rate": 1.0056270010867164e-06,
      "loss": 0.79760826,
      "num_input_tokens_seen": 120880915,
      "step": 5619,
      "time_per_iteration": 2.796499490737915
    },
    {
      "auxiliary_loss_clip": 0.01277678,
      "auxiliary_loss_mlp": 0.01021467,
      "balance_loss_clip": 1.0435183,
      "balance_loss_mlp": 1.01470494,
      "epoch": 0.6757650453916912,
      "flos": 21646521210240.0,
      "grad_norm": 3.814683059950269,
      "language_loss": 0.78349447,
      "learning_rate": 1.004951206686758e-06,
      "loss": 0.80648589,
      "num_input_tokens_seen": 120899190,
      "step": 5620,
      "time_per_iteration": 2.6802000999450684
    },
    {
      "auxiliary_loss_clip": 0.01226477,
      "auxiliary_loss_mlp": 0.01027312,
      "balance_loss_clip": 1.04898965,
      "balance_loss_mlp": 1.02073753,
      "epoch": 0.6758852882823303,
      "flos": 21795658479360.0,
      "grad_norm": 2.912257298731784,
      "language_loss": 0.71543908,
      "learning_rate": 1.0042755632349087e-06,
      "loss": 0.73797697,
      "num_input_tokens_seen": 120916080,
      "step": 5621,
      "time_per_iteration": 4.4816811084747314
    },
    {
      "auxiliary_loss_clip": 0.01324768,
      "auxiliary_loss_mlp": 0.01028947,
      "balance_loss_clip": 1.0461812,
      "balance_loss_mlp": 1.02183652,
      "epoch": 0.6760055311729694,
      "flos": 27088783580160.0,
      "grad_norm": 2.2349884272639815,
      "language_loss": 0.6306833,
      "learning_rate": 1.0036000708336653e-06,
      "loss": 0.65422046,
      "num_input_tokens_seen": 120935210,
      "step": 5622,
      "time_per_iteration": 2.777637481689453
    },
    {
      "auxiliary_loss_clip": 0.01279661,
      "auxiliary_loss_mlp": 0.01028025,
      "balance_loss_clip": 1.04988158,
      "balance_loss_mlp": 1.0207864,
      "epoch": 0.6761257740636085,
      "flos": 17999792922240.0,
      "grad_norm": 1.9246670159472827,
      "language_loss": 0.79580134,
      "learning_rate": 1.0029247295854984e-06,
      "loss": 0.81887817,
      "num_input_tokens_seen": 120951830,
      "step": 5623,
      "time_per_iteration": 2.700270891189575
    },
    {
      "auxiliary_loss_clip": 0.01334032,
      "auxiliary_loss_mlp": 0.01027735,
      "balance_loss_clip": 1.04906058,
      "balance_loss_mlp": 1.02080023,
      "epoch": 0.6762460169542476,
      "flos": 15121912273920.0,
      "grad_norm": 1.9346962504688938,
      "language_loss": 0.71888411,
      "learning_rate": 1.0022495395928588e-06,
      "loss": 0.74250174,
      "num_input_tokens_seen": 120970310,
      "step": 5624,
      "time_per_iteration": 2.7219483852386475
    },
    {
      "auxiliary_loss_clip": 0.01065605,
      "auxiliary_loss_mlp": 0.01001422,
      "balance_loss_clip": 1.00949347,
      "balance_loss_mlp": 1.00043857,
      "epoch": 0.6763662598448866,
      "flos": 67886970030720.0,
      "grad_norm": 0.7983165056331405,
      "language_loss": 0.62319434,
      "learning_rate": 1.0015745009581697e-06,
      "loss": 0.64386463,
      "num_input_tokens_seen": 121031915,
      "step": 5625,
      "time_per_iteration": 4.08286452293396
    },
    {
      "auxiliary_loss_clip": 0.01224823,
      "auxiliary_loss_mlp": 0.01026451,
      "balance_loss_clip": 1.04964983,
      "balance_loss_mlp": 1.01943576,
      "epoch": 0.6764865027355258,
      "flos": 20631829910400.0,
      "grad_norm": 2.9605406573615287,
      "language_loss": 0.67123133,
      "learning_rate": 1.0008996137838343e-06,
      "loss": 0.69374406,
      "num_input_tokens_seen": 121050890,
      "step": 5626,
      "time_per_iteration": 2.6052463054656982
    },
    {
      "auxiliary_loss_clip": 0.0118384,
      "auxiliary_loss_mlp": 0.01022908,
      "balance_loss_clip": 1.0523386,
      "balance_loss_mlp": 1.01479602,
      "epoch": 0.6766067456261649,
      "flos": 21215809226880.0,
      "grad_norm": 2.2697094566510785,
      "language_loss": 0.79923022,
      "learning_rate": 1.000224878172234e-06,
      "loss": 0.82129771,
      "num_input_tokens_seen": 121070015,
      "step": 5627,
      "time_per_iteration": 2.7874865531921387
    },
    {
      "auxiliary_loss_clip": 0.01128138,
      "auxiliary_loss_mlp": 0.01027418,
      "balance_loss_clip": 1.04912496,
      "balance_loss_mlp": 1.02066469,
      "epoch": 0.6767269885168039,
      "flos": 19938251220480.0,
      "grad_norm": 2.189889270254064,
      "language_loss": 0.72885346,
      "learning_rate": 9.99550294225724e-07,
      "loss": 0.75040901,
      "num_input_tokens_seen": 121089170,
      "step": 5628,
      "time_per_iteration": 2.679964065551758
    },
    {
      "auxiliary_loss_clip": 0.01383823,
      "auxiliary_loss_mlp": 0.0103051,
      "balance_loss_clip": 1.04408622,
      "balance_loss_mlp": 1.02337551,
      "epoch": 0.6768472314074431,
      "flos": 20814076540800.0,
      "grad_norm": 1.8561657310003972,
      "language_loss": 0.72422582,
      "learning_rate": 9.988758620466402e-07,
      "loss": 0.74836922,
      "num_input_tokens_seen": 121108040,
      "step": 5629,
      "time_per_iteration": 3.690739393234253
    },
    {
      "auxiliary_loss_clip": 0.01328637,
      "auxiliary_loss_mlp": 0.01023201,
      "balance_loss_clip": 1.04321611,
      "balance_loss_mlp": 1.01676643,
      "epoch": 0.6769674742980821,
      "flos": 23186012169600.0,
      "grad_norm": 1.8599517955722473,
      "language_loss": 0.76334089,
      "learning_rate": 9.982015817372917e-07,
      "loss": 0.78685933,
      "num_input_tokens_seen": 121128480,
      "step": 5630,
      "time_per_iteration": 2.914457082748413
    },
    {
      "auxiliary_loss_clip": 0.01375633,
      "auxiliary_loss_mlp": 0.01026304,
      "balance_loss_clip": 1.04180789,
      "balance_loss_mlp": 1.01882696,
      "epoch": 0.6770877171887212,
      "flos": 24242934885120.0,
      "grad_norm": 3.5349904504390555,
      "language_loss": 0.81821579,
      "learning_rate": 9.975274533999657e-07,
      "loss": 0.84223515,
      "num_input_tokens_seen": 121148010,
      "step": 5631,
      "time_per_iteration": 2.9193122386932373
    },
    {
      "auxiliary_loss_clip": 0.01176885,
      "auxiliary_loss_mlp": 0.01028246,
      "balance_loss_clip": 1.04884768,
      "balance_loss_mlp": 1.02013934,
      "epoch": 0.6772079600793603,
      "flos": 18141567903360.0,
      "grad_norm": 3.35708127981289,
      "language_loss": 0.84093606,
      "learning_rate": 9.96853477136929e-07,
      "loss": 0.86298728,
      "num_input_tokens_seen": 121162755,
      "step": 5632,
      "time_per_iteration": 2.672593116760254
    },
    {
      "auxiliary_loss_clip": 0.01317071,
      "auxiliary_loss_mlp": 0.01026348,
      "balance_loss_clip": 1.04186964,
      "balance_loss_mlp": 1.01933861,
      "epoch": 0.6773282029699994,
      "flos": 22452069571200.0,
      "grad_norm": 11.112346455058685,
      "language_loss": 0.75633907,
      "learning_rate": 9.96179653050422e-07,
      "loss": 0.77977324,
      "num_input_tokens_seen": 121182915,
      "step": 5633,
      "time_per_iteration": 2.721611976623535
    },
    {
      "auxiliary_loss_clip": 0.01323726,
      "auxiliary_loss_mlp": 0.01032415,
      "balance_loss_clip": 1.04776001,
      "balance_loss_mlp": 1.02509546,
      "epoch": 0.6774484458606385,
      "flos": 18693730748160.0,
      "grad_norm": 2.0443028294634558,
      "language_loss": 0.74201095,
      "learning_rate": 9.955059812426635e-07,
      "loss": 0.76557231,
      "num_input_tokens_seen": 121200445,
      "step": 5634,
      "time_per_iteration": 2.74883770942688
    },
    {
      "auxiliary_loss_clip": 0.01178223,
      "auxiliary_loss_mlp": 0.01027675,
      "balance_loss_clip": 1.05222487,
      "balance_loss_mlp": 1.02070117,
      "epoch": 0.6775686887512776,
      "flos": 25994046821760.0,
      "grad_norm": 2.3131419210518436,
      "language_loss": 0.82775629,
      "learning_rate": 9.948324618158493e-07,
      "loss": 0.84981531,
      "num_input_tokens_seen": 121220785,
      "step": 5635,
      "time_per_iteration": 2.7032361030578613
    },
    {
      "auxiliary_loss_clip": 0.01228238,
      "auxiliary_loss_mlp": 0.01025846,
      "balance_loss_clip": 1.04805148,
      "balance_loss_mlp": 1.01843131,
      "epoch": 0.6776889316419167,
      "flos": 13587987922560.0,
      "grad_norm": 2.2275301147943387,
      "language_loss": 0.77279955,
      "learning_rate": 9.941590948721502e-07,
      "loss": 0.79534042,
      "num_input_tokens_seen": 121237985,
      "step": 5636,
      "time_per_iteration": 2.668509006500244
    },
    {
      "auxiliary_loss_clip": 0.01269051,
      "auxiliary_loss_mlp": 0.01024663,
      "balance_loss_clip": 1.04831707,
      "balance_loss_mlp": 1.01824069,
      "epoch": 0.6778091745325557,
      "flos": 27601121220480.0,
      "grad_norm": 1.84171945740778,
      "language_loss": 0.76568866,
      "learning_rate": 9.934858805137188e-07,
      "loss": 0.78862584,
      "num_input_tokens_seen": 121258635,
      "step": 5637,
      "time_per_iteration": 2.705293893814087
    },
    {
      "auxiliary_loss_clip": 0.01220406,
      "auxiliary_loss_mlp": 0.01023388,
      "balance_loss_clip": 1.04871011,
      "balance_loss_mlp": 1.01654518,
      "epoch": 0.6779294174231949,
      "flos": 18734058743040.0,
      "grad_norm": 1.610733008742841,
      "language_loss": 0.80667698,
      "learning_rate": 9.92812818842677e-07,
      "loss": 0.82911497,
      "num_input_tokens_seen": 121277810,
      "step": 5638,
      "time_per_iteration": 2.633744239807129
    },
    {
      "auxiliary_loss_clip": 0.01228849,
      "auxiliary_loss_mlp": 0.01024362,
      "balance_loss_clip": 1.05001235,
      "balance_loss_mlp": 1.0171771,
      "epoch": 0.678049660313834,
      "flos": 45873797765760.0,
      "grad_norm": 1.9040382446368882,
      "language_loss": 0.64129335,
      "learning_rate": 9.921399099611306e-07,
      "loss": 0.66382551,
      "num_input_tokens_seen": 121298975,
      "step": 5639,
      "time_per_iteration": 2.857239246368408
    },
    {
      "auxiliary_loss_clip": 0.01274494,
      "auxiliary_loss_mlp": 0.01022252,
      "balance_loss_clip": 1.04551971,
      "balance_loss_mlp": 1.01572275,
      "epoch": 0.678169903204473,
      "flos": 19974556892160.0,
      "grad_norm": 1.964877109951649,
      "language_loss": 0.68859744,
      "learning_rate": 9.914671539711588e-07,
      "loss": 0.7115649,
      "num_input_tokens_seen": 121318495,
      "step": 5640,
      "time_per_iteration": 2.7055819034576416
    },
    {
      "auxiliary_loss_clip": 0.01523702,
      "auxiliary_loss_mlp": 0.02568455,
      "balance_loss_clip": 1.04381382,
      "balance_loss_mlp": 0.99991,
      "epoch": 0.6782901460951122,
      "flos": 21395613732480.0,
      "grad_norm": 2.6449471832844065,
      "language_loss": 0.78465658,
      "learning_rate": 9.90794550974817e-07,
      "loss": 0.82557821,
      "num_input_tokens_seen": 121338890,
      "step": 5641,
      "time_per_iteration": 2.949815511703491
    },
    {
      "auxiliary_loss_clip": 0.01324366,
      "auxiliary_loss_mlp": 0.01027033,
      "balance_loss_clip": 1.04647088,
      "balance_loss_mlp": 1.01979744,
      "epoch": 0.6784103889857512,
      "flos": 21434002392960.0,
      "grad_norm": 2.1310128730629283,
      "language_loss": 0.81480026,
      "learning_rate": 9.901221010741407e-07,
      "loss": 0.83831418,
      "num_input_tokens_seen": 121358210,
      "step": 5642,
      "time_per_iteration": 3.3170065879821777
    },
    {
      "auxiliary_loss_clip": 0.0122927,
      "auxiliary_loss_mlp": 0.01028583,
      "balance_loss_clip": 1.04843175,
      "balance_loss_mlp": 1.02156138,
      "epoch": 0.6785306318763903,
      "flos": 32671923091200.0,
      "grad_norm": 2.5321141212847595,
      "language_loss": 0.74980623,
      "learning_rate": 9.894498043711375e-07,
      "loss": 0.77238476,
      "num_input_tokens_seen": 121379955,
      "step": 5643,
      "time_per_iteration": 2.800082206726074
    },
    {
      "auxiliary_loss_clip": 0.01271461,
      "auxiliary_loss_mlp": 0.0102624,
      "balance_loss_clip": 1.04501092,
      "balance_loss_mlp": 1.0191294,
      "epoch": 0.6786508747670293,
      "flos": 25632139340160.0,
      "grad_norm": 2.6479819599431167,
      "language_loss": 0.69502234,
      "learning_rate": 9.887776609677962e-07,
      "loss": 0.71799934,
      "num_input_tokens_seen": 121401325,
      "step": 5644,
      "time_per_iteration": 2.7338764667510986
    },
    {
      "auxiliary_loss_clip": 0.01318775,
      "auxiliary_loss_mlp": 0.01028091,
      "balance_loss_clip": 1.04099846,
      "balance_loss_mlp": 1.0211556,
      "epoch": 0.6787711176576685,
      "flos": 19171881619200.0,
      "grad_norm": 2.1180849562701987,
      "language_loss": 0.72642684,
      "learning_rate": 9.88105670966079e-07,
      "loss": 0.74989545,
      "num_input_tokens_seen": 121419785,
      "step": 5645,
      "time_per_iteration": 2.7269327640533447
    },
    {
      "auxiliary_loss_clip": 0.01364505,
      "auxiliary_loss_mlp": 0.01027941,
      "balance_loss_clip": 1.04433072,
      "balance_loss_mlp": 1.02062774,
      "epoch": 0.6788913605483076,
      "flos": 13985159581440.0,
      "grad_norm": 2.6571087061554075,
      "language_loss": 0.79125094,
      "learning_rate": 9.874338344679283e-07,
      "loss": 0.81517541,
      "num_input_tokens_seen": 121435630,
      "step": 5646,
      "time_per_iteration": 2.727947235107422
    },
    {
      "auxiliary_loss_clip": 0.01173758,
      "auxiliary_loss_mlp": 0.01023384,
      "balance_loss_clip": 1.0502317,
      "balance_loss_mlp": 1.01692033,
      "epoch": 0.6790116034389466,
      "flos": 22017586659840.0,
      "grad_norm": 1.8081323050048514,
      "language_loss": 0.74201095,
      "learning_rate": 9.86762151575259e-07,
      "loss": 0.76398236,
      "num_input_tokens_seen": 121455625,
      "step": 5647,
      "time_per_iteration": 4.671745538711548
    },
    {
      "auxiliary_loss_clip": 0.01272844,
      "auxiliary_loss_mlp": 0.02561721,
      "balance_loss_clip": 1.04667222,
      "balance_loss_mlp": 0.99993587,
      "epoch": 0.6791318463295858,
      "flos": 20922454851840.0,
      "grad_norm": 1.7486181817525948,
      "language_loss": 0.80319262,
      "learning_rate": 9.860906223899651e-07,
      "loss": 0.84153819,
      "num_input_tokens_seen": 121475020,
      "step": 5648,
      "time_per_iteration": 2.794055700302124
    },
    {
      "auxiliary_loss_clip": 0.01282177,
      "auxiliary_loss_mlp": 0.01029555,
      "balance_loss_clip": 1.04848504,
      "balance_loss_mlp": 1.02245617,
      "epoch": 0.6792520892202248,
      "flos": 28512749422080.0,
      "grad_norm": 1.708280942138804,
      "language_loss": 0.75775123,
      "learning_rate": 9.854192470139184e-07,
      "loss": 0.78086853,
      "num_input_tokens_seen": 121496500,
      "step": 5649,
      "time_per_iteration": 2.7461514472961426
    },
    {
      "auxiliary_loss_clip": 0.01276285,
      "auxiliary_loss_mlp": 0.01022114,
      "balance_loss_clip": 1.05200863,
      "balance_loss_mlp": 1.01523542,
      "epoch": 0.6793723321108639,
      "flos": 20011904058240.0,
      "grad_norm": 2.6300929516984333,
      "language_loss": 0.71276814,
      "learning_rate": 9.847480255489645e-07,
      "loss": 0.73575211,
      "num_input_tokens_seen": 121515525,
      "step": 5650,
      "time_per_iteration": 3.5870859622955322
    },
    {
      "auxiliary_loss_clip": 0.01280027,
      "auxiliary_loss_mlp": 0.01027225,
      "balance_loss_clip": 1.04835701,
      "balance_loss_mlp": 1.02000403,
      "epoch": 0.6794925750015031,
      "flos": 26649488246400.0,
      "grad_norm": 1.8327972801879975,
      "language_loss": 0.69112813,
      "learning_rate": 9.840769580969295e-07,
      "loss": 0.71420068,
      "num_input_tokens_seen": 121535965,
      "step": 5651,
      "time_per_iteration": 2.724898338317871
    },
    {
      "auxiliary_loss_clip": 0.01220624,
      "auxiliary_loss_mlp": 0.01028632,
      "balance_loss_clip": 1.04625058,
      "balance_loss_mlp": 1.02137184,
      "epoch": 0.6796128178921421,
      "flos": 21580374314880.0,
      "grad_norm": 2.2736757935076013,
      "language_loss": 0.7986744,
      "learning_rate": 9.834060447596114e-07,
      "loss": 0.82116693,
      "num_input_tokens_seen": 121555235,
      "step": 5652,
      "time_per_iteration": 2.593921661376953
    },
    {
      "auxiliary_loss_clip": 0.01225336,
      "auxiliary_loss_mlp": 0.01029119,
      "balance_loss_clip": 1.04667902,
      "balance_loss_mlp": 1.02171004,
      "epoch": 0.6797330607827812,
      "flos": 22492002516480.0,
      "grad_norm": 2.0713616863105737,
      "language_loss": 0.78383338,
      "learning_rate": 9.827352856387868e-07,
      "loss": 0.80637795,
      "num_input_tokens_seen": 121574945,
      "step": 5653,
      "time_per_iteration": 2.7752583026885986
    },
    {
      "auxiliary_loss_clip": 0.01277022,
      "auxiliary_loss_mlp": 0.0099963,
      "balance_loss_clip": 1.012429,
      "balance_loss_mlp": 0.99873585,
      "epoch": 0.6798533036734203,
      "flos": 66306648286080.0,
      "grad_norm": 0.7781990583417483,
      "language_loss": 0.64292991,
      "learning_rate": 9.820646808362118e-07,
      "loss": 0.66569638,
      "num_input_tokens_seen": 121641200,
      "step": 5654,
      "time_per_iteration": 4.481834411621094
    },
    {
      "auxiliary_loss_clip": 0.01271924,
      "auxiliary_loss_mlp": 0.0102282,
      "balance_loss_clip": 1.04949772,
      "balance_loss_mlp": 1.01579881,
      "epoch": 0.6799735465640594,
      "flos": 16180163792640.0,
      "grad_norm": 2.0720199815409446,
      "language_loss": 0.72724295,
      "learning_rate": 9.813942304536154e-07,
      "loss": 0.75019032,
      "num_input_tokens_seen": 121659170,
      "step": 5655,
      "time_per_iteration": 2.808762788772583
    },
    {
      "auxiliary_loss_clip": 0.01277536,
      "auxiliary_loss_mlp": 0.01024071,
      "balance_loss_clip": 1.04740393,
      "balance_loss_mlp": 1.01691198,
      "epoch": 0.6800937894546984,
      "flos": 22125749489280.0,
      "grad_norm": 1.8692674359696668,
      "language_loss": 0.63610029,
      "learning_rate": 9.807239345927043e-07,
      "loss": 0.65911639,
      "num_input_tokens_seen": 121679180,
      "step": 5656,
      "time_per_iteration": 2.737557888031006
    },
    {
      "auxiliary_loss_clip": 0.0117944,
      "auxiliary_loss_mlp": 0.01024556,
      "balance_loss_clip": 1.04568911,
      "balance_loss_mlp": 1.01722813,
      "epoch": 0.6802140323453376,
      "flos": 31612953300480.0,
      "grad_norm": 2.333451316877656,
      "language_loss": 0.72252059,
      "learning_rate": 9.80053793355162e-07,
      "loss": 0.7445606,
      "num_input_tokens_seen": 121697875,
      "step": 5657,
      "time_per_iteration": 2.7595248222351074
    },
    {
      "auxiliary_loss_clip": 0.01379949,
      "auxiliary_loss_mlp": 0.01024768,
      "balance_loss_clip": 1.04919052,
      "balance_loss_mlp": 1.01768684,
      "epoch": 0.6803342752359767,
      "flos": 17712938908800.0,
      "grad_norm": 2.0869753363975905,
      "language_loss": 0.74988353,
      "learning_rate": 9.793838068426472e-07,
      "loss": 0.77393073,
      "num_input_tokens_seen": 121715570,
      "step": 5658,
      "time_per_iteration": 2.788546562194824
    },
    {
      "auxiliary_loss_clip": 0.01177691,
      "auxiliary_loss_mlp": 0.01029408,
      "balance_loss_clip": 1.05215764,
      "balance_loss_mlp": 1.02197492,
      "epoch": 0.6804545181266157,
      "flos": 11326800902400.0,
      "grad_norm": 2.029726971046754,
      "language_loss": 0.61124218,
      "learning_rate": 9.78713975156799e-07,
      "loss": 0.63331318,
      "num_input_tokens_seen": 121731435,
      "step": 5659,
      "time_per_iteration": 2.571359872817993
    },
    {
      "auxiliary_loss_clip": 0.01332293,
      "auxiliary_loss_mlp": 0.01026743,
      "balance_loss_clip": 1.05124009,
      "balance_loss_mlp": 1.01934659,
      "epoch": 0.6805747610172549,
      "flos": 29350976181120.0,
      "grad_norm": 1.7939877098624293,
      "language_loss": 0.71689057,
      "learning_rate": 9.780442983992273e-07,
      "loss": 0.74048102,
      "num_input_tokens_seen": 121749950,
      "step": 5660,
      "time_per_iteration": 2.75433611869812
    },
    {
      "auxiliary_loss_clip": 0.01270659,
      "auxiliary_loss_mlp": 0.0102819,
      "balance_loss_clip": 1.04613066,
      "balance_loss_mlp": 1.02109742,
      "epoch": 0.680695003907894,
      "flos": 37631868612480.0,
      "grad_norm": 1.934409976476223,
      "language_loss": 0.71808815,
      "learning_rate": 9.773747766715238e-07,
      "loss": 0.74107659,
      "num_input_tokens_seen": 121770770,
      "step": 5661,
      "time_per_iteration": 2.8214199542999268
    },
    {
      "auxiliary_loss_clip": 0.01274635,
      "auxiliary_loss_mlp": 0.01023822,
      "balance_loss_clip": 1.04496086,
      "balance_loss_mlp": 1.01665783,
      "epoch": 0.680815246798533,
      "flos": 22127365601280.0,
      "grad_norm": 1.6886575844507006,
      "language_loss": 0.80234563,
      "learning_rate": 9.767054100752536e-07,
      "loss": 0.82533026,
      "num_input_tokens_seen": 121790720,
      "step": 5662,
      "time_per_iteration": 2.612154960632324
    },
    {
      "auxiliary_loss_clip": 0.01226437,
      "auxiliary_loss_mlp": 0.0102362,
      "balance_loss_clip": 1.04737556,
      "balance_loss_mlp": 1.016271,
      "epoch": 0.6809354896891722,
      "flos": 17201822330880.0,
      "grad_norm": 2.3747885577257892,
      "language_loss": 0.81571734,
      "learning_rate": 9.760361987119584e-07,
      "loss": 0.83821785,
      "num_input_tokens_seen": 121808455,
      "step": 5663,
      "time_per_iteration": 2.676050901412964
    },
    {
      "auxiliary_loss_clip": 0.01272944,
      "auxiliary_loss_mlp": 0.01024654,
      "balance_loss_clip": 1.04907846,
      "balance_loss_mlp": 1.01740575,
      "epoch": 0.6810557325798112,
      "flos": 12458166554880.0,
      "grad_norm": 3.538853926142097,
      "language_loss": 0.67850333,
      "learning_rate": 9.753671426831592e-07,
      "loss": 0.70147932,
      "num_input_tokens_seen": 121824470,
      "step": 5664,
      "time_per_iteration": 2.608029842376709
    },
    {
      "auxiliary_loss_clip": 0.01220018,
      "auxiliary_loss_mlp": 0.01024268,
      "balance_loss_clip": 1.04518378,
      "balance_loss_mlp": 1.01690078,
      "epoch": 0.6811759754704503,
      "flos": 22156165330560.0,
      "grad_norm": 2.051530452164032,
      "language_loss": 0.79804707,
      "learning_rate": 9.746982420903483e-07,
      "loss": 0.82049,
      "num_input_tokens_seen": 121842665,
      "step": 5665,
      "time_per_iteration": 2.666740894317627
    },
    {
      "auxiliary_loss_clip": 0.01228484,
      "auxiliary_loss_mlp": 0.01022784,
      "balance_loss_clip": 1.05262399,
      "balance_loss_mlp": 1.01579571,
      "epoch": 0.6812962183610894,
      "flos": 17525377065600.0,
      "grad_norm": 2.0549577069544878,
      "language_loss": 0.74642146,
      "learning_rate": 9.740294970349993e-07,
      "loss": 0.76893413,
      "num_input_tokens_seen": 121859080,
      "step": 5666,
      "time_per_iteration": 2.5737645626068115
    },
    {
      "auxiliary_loss_clip": 0.01169732,
      "auxiliary_loss_mlp": 0.00999722,
      "balance_loss_clip": 1.0086422,
      "balance_loss_mlp": 0.99876803,
      "epoch": 0.6814164612517285,
      "flos": 60274480855680.0,
      "grad_norm": 0.9031821921517337,
      "language_loss": 0.60845721,
      "learning_rate": 9.733609076185594e-07,
      "loss": 0.63015163,
      "num_input_tokens_seen": 121915485,
      "step": 5667,
      "time_per_iteration": 3.115020751953125
    },
    {
      "auxiliary_loss_clip": 0.01226088,
      "auxiliary_loss_mlp": 0.0102764,
      "balance_loss_clip": 1.04991984,
      "balance_loss_mlp": 1.02066934,
      "epoch": 0.6815367041423676,
      "flos": 19317750750720.0,
      "grad_norm": 2.588583308675847,
      "language_loss": 0.84142435,
      "learning_rate": 9.72692473942455e-07,
      "loss": 0.86396158,
      "num_input_tokens_seen": 121932710,
      "step": 5668,
      "time_per_iteration": 2.6086370944976807
    },
    {
      "auxiliary_loss_clip": 0.01384823,
      "auxiliary_loss_mlp": 0.01029237,
      "balance_loss_clip": 1.05037212,
      "balance_loss_mlp": 1.02186394,
      "epoch": 0.6816569470330067,
      "flos": 22161696024960.0,
      "grad_norm": 1.735544058324277,
      "language_loss": 0.77616847,
      "learning_rate": 9.720241961080849e-07,
      "loss": 0.80030912,
      "num_input_tokens_seen": 121952025,
      "step": 5669,
      "time_per_iteration": 2.7292754650115967
    },
    {
      "auxiliary_loss_clip": 0.01176123,
      "auxiliary_loss_mlp": 0.01024447,
      "balance_loss_clip": 1.04936028,
      "balance_loss_mlp": 1.01748478,
      "epoch": 0.6817771899236458,
      "flos": 41463501137280.0,
      "grad_norm": 2.010274318268284,
      "language_loss": 0.72936106,
      "learning_rate": 9.713560742168259e-07,
      "loss": 0.75136673,
      "num_input_tokens_seen": 121974650,
      "step": 5670,
      "time_per_iteration": 2.7083382606506348
    },
    {
      "auxiliary_loss_clip": 0.01225128,
      "auxiliary_loss_mlp": 0.01025326,
      "balance_loss_clip": 1.04628086,
      "balance_loss_mlp": 1.01812017,
      "epoch": 0.6818974328142848,
      "flos": 21106138026240.0,
      "grad_norm": 4.246766151782642,
      "language_loss": 0.71378565,
      "learning_rate": 9.706881083700333e-07,
      "loss": 0.7362901,
      "num_input_tokens_seen": 121994335,
      "step": 5671,
      "time_per_iteration": 2.6749114990234375
    },
    {
      "auxiliary_loss_clip": 0.01425416,
      "auxiliary_loss_mlp": 0.01024422,
      "balance_loss_clip": 1.04872787,
      "balance_loss_mlp": 1.01716506,
      "epoch": 0.682017675704924,
      "flos": 20441897769600.0,
      "grad_norm": 2.2108689901389864,
      "language_loss": 0.82539642,
      "learning_rate": 9.700202986690357e-07,
      "loss": 0.84989482,
      "num_input_tokens_seen": 122012635,
      "step": 5672,
      "time_per_iteration": 2.8656537532806396
    },
    {
      "auxiliary_loss_clip": 0.01222395,
      "auxiliary_loss_mlp": 0.02568969,
      "balance_loss_clip": 1.04724967,
      "balance_loss_mlp": 0.99990785,
      "epoch": 0.682137918595563,
      "flos": 20044438801920.0,
      "grad_norm": 2.2904661231886148,
      "language_loss": 0.66750574,
      "learning_rate": 9.693526452151413e-07,
      "loss": 0.70541942,
      "num_input_tokens_seen": 122031685,
      "step": 5673,
      "time_per_iteration": 4.341189622879028
    },
    {
      "auxiliary_loss_clip": 0.01333591,
      "auxiliary_loss_mlp": 0.01027269,
      "balance_loss_clip": 1.04561198,
      "balance_loss_mlp": 1.01896048,
      "epoch": 0.6822581614862021,
      "flos": 31684559063040.0,
      "grad_norm": 1.8363575606245934,
      "language_loss": 0.75870782,
      "learning_rate": 9.686851481096305e-07,
      "loss": 0.78231645,
      "num_input_tokens_seen": 122052995,
      "step": 5674,
      "time_per_iteration": 2.7774531841278076
    },
    {
      "auxiliary_loss_clip": 0.01325364,
      "auxiliary_loss_mlp": 0.0102618,
      "balance_loss_clip": 1.04450834,
      "balance_loss_mlp": 1.01869321,
      "epoch": 0.6823784043768413,
      "flos": 23477570864640.0,
      "grad_norm": 2.2081260206024287,
      "language_loss": 0.71577311,
      "learning_rate": 9.68017807453762e-07,
      "loss": 0.73928851,
      "num_input_tokens_seen": 122071740,
      "step": 5675,
      "time_per_iteration": 2.7937445640563965
    },
    {
      "auxiliary_loss_clip": 0.01277966,
      "auxiliary_loss_mlp": 0.02565002,
      "balance_loss_clip": 1.0500989,
      "balance_loss_mlp": 0.99986863,
      "epoch": 0.6824986472674803,
      "flos": 14137134024960.0,
      "grad_norm": 1.7752346636873464,
      "language_loss": 0.73133993,
      "learning_rate": 9.673506233487721e-07,
      "loss": 0.76976967,
      "num_input_tokens_seen": 122089705,
      "step": 5676,
      "time_per_iteration": 3.5410282611846924
    },
    {
      "auxiliary_loss_clip": 0.01273209,
      "auxiliary_loss_mlp": 0.02560392,
      "balance_loss_clip": 1.0461396,
      "balance_loss_mlp": 0.99990594,
      "epoch": 0.6826188901581194,
      "flos": 21504997624320.0,
      "grad_norm": 5.035075896186554,
      "language_loss": 0.86107206,
      "learning_rate": 9.666835958958717e-07,
      "loss": 0.8994081,
      "num_input_tokens_seen": 122109025,
      "step": 5677,
      "time_per_iteration": 2.707176923751831
    },
    {
      "auxiliary_loss_clip": 0.01175205,
      "auxiliary_loss_mlp": 0.01024164,
      "balance_loss_clip": 1.05051231,
      "balance_loss_mlp": 1.01727343,
      "epoch": 0.6827391330487584,
      "flos": 20810126044800.0,
      "grad_norm": 2.1289830886127863,
      "language_loss": 0.80723929,
      "learning_rate": 9.660167251962484e-07,
      "loss": 0.82923299,
      "num_input_tokens_seen": 122127385,
      "step": 5678,
      "time_per_iteration": 2.616446018218994
    },
    {
      "auxiliary_loss_clip": 0.01326993,
      "auxiliary_loss_mlp": 0.01025687,
      "balance_loss_clip": 1.04552627,
      "balance_loss_mlp": 1.01872838,
      "epoch": 0.6828593759393976,
      "flos": 21688788539520.0,
      "grad_norm": 1.6678969836363005,
      "language_loss": 0.77877629,
      "learning_rate": 9.653500113510654e-07,
      "loss": 0.80230308,
      "num_input_tokens_seen": 122146500,
      "step": 5679,
      "time_per_iteration": 2.7183127403259277
    },
    {
      "auxiliary_loss_clip": 0.01273302,
      "auxiliary_loss_mlp": 0.01025706,
      "balance_loss_clip": 1.04613006,
      "balance_loss_mlp": 1.01830554,
      "epoch": 0.6829796188300367,
      "flos": 25337707557120.0,
      "grad_norm": 2.7828764047272916,
      "language_loss": 0.66872585,
      "learning_rate": 9.646834544614627e-07,
      "loss": 0.69171596,
      "num_input_tokens_seen": 122167000,
      "step": 5680,
      "time_per_iteration": 3.8200297355651855
    },
    {
      "auxiliary_loss_clip": 0.01270601,
      "auxiliary_loss_mlp": 0.01022447,
      "balance_loss_clip": 1.04685318,
      "balance_loss_mlp": 1.01564562,
      "epoch": 0.6830998617206757,
      "flos": 20704800389760.0,
      "grad_norm": 2.0113858275556336,
      "language_loss": 0.76183271,
      "learning_rate": 9.64017054628558e-07,
      "loss": 0.78476322,
      "num_input_tokens_seen": 122185825,
      "step": 5681,
      "time_per_iteration": 2.654698133468628
    },
    {
      "auxiliary_loss_clip": 0.01373429,
      "auxiliary_loss_mlp": 0.01027854,
      "balance_loss_clip": 1.04248595,
      "balance_loss_mlp": 1.0209372,
      "epoch": 0.6832201046113149,
      "flos": 21726638496000.0,
      "grad_norm": 2.3052481490638224,
      "language_loss": 0.78986406,
      "learning_rate": 9.63350811953441e-07,
      "loss": 0.81387687,
      "num_input_tokens_seen": 122206200,
      "step": 5682,
      "time_per_iteration": 2.7597062587738037
    },
    {
      "auxiliary_loss_clip": 0.01322535,
      "auxiliary_loss_mlp": 0.01023296,
      "balance_loss_clip": 1.04491901,
      "balance_loss_mlp": 1.01651609,
      "epoch": 0.6833403475019539,
      "flos": 19536554448000.0,
      "grad_norm": 2.205399457845408,
      "language_loss": 0.70806134,
      "learning_rate": 9.626847265371826e-07,
      "loss": 0.73151964,
      "num_input_tokens_seen": 122225520,
      "step": 5683,
      "time_per_iteration": 2.6784133911132812
    },
    {
      "auxiliary_loss_clip": 0.01268935,
      "auxiliary_loss_mlp": 0.01020399,
      "balance_loss_clip": 1.04487109,
      "balance_loss_mlp": 1.01352644,
      "epoch": 0.683460590392593,
      "flos": 19352153001600.0,
      "grad_norm": 2.383593930829791,
      "language_loss": 0.78942871,
      "learning_rate": 9.620187984808262e-07,
      "loss": 0.81232202,
      "num_input_tokens_seen": 122244320,
      "step": 5684,
      "time_per_iteration": 2.661470651626587
    },
    {
      "auxiliary_loss_clip": 0.01276095,
      "auxiliary_loss_mlp": 0.02565048,
      "balance_loss_clip": 1.04953218,
      "balance_loss_mlp": 0.9998818,
      "epoch": 0.6835808332832322,
      "flos": 23288500650240.0,
      "grad_norm": 2.083440043596199,
      "language_loss": 0.86249906,
      "learning_rate": 9.613530278853919e-07,
      "loss": 0.9009105,
      "num_input_tokens_seen": 122264295,
      "step": 5685,
      "time_per_iteration": 2.6823244094848633
    },
    {
      "auxiliary_loss_clip": 0.01228775,
      "auxiliary_loss_mlp": 0.01029829,
      "balance_loss_clip": 1.05117726,
      "balance_loss_mlp": 1.0223664,
      "epoch": 0.6837010761738712,
      "flos": 21653416621440.0,
      "grad_norm": 2.3208347545979264,
      "language_loss": 0.74395192,
      "learning_rate": 9.60687414851879e-07,
      "loss": 0.76653796,
      "num_input_tokens_seen": 122285300,
      "step": 5686,
      "time_per_iteration": 2.736229658126831
    },
    {
      "auxiliary_loss_clip": 0.01176537,
      "auxiliary_loss_mlp": 0.01029722,
      "balance_loss_clip": 1.04839945,
      "balance_loss_mlp": 1.02181292,
      "epoch": 0.6838213190645103,
      "flos": 17566387418880.0,
      "grad_norm": 2.8525911160431114,
      "language_loss": 0.78008795,
      "learning_rate": 9.600219594812575e-07,
      "loss": 0.80215055,
      "num_input_tokens_seen": 122303240,
      "step": 5687,
      "time_per_iteration": 2.6897788047790527
    },
    {
      "auxiliary_loss_clip": 0.0117279,
      "auxiliary_loss_mlp": 0.0102467,
      "balance_loss_clip": 1.04910231,
      "balance_loss_mlp": 1.01783633,
      "epoch": 0.6839415619551494,
      "flos": 23112538899840.0,
      "grad_norm": 2.940916159499002,
      "language_loss": 0.72783267,
      "learning_rate": 9.593566618744786e-07,
      "loss": 0.74980724,
      "num_input_tokens_seen": 122323390,
      "step": 5688,
      "time_per_iteration": 2.5747857093811035
    },
    {
      "auxiliary_loss_clip": 0.0117699,
      "auxiliary_loss_mlp": 0.01025186,
      "balance_loss_clip": 1.05086064,
      "balance_loss_mlp": 1.01790786,
      "epoch": 0.6840618048457885,
      "flos": 22127868391680.0,
      "grad_norm": 1.9213602091676687,
      "language_loss": 0.74121422,
      "learning_rate": 9.58691522132466e-07,
      "loss": 0.76323599,
      "num_input_tokens_seen": 122342200,
      "step": 5689,
      "time_per_iteration": 2.5558040142059326
    },
    {
      "auxiliary_loss_clip": 0.01275476,
      "auxiliary_loss_mlp": 0.01026663,
      "balance_loss_clip": 1.0491178,
      "balance_loss_mlp": 1.01917052,
      "epoch": 0.6841820477364275,
      "flos": 22015898720640.0,
      "grad_norm": 2.0714248252195535,
      "language_loss": 0.850766,
      "learning_rate": 9.58026540356123e-07,
      "loss": 0.8737874,
      "num_input_tokens_seen": 122360465,
      "step": 5690,
      "time_per_iteration": 2.678323268890381
    },
    {
      "auxiliary_loss_clip": 0.01229198,
      "auxiliary_loss_mlp": 0.01029029,
      "balance_loss_clip": 1.04907966,
      "balance_loss_mlp": 1.02096474,
      "epoch": 0.6843022906270667,
      "flos": 24900531125760.0,
      "grad_norm": 1.5680000359587742,
      "language_loss": 0.86724508,
      "learning_rate": 9.573617166463246e-07,
      "loss": 0.88982737,
      "num_input_tokens_seen": 122381680,
      "step": 5691,
      "time_per_iteration": 2.6601693630218506
    },
    {
      "auxiliary_loss_clip": 0.01275506,
      "auxiliary_loss_mlp": 0.01025735,
      "balance_loss_clip": 1.0446105,
      "balance_loss_mlp": 1.01847529,
      "epoch": 0.6844225335177058,
      "flos": 19969924037760.0,
      "grad_norm": 1.9940667182706293,
      "language_loss": 0.60289884,
      "learning_rate": 9.56697051103924e-07,
      "loss": 0.62591124,
      "num_input_tokens_seen": 122399120,
      "step": 5692,
      "time_per_iteration": 2.6190919876098633
    },
    {
      "auxiliary_loss_clip": 0.01272387,
      "auxiliary_loss_mlp": 0.0102445,
      "balance_loss_clip": 1.04605651,
      "balance_loss_mlp": 1.01813793,
      "epoch": 0.6845427764083448,
      "flos": 25883334126720.0,
      "grad_norm": 2.1276339359785696,
      "language_loss": 0.81294072,
      "learning_rate": 9.560325438297522e-07,
      "loss": 0.83590907,
      "num_input_tokens_seen": 122417430,
      "step": 5693,
      "time_per_iteration": 2.7365951538085938
    },
    {
      "auxiliary_loss_clip": 0.01281813,
      "auxiliary_loss_mlp": 0.01030866,
      "balance_loss_clip": 1.05187249,
      "balance_loss_mlp": 1.02365112,
      "epoch": 0.684663019298984,
      "flos": 18880143356160.0,
      "grad_norm": 2.0131217826635215,
      "language_loss": 0.86571622,
      "learning_rate": 9.553681949246127e-07,
      "loss": 0.88884306,
      "num_input_tokens_seen": 122435055,
      "step": 5694,
      "time_per_iteration": 2.6193203926086426
    },
    {
      "auxiliary_loss_clip": 0.01333971,
      "auxiliary_loss_mlp": 0.01025638,
      "balance_loss_clip": 1.04902065,
      "balance_loss_mlp": 1.01869404,
      "epoch": 0.684783262189623,
      "flos": 54193725302400.0,
      "grad_norm": 1.7837101792795929,
      "language_loss": 0.75307715,
      "learning_rate": 9.547040044892886e-07,
      "loss": 0.7766732,
      "num_input_tokens_seen": 122462570,
      "step": 5695,
      "time_per_iteration": 2.988497495651245
    },
    {
      "auxiliary_loss_clip": 0.01122943,
      "auxiliary_loss_mlp": 0.0100201,
      "balance_loss_clip": 1.01030159,
      "balance_loss_mlp": 1.00109768,
      "epoch": 0.6849035050802621,
      "flos": 63970264143360.0,
      "grad_norm": 0.8539573509703516,
      "language_loss": 0.60055006,
      "learning_rate": 9.540399726245354e-07,
      "loss": 0.62179959,
      "num_input_tokens_seen": 122519275,
      "step": 5696,
      "time_per_iteration": 3.068936347961426
    },
    {
      "auxiliary_loss_clip": 0.01273194,
      "auxiliary_loss_mlp": 0.01025494,
      "balance_loss_clip": 1.04526818,
      "balance_loss_mlp": 1.01784086,
      "epoch": 0.6850237479709013,
      "flos": 25224121774080.0,
      "grad_norm": 1.803645209650667,
      "language_loss": 0.69132644,
      "learning_rate": 9.533760994310859e-07,
      "loss": 0.71431327,
      "num_input_tokens_seen": 122539675,
      "step": 5697,
      "time_per_iteration": 2.7191457748413086
    },
    {
      "auxiliary_loss_clip": 0.01179696,
      "auxiliary_loss_mlp": 0.01027568,
      "balance_loss_clip": 1.0523752,
      "balance_loss_mlp": 1.02014709,
      "epoch": 0.6851439908615403,
      "flos": 19354128249600.0,
      "grad_norm": 1.9510714830900895,
      "language_loss": 0.75362349,
      "learning_rate": 9.527123850096508e-07,
      "loss": 0.77569616,
      "num_input_tokens_seen": 122558035,
      "step": 5698,
      "time_per_iteration": 2.544518232345581
    },
    {
      "auxiliary_loss_clip": 0.01128966,
      "auxiliary_loss_mlp": 0.01027735,
      "balance_loss_clip": 1.04858065,
      "balance_loss_mlp": 1.02077055,
      "epoch": 0.6852642337521794,
      "flos": 23182133500800.0,
      "grad_norm": 2.051018936946761,
      "language_loss": 0.72021872,
      "learning_rate": 9.520488294609142e-07,
      "loss": 0.74178571,
      "num_input_tokens_seen": 122576815,
      "step": 5699,
      "time_per_iteration": 3.7030558586120605
    },
    {
      "auxiliary_loss_clip": 0.0127279,
      "auxiliary_loss_mlp": 0.00998161,
      "balance_loss_clip": 1.01056111,
      "balance_loss_mlp": 0.99715942,
      "epoch": 0.6853844766428185,
      "flos": 62647206583680.0,
      "grad_norm": 0.7363687255039579,
      "language_loss": 0.53787374,
      "learning_rate": 9.513854328855368e-07,
      "loss": 0.56058329,
      "num_input_tokens_seen": 122634690,
      "step": 5700,
      "time_per_iteration": 4.061933755874634
    },
    {
      "auxiliary_loss_clip": 0.01172113,
      "auxiliary_loss_mlp": 0.01025175,
      "balance_loss_clip": 1.048733,
      "balance_loss_mlp": 1.01872921,
      "epoch": 0.6855047195334576,
      "flos": 23437242869760.0,
      "grad_norm": 4.559240214138372,
      "language_loss": 0.81215578,
      "learning_rate": 9.507221953841558e-07,
      "loss": 0.83412874,
      "num_input_tokens_seen": 122652320,
      "step": 5701,
      "time_per_iteration": 2.5772759914398193
    },
    {
      "auxiliary_loss_clip": 0.01226575,
      "auxiliary_loss_mlp": 0.01027547,
      "balance_loss_clip": 1.05020499,
      "balance_loss_mlp": 1.02023399,
      "epoch": 0.6856249624240967,
      "flos": 20664831530880.0,
      "grad_norm": 1.5580689648306694,
      "language_loss": 0.77861047,
      "learning_rate": 9.500591170573824e-07,
      "loss": 0.80115169,
      "num_input_tokens_seen": 122672340,
      "step": 5702,
      "time_per_iteration": 3.487673282623291
    },
    {
      "auxiliary_loss_clip": 0.01366323,
      "auxiliary_loss_mlp": 0.01023686,
      "balance_loss_clip": 1.04440951,
      "balance_loss_mlp": 1.01678085,
      "epoch": 0.6857452053147358,
      "flos": 17087302794240.0,
      "grad_norm": 1.9473316161229444,
      "language_loss": 0.74156773,
      "learning_rate": 9.493961980058078e-07,
      "loss": 0.76546782,
      "num_input_tokens_seen": 122689935,
      "step": 5703,
      "time_per_iteration": 2.691288948059082
    },
    {
      "auxiliary_loss_clip": 0.01412827,
      "auxiliary_loss_mlp": 0.01026087,
      "balance_loss_clip": 1.03941441,
      "balance_loss_mlp": 1.01934576,
      "epoch": 0.6858654482053749,
      "flos": 30847266057600.0,
      "grad_norm": 1.7984888693962273,
      "language_loss": 0.67054278,
      "learning_rate": 9.48733438329993e-07,
      "loss": 0.69493192,
      "num_input_tokens_seen": 122710200,
      "step": 5704,
      "time_per_iteration": 2.84193754196167
    },
    {
      "auxiliary_loss_clip": 0.01173491,
      "auxiliary_loss_mlp": 0.02565207,
      "balance_loss_clip": 1.05013168,
      "balance_loss_mlp": 0.9998731,
      "epoch": 0.6859856910960139,
      "flos": 28877314510080.0,
      "grad_norm": 1.6352658294007683,
      "language_loss": 0.74609184,
      "learning_rate": 9.480708381304807e-07,
      "loss": 0.78347886,
      "num_input_tokens_seen": 122731495,
      "step": 5705,
      "time_per_iteration": 2.6821491718292236
    },
    {
      "auxiliary_loss_clip": 0.01366056,
      "auxiliary_loss_mlp": 0.01032536,
      "balance_loss_clip": 1.04624176,
      "balance_loss_mlp": 1.02481771,
      "epoch": 0.6861059339866531,
      "flos": 19354523299200.0,
      "grad_norm": 2.105814549809959,
      "language_loss": 0.83511019,
      "learning_rate": 9.474083975077858e-07,
      "loss": 0.85909611,
      "num_input_tokens_seen": 122748620,
      "step": 5706,
      "time_per_iteration": 3.641396999359131
    },
    {
      "auxiliary_loss_clip": 0.01222068,
      "auxiliary_loss_mlp": 0.01030754,
      "balance_loss_clip": 1.0471164,
      "balance_loss_mlp": 1.02338135,
      "epoch": 0.6862261768772921,
      "flos": 22199976944640.0,
      "grad_norm": 2.792949141537391,
      "language_loss": 0.80726814,
      "learning_rate": 9.467461165623994e-07,
      "loss": 0.82979637,
      "num_input_tokens_seen": 122767670,
      "step": 5707,
      "time_per_iteration": 2.6267364025115967
    },
    {
      "auxiliary_loss_clip": 0.01225909,
      "auxiliary_loss_mlp": 0.01027481,
      "balance_loss_clip": 1.04682183,
      "balance_loss_mlp": 1.0207994,
      "epoch": 0.6863464197679312,
      "flos": 26285677344000.0,
      "grad_norm": 2.329033601036418,
      "language_loss": 0.79694533,
      "learning_rate": 9.46083995394791e-07,
      "loss": 0.81947923,
      "num_input_tokens_seen": 122785480,
      "step": 5708,
      "time_per_iteration": 2.687666416168213
    },
    {
      "auxiliary_loss_clip": 0.01225084,
      "auxiliary_loss_mlp": 0.02562198,
      "balance_loss_clip": 1.04887009,
      "balance_loss_mlp": 0.99988246,
      "epoch": 0.6864666626585703,
      "flos": 37815228564480.0,
      "grad_norm": 1.9641969576923344,
      "language_loss": 0.63212156,
      "learning_rate": 9.454220341054012e-07,
      "loss": 0.66999435,
      "num_input_tokens_seen": 122810265,
      "step": 5709,
      "time_per_iteration": 2.78898286819458
    },
    {
      "auxiliary_loss_clip": 0.01319669,
      "auxiliary_loss_mlp": 0.01023767,
      "balance_loss_clip": 1.04450226,
      "balance_loss_mlp": 1.01689744,
      "epoch": 0.6865869055492094,
      "flos": 19391152193280.0,
      "grad_norm": 5.227472392339175,
      "language_loss": 0.80553579,
      "learning_rate": 9.447602327946512e-07,
      "loss": 0.82897019,
      "num_input_tokens_seen": 122828905,
      "step": 5710,
      "time_per_iteration": 2.7231011390686035
    },
    {
      "auxiliary_loss_clip": 0.0127723,
      "auxiliary_loss_mlp": 0.01026585,
      "balance_loss_clip": 1.04530954,
      "balance_loss_mlp": 1.01899755,
      "epoch": 0.6867071484398485,
      "flos": 20375966355840.0,
      "grad_norm": 1.8317816552749706,
      "language_loss": 0.76462305,
      "learning_rate": 9.440985915629338e-07,
      "loss": 0.78766114,
      "num_input_tokens_seen": 122846235,
      "step": 5711,
      "time_per_iteration": 2.6180999279022217
    },
    {
      "auxiliary_loss_clip": 0.01175071,
      "auxiliary_loss_mlp": 0.0103209,
      "balance_loss_clip": 1.05173755,
      "balance_loss_mlp": 1.02522087,
      "epoch": 0.6868273913304875,
      "flos": 15889143801600.0,
      "grad_norm": 2.2354838289724506,
      "language_loss": 0.73151171,
      "learning_rate": 9.434371105106223e-07,
      "loss": 0.75358331,
      "num_input_tokens_seen": 122863835,
      "step": 5712,
      "time_per_iteration": 2.5748751163482666
    },
    {
      "auxiliary_loss_clip": 0.01325881,
      "auxiliary_loss_mlp": 0.01030242,
      "balance_loss_clip": 1.04608881,
      "balance_loss_mlp": 1.02268088,
      "epoch": 0.6869476342211267,
      "flos": 24462492768000.0,
      "grad_norm": 2.797659046559257,
      "language_loss": 0.70602477,
      "learning_rate": 9.427757897380602e-07,
      "loss": 0.72958601,
      "num_input_tokens_seen": 122883235,
      "step": 5713,
      "time_per_iteration": 2.7238268852233887
    },
    {
      "auxiliary_loss_clip": 0.01315089,
      "auxiliary_loss_mlp": 0.01028024,
      "balance_loss_clip": 1.04656994,
      "balance_loss_mlp": 1.02043056,
      "epoch": 0.6870678771117658,
      "flos": 18442571875200.0,
      "grad_norm": 2.4805680228557794,
      "language_loss": 0.85022008,
      "learning_rate": 9.421146293455695e-07,
      "loss": 0.87365127,
      "num_input_tokens_seen": 122898975,
      "step": 5714,
      "time_per_iteration": 2.6656785011291504
    },
    {
      "auxiliary_loss_clip": 0.01275414,
      "auxiliary_loss_mlp": 0.01027551,
      "balance_loss_clip": 1.04613006,
      "balance_loss_mlp": 1.02049136,
      "epoch": 0.6871881200024048,
      "flos": 22200371994240.0,
      "grad_norm": 1.9077841215875244,
      "language_loss": 0.68432462,
      "learning_rate": 9.414536294334489e-07,
      "loss": 0.70735425,
      "num_input_tokens_seen": 122918995,
      "step": 5715,
      "time_per_iteration": 2.69136905670166
    },
    {
      "auxiliary_loss_clip": 0.01277298,
      "auxiliary_loss_mlp": 0.01025327,
      "balance_loss_clip": 1.04454398,
      "balance_loss_mlp": 1.01808846,
      "epoch": 0.687308362893044,
      "flos": 22127724737280.0,
      "grad_norm": 2.0196915890800535,
      "language_loss": 0.6991452,
      "learning_rate": 9.407927901019708e-07,
      "loss": 0.72217143,
      "num_input_tokens_seen": 122938125,
      "step": 5716,
      "time_per_iteration": 2.65014386177063
    },
    {
      "auxiliary_loss_clip": 0.01226173,
      "auxiliary_loss_mlp": 0.0102271,
      "balance_loss_clip": 1.04919529,
      "balance_loss_mlp": 1.01562572,
      "epoch": 0.687428605783683,
      "flos": 25040546340480.0,
      "grad_norm": 2.0653014308526454,
      "language_loss": 0.76750237,
      "learning_rate": 9.401321114513854e-07,
      "loss": 0.7899912,
      "num_input_tokens_seen": 122957020,
      "step": 5717,
      "time_per_iteration": 2.604227304458618
    },
    {
      "auxiliary_loss_clip": 0.01177054,
      "auxiliary_loss_mlp": 0.01024415,
      "balance_loss_clip": 1.05129433,
      "balance_loss_mlp": 1.01718211,
      "epoch": 0.6875488486743221,
      "flos": 23770063313280.0,
      "grad_norm": 2.4708513584558056,
      "language_loss": 0.75512773,
      "learning_rate": 9.394715935819155e-07,
      "loss": 0.77714241,
      "num_input_tokens_seen": 122977410,
      "step": 5718,
      "time_per_iteration": 2.6273937225341797
    },
    {
      "auxiliary_loss_clip": 0.01229285,
      "auxiliary_loss_mlp": 0.01028354,
      "balance_loss_clip": 1.04810572,
      "balance_loss_mlp": 1.02129936,
      "epoch": 0.6876690915649613,
      "flos": 25516937445120.0,
      "grad_norm": 2.1028429462361538,
      "language_loss": 0.62164557,
      "learning_rate": 9.388112365937608e-07,
      "loss": 0.6442219,
      "num_input_tokens_seen": 122996875,
      "step": 5719,
      "time_per_iteration": 2.6580827236175537
    },
    {
      "auxiliary_loss_clip": 0.01328323,
      "auxiliary_loss_mlp": 0.01029767,
      "balance_loss_clip": 1.04733849,
      "balance_loss_mlp": 1.02275181,
      "epoch": 0.6877893344556003,
      "flos": 19427996568960.0,
      "grad_norm": 2.1766364624415795,
      "language_loss": 0.82616538,
      "learning_rate": 9.381510405870985e-07,
      "loss": 0.84974623,
      "num_input_tokens_seen": 123015890,
      "step": 5720,
      "time_per_iteration": 2.7399609088897705
    },
    {
      "auxiliary_loss_clip": 0.01226628,
      "auxiliary_loss_mlp": 0.01032057,
      "balance_loss_clip": 1.04673696,
      "balance_loss_mlp": 1.02454662,
      "epoch": 0.6879095773462394,
      "flos": 18661303745280.0,
      "grad_norm": 2.6044894746204257,
      "language_loss": 0.77330565,
      "learning_rate": 9.374910056620791e-07,
      "loss": 0.79589248,
      "num_input_tokens_seen": 123034955,
      "step": 5721,
      "time_per_iteration": 2.673384666442871
    },
    {
      "auxiliary_loss_clip": 0.01224441,
      "auxiliary_loss_mlp": 0.01027222,
      "balance_loss_clip": 1.04847407,
      "balance_loss_mlp": 1.01949155,
      "epoch": 0.6880298202368785,
      "flos": 20883132437760.0,
      "grad_norm": 1.970407040502931,
      "language_loss": 0.81097031,
      "learning_rate": 9.368311319188293e-07,
      "loss": 0.83348691,
      "num_input_tokens_seen": 123052770,
      "step": 5722,
      "time_per_iteration": 2.5886852741241455
    },
    {
      "auxiliary_loss_clip": 0.01319887,
      "auxiliary_loss_mlp": 0.01030966,
      "balance_loss_clip": 1.04333854,
      "balance_loss_mlp": 1.0242157,
      "epoch": 0.6881500631275176,
      "flos": 30153292318080.0,
      "grad_norm": 1.7505810711663898,
      "language_loss": 0.79286361,
      "learning_rate": 9.361714194574515e-07,
      "loss": 0.81637216,
      "num_input_tokens_seen": 123075105,
      "step": 5723,
      "time_per_iteration": 2.8434510231018066
    },
    {
      "auxiliary_loss_clip": 0.01067686,
      "auxiliary_loss_mlp": 0.0100321,
      "balance_loss_clip": 1.0118407,
      "balance_loss_mlp": 1.00222683,
      "epoch": 0.6882703060181566,
      "flos": 66181537215360.0,
      "grad_norm": 0.7354427154682996,
      "language_loss": 0.58275026,
      "learning_rate": 9.355118683780228e-07,
      "loss": 0.60345924,
      "num_input_tokens_seen": 123145175,
      "step": 5724,
      "time_per_iteration": 4.211063861846924
    },
    {
      "auxiliary_loss_clip": 0.01174732,
      "auxiliary_loss_mlp": 0.01025767,
      "balance_loss_clip": 1.04820657,
      "balance_loss_mlp": 1.01842928,
      "epoch": 0.6883905489087958,
      "flos": 18214646123520.0,
      "grad_norm": 2.298206653585592,
      "language_loss": 0.79184395,
      "learning_rate": 9.348524787805987e-07,
      "loss": 0.81384891,
      "num_input_tokens_seen": 123160365,
      "step": 5725,
      "time_per_iteration": 2.5176753997802734
    },
    {
      "auxiliary_loss_clip": 0.01324894,
      "auxiliary_loss_mlp": 0.01023451,
      "balance_loss_clip": 1.04112267,
      "balance_loss_mlp": 1.01612234,
      "epoch": 0.6885107917994349,
      "flos": 14056262553600.0,
      "grad_norm": 2.387149131902642,
      "language_loss": 0.84804809,
      "learning_rate": 9.341932507652053e-07,
      "loss": 0.87153161,
      "num_input_tokens_seen": 123174855,
      "step": 5726,
      "time_per_iteration": 3.563671827316284
    },
    {
      "auxiliary_loss_clip": 0.01276044,
      "auxiliary_loss_mlp": 0.01026817,
      "balance_loss_clip": 1.04338956,
      "balance_loss_mlp": 1.01976919,
      "epoch": 0.6886310346900739,
      "flos": 28690722334080.0,
      "grad_norm": 1.7221766865107224,
      "language_loss": 0.78495193,
      "learning_rate": 9.335341844318489e-07,
      "loss": 0.80798054,
      "num_input_tokens_seen": 123194995,
      "step": 5727,
      "time_per_iteration": 2.6723580360412598
    },
    {
      "auxiliary_loss_clip": 0.01269657,
      "auxiliary_loss_mlp": 0.01023646,
      "balance_loss_clip": 1.04691947,
      "balance_loss_mlp": 1.01659465,
      "epoch": 0.6887512775807131,
      "flos": 24535319592960.0,
      "grad_norm": 1.8229815459948806,
      "language_loss": 0.73574519,
      "learning_rate": 9.328752798805091e-07,
      "loss": 0.75867826,
      "num_input_tokens_seen": 123213465,
      "step": 5728,
      "time_per_iteration": 3.570526123046875
    },
    {
      "auxiliary_loss_clip": 0.01226201,
      "auxiliary_loss_mlp": 0.0102581,
      "balance_loss_clip": 1.04951692,
      "balance_loss_mlp": 1.01834476,
      "epoch": 0.6888715204713521,
      "flos": 22414363269120.0,
      "grad_norm": 3.0406886795921513,
      "language_loss": 0.75993991,
      "learning_rate": 9.322165372111399e-07,
      "loss": 0.78245997,
      "num_input_tokens_seen": 123231610,
      "step": 5729,
      "time_per_iteration": 2.62888765335083
    },
    {
      "auxiliary_loss_clip": 0.01315241,
      "auxiliary_loss_mlp": 0.01025607,
      "balance_loss_clip": 1.04677022,
      "balance_loss_mlp": 1.01921439,
      "epoch": 0.6889917633619912,
      "flos": 22054323294720.0,
      "grad_norm": 2.2651621390276326,
      "language_loss": 0.76290661,
      "learning_rate": 9.315579565236747e-07,
      "loss": 0.78631508,
      "num_input_tokens_seen": 123250715,
      "step": 5730,
      "time_per_iteration": 2.7067768573760986
    },
    {
      "auxiliary_loss_clip": 0.01276477,
      "auxiliary_loss_mlp": 0.01025967,
      "balance_loss_clip": 1.05184245,
      "balance_loss_mlp": 1.01864123,
      "epoch": 0.6891120062526304,
      "flos": 23949724164480.0,
      "grad_norm": 1.7954511455153153,
      "language_loss": 0.74206942,
      "learning_rate": 9.308995379180162e-07,
      "loss": 0.76509386,
      "num_input_tokens_seen": 123270270,
      "step": 5731,
      "time_per_iteration": 2.66896390914917
    },
    {
      "auxiliary_loss_clip": 0.01122403,
      "auxiliary_loss_mlp": 0.01007456,
      "balance_loss_clip": 1.01085567,
      "balance_loss_mlp": 1.00649595,
      "epoch": 0.6892322491432694,
      "flos": 64117354337280.0,
      "grad_norm": 0.7386078759834017,
      "language_loss": 0.5948649,
      "learning_rate": 9.302412814940488e-07,
      "loss": 0.61616349,
      "num_input_tokens_seen": 123333045,
      "step": 5732,
      "time_per_iteration": 4.155320644378662
    },
    {
      "auxiliary_loss_clip": 0.01274517,
      "auxiliary_loss_mlp": 0.01026956,
      "balance_loss_clip": 1.04469228,
      "balance_loss_mlp": 1.01982772,
      "epoch": 0.6893524920339085,
      "flos": 23002436736000.0,
      "grad_norm": 2.1534541563585874,
      "language_loss": 0.7081399,
      "learning_rate": 9.295831873516276e-07,
      "loss": 0.73115456,
      "num_input_tokens_seen": 123352320,
      "step": 5733,
      "time_per_iteration": 2.6761391162872314
    },
    {
      "auxiliary_loss_clip": 0.01175246,
      "auxiliary_loss_mlp": 0.01029526,
      "balance_loss_clip": 1.05158114,
      "balance_loss_mlp": 1.02226353,
      "epoch": 0.6894727349245476,
      "flos": 21396260177280.0,
      "grad_norm": 1.8666401424307602,
      "language_loss": 0.76105094,
      "learning_rate": 9.289252555905873e-07,
      "loss": 0.7830987,
      "num_input_tokens_seen": 123372400,
      "step": 5734,
      "time_per_iteration": 2.5758371353149414
    },
    {
      "auxiliary_loss_clip": 0.01228029,
      "auxiliary_loss_mlp": 0.01028568,
      "balance_loss_clip": 1.05195653,
      "balance_loss_mlp": 1.02129936,
      "epoch": 0.6895929778151867,
      "flos": 19865316654720.0,
      "grad_norm": 2.4974901719934066,
      "language_loss": 0.75881422,
      "learning_rate": 9.282674863107334e-07,
      "loss": 0.78138018,
      "num_input_tokens_seen": 123390215,
      "step": 5735,
      "time_per_iteration": 2.6942427158355713
    },
    {
      "auxiliary_loss_clip": 0.01220484,
      "auxiliary_loss_mlp": 0.01024196,
      "balance_loss_clip": 1.04798841,
      "balance_loss_mlp": 1.01751757,
      "epoch": 0.6897132207058257,
      "flos": 18179166464640.0,
      "grad_norm": 2.2904406945105626,
      "language_loss": 0.76139116,
      "learning_rate": 9.276098796118488e-07,
      "loss": 0.78383791,
      "num_input_tokens_seen": 123406700,
      "step": 5736,
      "time_per_iteration": 2.6272382736206055
    },
    {
      "auxiliary_loss_clip": 0.01270393,
      "auxiliary_loss_mlp": 0.01024993,
      "balance_loss_clip": 1.04684424,
      "balance_loss_mlp": 1.01805472,
      "epoch": 0.6898334635964649,
      "flos": 32561641359360.0,
      "grad_norm": 2.2076571718658156,
      "language_loss": 0.65964866,
      "learning_rate": 9.269524355936938e-07,
      "loss": 0.68260247,
      "num_input_tokens_seen": 123429880,
      "step": 5737,
      "time_per_iteration": 2.7755043506622314
    },
    {
      "auxiliary_loss_clip": 0.01268469,
      "auxiliary_loss_mlp": 0.01028929,
      "balance_loss_clip": 1.04303598,
      "balance_loss_mlp": 1.02185738,
      "epoch": 0.689953706487104,
      "flos": 22819004956800.0,
      "grad_norm": 1.7776846696691546,
      "language_loss": 0.84649372,
      "learning_rate": 9.262951543560002e-07,
      "loss": 0.86946774,
      "num_input_tokens_seen": 123449105,
      "step": 5738,
      "time_per_iteration": 2.8290388584136963
    },
    {
      "auxiliary_loss_clip": 0.0128042,
      "auxiliary_loss_mlp": 0.01025811,
      "balance_loss_clip": 1.05380237,
      "balance_loss_mlp": 1.0179137,
      "epoch": 0.690073949377743,
      "flos": 18515362786560.0,
      "grad_norm": 2.475076819173484,
      "language_loss": 0.85925663,
      "learning_rate": 9.256380359984795e-07,
      "loss": 0.88231885,
      "num_input_tokens_seen": 123466215,
      "step": 5739,
      "time_per_iteration": 2.6272664070129395
    },
    {
      "auxiliary_loss_clip": 0.01380177,
      "auxiliary_loss_mlp": 0.01022572,
      "balance_loss_clip": 1.03934109,
      "balance_loss_mlp": 1.01549387,
      "epoch": 0.6901941922683821,
      "flos": 34857194716800.0,
      "grad_norm": 2.184700708538197,
      "language_loss": 0.74386454,
      "learning_rate": 9.249810806208139e-07,
      "loss": 0.767892,
      "num_input_tokens_seen": 123485480,
      "step": 5740,
      "time_per_iteration": 2.8792734146118164
    },
    {
      "auxiliary_loss_clip": 0.01370667,
      "auxiliary_loss_mlp": 0.02563464,
      "balance_loss_clip": 1.03880596,
      "balance_loss_mlp": 0.99984843,
      "epoch": 0.6903144351590212,
      "flos": 16253672976000.0,
      "grad_norm": 1.9202367721768152,
      "language_loss": 0.79793119,
      "learning_rate": 9.243242883226627e-07,
      "loss": 0.83727252,
      "num_input_tokens_seen": 123504575,
      "step": 5741,
      "time_per_iteration": 2.7153124809265137
    },
    {
      "auxiliary_loss_clip": 0.01227361,
      "auxiliary_loss_mlp": 0.01028746,
      "balance_loss_clip": 1.0450778,
      "balance_loss_mlp": 1.021474,
      "epoch": 0.6904346780496603,
      "flos": 28035137255040.0,
      "grad_norm": 2.0021021892468993,
      "language_loss": 0.69746673,
      "learning_rate": 9.236676592036628e-07,
      "loss": 0.7200278,
      "num_input_tokens_seen": 123524250,
      "step": 5742,
      "time_per_iteration": 2.714186191558838
    },
    {
      "auxiliary_loss_clip": 0.01269372,
      "auxiliary_loss_mlp": 0.01029554,
      "balance_loss_clip": 1.05026281,
      "balance_loss_mlp": 1.02250552,
      "epoch": 0.6905549209402994,
      "flos": 23624266008960.0,
      "grad_norm": 1.990467399179294,
      "language_loss": 0.73813164,
      "learning_rate": 9.230111933634228e-07,
      "loss": 0.76112092,
      "num_input_tokens_seen": 123545845,
      "step": 5743,
      "time_per_iteration": 2.7219197750091553
    },
    {
      "auxiliary_loss_clip": 0.01224368,
      "auxiliary_loss_mlp": 0.01036428,
      "balance_loss_clip": 1.04896605,
      "balance_loss_mlp": 1.02957654,
      "epoch": 0.6906751638309385,
      "flos": 23114945111040.0,
      "grad_norm": 1.5977270717397627,
      "language_loss": 0.80874312,
      "learning_rate": 9.223548909015288e-07,
      "loss": 0.83135104,
      "num_input_tokens_seen": 123567535,
      "step": 5744,
      "time_per_iteration": 2.6736762523651123
    },
    {
      "auxiliary_loss_clip": 0.01364655,
      "auxiliary_loss_mlp": 0.01021407,
      "balance_loss_clip": 1.04238665,
      "balance_loss_mlp": 1.01446295,
      "epoch": 0.6907954067215776,
      "flos": 27305468375040.0,
      "grad_norm": 1.9193185631664957,
      "language_loss": 0.72465229,
      "learning_rate": 9.216987519175407e-07,
      "loss": 0.74851286,
      "num_input_tokens_seen": 123587710,
      "step": 5745,
      "time_per_iteration": 2.832808494567871
    },
    {
      "auxiliary_loss_clip": 0.0121973,
      "auxiliary_loss_mlp": 0.01027233,
      "balance_loss_clip": 1.04877746,
      "balance_loss_mlp": 1.0199523,
      "epoch": 0.6909156496122166,
      "flos": 21689399070720.0,
      "grad_norm": 1.708292947728412,
      "language_loss": 0.68864667,
      "learning_rate": 9.210427765109942e-07,
      "loss": 0.71111625,
      "num_input_tokens_seen": 123607385,
      "step": 5746,
      "time_per_iteration": 2.6491191387176514
    },
    {
      "auxiliary_loss_clip": 0.01276814,
      "auxiliary_loss_mlp": 0.0103295,
      "balance_loss_clip": 1.046103,
      "balance_loss_mlp": 1.02447772,
      "epoch": 0.6910358925028558,
      "flos": 22561453463040.0,
      "grad_norm": 2.090376798382181,
      "language_loss": 0.81079096,
      "learning_rate": 9.20386964781402e-07,
      "loss": 0.83388853,
      "num_input_tokens_seen": 123625405,
      "step": 5747,
      "time_per_iteration": 2.6853320598602295
    },
    {
      "auxiliary_loss_clip": 0.01272165,
      "auxiliary_loss_mlp": 0.01024942,
      "balance_loss_clip": 1.04550433,
      "balance_loss_mlp": 1.01822162,
      "epoch": 0.6911561353934949,
      "flos": 22054107813120.0,
      "grad_norm": 1.8876306575425497,
      "language_loss": 0.84590769,
      "learning_rate": 9.197313168282472e-07,
      "loss": 0.86887872,
      "num_input_tokens_seen": 123642850,
      "step": 5748,
      "time_per_iteration": 2.666729688644409
    },
    {
      "auxiliary_loss_clip": 0.01220658,
      "auxiliary_loss_mlp": 0.01026427,
      "balance_loss_clip": 1.0452714,
      "balance_loss_mlp": 1.01902759,
      "epoch": 0.6912763782841339,
      "flos": 24206557386240.0,
      "grad_norm": 2.2784311512602797,
      "language_loss": 0.72617632,
      "learning_rate": 9.190758327509935e-07,
      "loss": 0.74864709,
      "num_input_tokens_seen": 123661595,
      "step": 5749,
      "time_per_iteration": 2.6466825008392334
    },
    {
      "auxiliary_loss_clip": 0.01188266,
      "auxiliary_loss_mlp": 0.02506638,
      "balance_loss_clip": 1.01379108,
      "balance_loss_mlp": 0.99972993,
      "epoch": 0.6913966211747731,
      "flos": 52329641091840.0,
      "grad_norm": 0.9319355066481709,
      "language_loss": 0.64456022,
      "learning_rate": 9.184205126490767e-07,
      "loss": 0.68150926,
      "num_input_tokens_seen": 123710490,
      "step": 5750,
      "time_per_iteration": 4.051300764083862
    },
    {
      "auxiliary_loss_clip": 0.01226656,
      "auxiliary_loss_mlp": 0.02507436,
      "balance_loss_clip": 1.01177478,
      "balance_loss_mlp": 0.99977225,
      "epoch": 0.6915168640654121,
      "flos": 66741274851840.0,
      "grad_norm": 1.1077367510146883,
      "language_loss": 0.59615487,
      "learning_rate": 9.177653566219075e-07,
      "loss": 0.63349581,
      "num_input_tokens_seen": 123765215,
      "step": 5751,
      "time_per_iteration": 4.09380316734314
    },
    {
      "auxiliary_loss_clip": 0.01327331,
      "auxiliary_loss_mlp": 0.01022502,
      "balance_loss_clip": 1.04347146,
      "balance_loss_mlp": 1.01584089,
      "epoch": 0.6916371069560512,
      "flos": 18296523175680.0,
      "grad_norm": 2.580845649908458,
      "language_loss": 0.76540005,
      "learning_rate": 9.171103647688744e-07,
      "loss": 0.78889835,
      "num_input_tokens_seen": 123783955,
      "step": 5752,
      "time_per_iteration": 2.715479612350464
    },
    {
      "auxiliary_loss_clip": 0.01464297,
      "auxiliary_loss_mlp": 0.01027145,
      "balance_loss_clip": 1.0383867,
      "balance_loss_mlp": 1.02074051,
      "epoch": 0.6917573498466904,
      "flos": 19645794685440.0,
      "grad_norm": 3.179459065167304,
      "language_loss": 0.6915217,
      "learning_rate": 9.164555371893367e-07,
      "loss": 0.71643615,
      "num_input_tokens_seen": 123803885,
      "step": 5753,
      "time_per_iteration": 3.0214009284973145
    },
    {
      "auxiliary_loss_clip": 0.01222808,
      "auxiliary_loss_mlp": 0.02563963,
      "balance_loss_clip": 1.04837573,
      "balance_loss_mlp": 0.99989271,
      "epoch": 0.6918775927373294,
      "flos": 14210319985920.0,
      "grad_norm": 2.0638989344165766,
      "language_loss": 0.75377417,
      "learning_rate": 9.158008739826333e-07,
      "loss": 0.79164183,
      "num_input_tokens_seen": 123821485,
      "step": 5754,
      "time_per_iteration": 3.611166477203369
    },
    {
      "auxiliary_loss_clip": 0.01269143,
      "auxiliary_loss_mlp": 0.01030817,
      "balance_loss_clip": 1.04699111,
      "balance_loss_mlp": 1.02374244,
      "epoch": 0.6919978356279685,
      "flos": 23985455218560.0,
      "grad_norm": 1.7688947053935131,
      "language_loss": 0.86761785,
      "learning_rate": 9.151463752480744e-07,
      "loss": 0.89061749,
      "num_input_tokens_seen": 123840215,
      "step": 5755,
      "time_per_iteration": 2.698784589767456
    },
    {
      "auxiliary_loss_clip": 0.01314809,
      "auxiliary_loss_mlp": 0.01026369,
      "balance_loss_clip": 1.04235172,
      "balance_loss_mlp": 1.01934803,
      "epoch": 0.6921180785186076,
      "flos": 23622937205760.0,
      "grad_norm": 1.8372690556565452,
      "language_loss": 0.80221313,
      "learning_rate": 9.144920410849493e-07,
      "loss": 0.82562494,
      "num_input_tokens_seen": 123861450,
      "step": 5756,
      "time_per_iteration": 2.6950252056121826
    },
    {
      "auxiliary_loss_clip": 0.01279356,
      "auxiliary_loss_mlp": 0.01023397,
      "balance_loss_clip": 1.04832864,
      "balance_loss_mlp": 1.01678705,
      "epoch": 0.6922383214092467,
      "flos": 21142623265920.0,
      "grad_norm": 1.7332417315961068,
      "language_loss": 0.80472088,
      "learning_rate": 9.138378715925176e-07,
      "loss": 0.82774836,
      "num_input_tokens_seen": 123880545,
      "step": 5757,
      "time_per_iteration": 2.6598691940307617
    },
    {
      "auxiliary_loss_clip": 0.01264155,
      "auxiliary_loss_mlp": 0.01026977,
      "balance_loss_clip": 1.04396677,
      "balance_loss_mlp": 1.01997364,
      "epoch": 0.6923585642998857,
      "flos": 21470667200640.0,
      "grad_norm": 1.8416223329557695,
      "language_loss": 0.80859488,
      "learning_rate": 9.131838668700167e-07,
      "loss": 0.83150625,
      "num_input_tokens_seen": 123900615,
      "step": 5758,
      "time_per_iteration": 4.138831377029419
    },
    {
      "auxiliary_loss_clip": 0.01328809,
      "auxiliary_loss_mlp": 0.01029787,
      "balance_loss_clip": 1.04461551,
      "balance_loss_mlp": 1.02253604,
      "epoch": 0.6924788071905249,
      "flos": 21105204272640.0,
      "grad_norm": 1.9793759597275544,
      "language_loss": 0.86679941,
      "learning_rate": 9.125300270166598e-07,
      "loss": 0.89038539,
      "num_input_tokens_seen": 123921220,
      "step": 5759,
      "time_per_iteration": 2.771765947341919
    },
    {
      "auxiliary_loss_clip": 0.01227297,
      "auxiliary_loss_mlp": 0.01024966,
      "balance_loss_clip": 1.04322529,
      "balance_loss_mlp": 1.01815879,
      "epoch": 0.692599050081164,
      "flos": 26250018117120.0,
      "grad_norm": 1.6944333321573566,
      "language_loss": 0.85789722,
      "learning_rate": 9.118763521316324e-07,
      "loss": 0.88041985,
      "num_input_tokens_seen": 123941795,
      "step": 5760,
      "time_per_iteration": 2.7644693851470947
    },
    {
      "auxiliary_loss_clip": 0.01173402,
      "auxiliary_loss_mlp": 0.02564862,
      "balance_loss_clip": 1.04792702,
      "balance_loss_mlp": 0.99986935,
      "epoch": 0.692719292971803,
      "flos": 20885215426560.0,
      "grad_norm": 2.036379971974148,
      "language_loss": 0.7601198,
      "learning_rate": 9.112228423140987e-07,
      "loss": 0.7975024,
      "num_input_tokens_seen": 123960715,
      "step": 5761,
      "time_per_iteration": 2.575529098510742
    },
    {
      "auxiliary_loss_clip": 0.01277473,
      "auxiliary_loss_mlp": 0.01025081,
      "balance_loss_clip": 1.04738939,
      "balance_loss_mlp": 1.01791096,
      "epoch": 0.6928395358624422,
      "flos": 25921938268800.0,
      "grad_norm": 2.5434413975282206,
      "language_loss": 0.86881232,
      "learning_rate": 9.105694976631932e-07,
      "loss": 0.89183789,
      "num_input_tokens_seen": 123978625,
      "step": 5762,
      "time_per_iteration": 2.817011833190918
    },
    {
      "auxiliary_loss_clip": 0.01225808,
      "auxiliary_loss_mlp": 0.01021775,
      "balance_loss_clip": 1.05078506,
      "balance_loss_mlp": 1.01472998,
      "epoch": 0.6929597787530812,
      "flos": 23586559706880.0,
      "grad_norm": 3.1176795627321634,
      "language_loss": 0.7261737,
      "learning_rate": 9.099163182780283e-07,
      "loss": 0.74864948,
      "num_input_tokens_seen": 123996780,
      "step": 5763,
      "time_per_iteration": 2.646911382675171
    },
    {
      "auxiliary_loss_clip": 0.01270213,
      "auxiliary_loss_mlp": 0.01024289,
      "balance_loss_clip": 1.04795706,
      "balance_loss_mlp": 1.01721704,
      "epoch": 0.6930800216437203,
      "flos": 18255656476800.0,
      "grad_norm": 9.561806247417692,
      "language_loss": 0.49623895,
      "learning_rate": 9.092633042576916e-07,
      "loss": 0.51918399,
      "num_input_tokens_seen": 124014045,
      "step": 5764,
      "time_per_iteration": 2.716426372528076
    },
    {
      "auxiliary_loss_clip": 0.01264792,
      "auxiliary_loss_mlp": 0.01023668,
      "balance_loss_clip": 1.04686868,
      "balance_loss_mlp": 1.01703668,
      "epoch": 0.6932002645343595,
      "flos": 29168621809920.0,
      "grad_norm": 1.8979134024347013,
      "language_loss": 0.56385636,
      "learning_rate": 9.086104557012446e-07,
      "loss": 0.58674103,
      "num_input_tokens_seen": 124034615,
      "step": 5765,
      "time_per_iteration": 2.7018938064575195
    },
    {
      "auxiliary_loss_clip": 0.01219691,
      "auxiliary_loss_mlp": 0.0102387,
      "balance_loss_clip": 1.04665208,
      "balance_loss_mlp": 1.01712263,
      "epoch": 0.6933205074249985,
      "flos": 23842746483840.0,
      "grad_norm": 2.2755272907052544,
      "language_loss": 0.65773249,
      "learning_rate": 9.079577727077239e-07,
      "loss": 0.68016815,
      "num_input_tokens_seen": 124053445,
      "step": 5766,
      "time_per_iteration": 2.6521592140197754
    },
    {
      "auxiliary_loss_clip": 0.01226515,
      "auxiliary_loss_mlp": 0.0102558,
      "balance_loss_clip": 1.05043721,
      "balance_loss_mlp": 1.01827824,
      "epoch": 0.6934407503156376,
      "flos": 24166696268160.0,
      "grad_norm": 7.454943879645916,
      "language_loss": 0.72225535,
      "learning_rate": 9.073052553761404e-07,
      "loss": 0.74477637,
      "num_input_tokens_seen": 124072810,
      "step": 5767,
      "time_per_iteration": 2.608692169189453
    },
    {
      "auxiliary_loss_clip": 0.01371753,
      "auxiliary_loss_mlp": 0.01028706,
      "balance_loss_clip": 1.04295695,
      "balance_loss_mlp": 1.02120757,
      "epoch": 0.6935609932062767,
      "flos": 20631327120000.0,
      "grad_norm": 1.7849821638605738,
      "language_loss": 0.78236842,
      "learning_rate": 9.066529038054805e-07,
      "loss": 0.806373,
      "num_input_tokens_seen": 124092875,
      "step": 5768,
      "time_per_iteration": 2.7513322830200195
    },
    {
      "auxiliary_loss_clip": 0.01273377,
      "auxiliary_loss_mlp": 0.01027389,
      "balance_loss_clip": 1.04641879,
      "balance_loss_mlp": 1.02075505,
      "epoch": 0.6936812360969158,
      "flos": 18254184019200.0,
      "grad_norm": 1.803670710177161,
      "language_loss": 0.74201149,
      "learning_rate": 9.060007180947071e-07,
      "loss": 0.76501918,
      "num_input_tokens_seen": 124110930,
      "step": 5769,
      "time_per_iteration": 2.619276523590088
    },
    {
      "auxiliary_loss_clip": 0.01275214,
      "auxiliary_loss_mlp": 0.0103144,
      "balance_loss_clip": 1.04236782,
      "balance_loss_mlp": 1.02349472,
      "epoch": 0.6938014789875548,
      "flos": 31317336368640.0,
      "grad_norm": 2.938524968408278,
      "language_loss": 0.72987187,
      "learning_rate": 9.053486983427534e-07,
      "loss": 0.75293839,
      "num_input_tokens_seen": 124132180,
      "step": 5770,
      "time_per_iteration": 2.818345785140991
    },
    {
      "auxiliary_loss_clip": 0.01277229,
      "auxiliary_loss_mlp": 0.01028719,
      "balance_loss_clip": 1.04521799,
      "balance_loss_mlp": 1.02187634,
      "epoch": 0.6939217218781939,
      "flos": 17528429721600.0,
      "grad_norm": 1.9155228687975974,
      "language_loss": 0.71127832,
      "learning_rate": 9.046968446485326e-07,
      "loss": 0.73433775,
      "num_input_tokens_seen": 124150585,
      "step": 5771,
      "time_per_iteration": 2.693904161453247
    },
    {
      "auxiliary_loss_clip": 0.0122539,
      "auxiliary_loss_mlp": 0.01026463,
      "balance_loss_clip": 1.04961479,
      "balance_loss_mlp": 1.01889324,
      "epoch": 0.6940419647688331,
      "flos": 18551776199040.0,
      "grad_norm": 2.1803383682428494,
      "language_loss": 0.7060945,
      "learning_rate": 9.040451571109295e-07,
      "loss": 0.72861302,
      "num_input_tokens_seen": 124166205,
      "step": 5772,
      "time_per_iteration": 2.557554006576538
    },
    {
      "auxiliary_loss_clip": 0.01239787,
      "auxiliary_loss_mlp": 0.00998778,
      "balance_loss_clip": 1.0205394,
      "balance_loss_mlp": 0.99771065,
      "epoch": 0.6941622076594721,
      "flos": 66926286829440.0,
      "grad_norm": 0.8281672444191933,
      "language_loss": 0.60397536,
      "learning_rate": 9.033936358288042e-07,
      "loss": 0.62636101,
      "num_input_tokens_seen": 124219940,
      "step": 5773,
      "time_per_iteration": 3.203559160232544
    },
    {
      "auxiliary_loss_clip": 0.01176076,
      "auxiliary_loss_mlp": 0.01024968,
      "balance_loss_clip": 1.04920089,
      "balance_loss_mlp": 1.01739264,
      "epoch": 0.6942824505501112,
      "flos": 26578062051840.0,
      "grad_norm": 2.3791559168784544,
      "language_loss": 0.82323587,
      "learning_rate": 9.027422809009937e-07,
      "loss": 0.84524632,
      "num_input_tokens_seen": 124239885,
      "step": 5774,
      "time_per_iteration": 2.6352412700653076
    },
    {
      "auxiliary_loss_clip": 0.01227001,
      "auxiliary_loss_mlp": 0.01023492,
      "balance_loss_clip": 1.04749107,
      "balance_loss_mlp": 1.01657462,
      "epoch": 0.6944026934407503,
      "flos": 21248308056960.0,
      "grad_norm": 1.749950572012988,
      "language_loss": 0.83178169,
      "learning_rate": 9.020910924263054e-07,
      "loss": 0.85428667,
      "num_input_tokens_seen": 124258410,
      "step": 5775,
      "time_per_iteration": 2.693902015686035
    },
    {
      "auxiliary_loss_clip": 0.01238048,
      "auxiliary_loss_mlp": 0.00999291,
      "balance_loss_clip": 1.02091932,
      "balance_loss_mlp": 0.99828339,
      "epoch": 0.6945229363313894,
      "flos": 70677191537280.0,
      "grad_norm": 0.8179131408682453,
      "language_loss": 0.58109522,
      "learning_rate": 9.014400705035261e-07,
      "loss": 0.6034686,
      "num_input_tokens_seen": 124315315,
      "step": 5776,
      "time_per_iteration": 4.260440826416016
    },
    {
      "auxiliary_loss_clip": 0.01176376,
      "auxiliary_loss_mlp": 0.01025675,
      "balance_loss_clip": 1.05203319,
      "balance_loss_mlp": 1.01865983,
      "epoch": 0.6946431792220285,
      "flos": 18952934267520.0,
      "grad_norm": 1.861198904612545,
      "language_loss": 0.7668494,
      "learning_rate": 9.00789215231414e-07,
      "loss": 0.78886986,
      "num_input_tokens_seen": 124333710,
      "step": 5777,
      "time_per_iteration": 3.626063823699951
    },
    {
      "auxiliary_loss_clip": 0.01327085,
      "auxiliary_loss_mlp": 0.02568952,
      "balance_loss_clip": 1.0421989,
      "balance_loss_mlp": 0.99988544,
      "epoch": 0.6947634221126676,
      "flos": 20338834671360.0,
      "grad_norm": 1.8225606397056584,
      "language_loss": 0.81981361,
      "learning_rate": 9.001385267087056e-07,
      "loss": 0.85877395,
      "num_input_tokens_seen": 124352855,
      "step": 5778,
      "time_per_iteration": 2.708718776702881
    },
    {
      "auxiliary_loss_clip": 0.01228678,
      "auxiliary_loss_mlp": 0.01024011,
      "balance_loss_clip": 1.04962897,
      "balance_loss_mlp": 1.0171957,
      "epoch": 0.6948836650033067,
      "flos": 21833723917440.0,
      "grad_norm": 1.672500466228321,
      "language_loss": 0.70606899,
      "learning_rate": 8.994880050341072e-07,
      "loss": 0.72859585,
      "num_input_tokens_seen": 124372960,
      "step": 5779,
      "time_per_iteration": 2.6294257640838623
    },
    {
      "auxiliary_loss_clip": 0.01273854,
      "auxiliary_loss_mlp": 0.01028608,
      "balance_loss_clip": 1.04737341,
      "balance_loss_mlp": 1.02169991,
      "epoch": 0.6950039078939457,
      "flos": 23657519024640.0,
      "grad_norm": 1.8927650005237129,
      "language_loss": 0.77766377,
      "learning_rate": 8.988376503063026e-07,
      "loss": 0.80068845,
      "num_input_tokens_seen": 124394220,
      "step": 5780,
      "time_per_iteration": 3.5834269523620605
    },
    {
      "auxiliary_loss_clip": 0.01383419,
      "auxiliary_loss_mlp": 0.01024342,
      "balance_loss_clip": 1.04663134,
      "balance_loss_mlp": 1.01679897,
      "epoch": 0.6951241507845849,
      "flos": 21792462168960.0,
      "grad_norm": 2.8118698563458273,
      "language_loss": 0.81509489,
      "learning_rate": 8.981874626239521e-07,
      "loss": 0.83917248,
      "num_input_tokens_seen": 124412795,
      "step": 5781,
      "time_per_iteration": 2.765078067779541
    },
    {
      "auxiliary_loss_clip": 0.01227229,
      "auxiliary_loss_mlp": 0.01029765,
      "balance_loss_clip": 1.05100536,
      "balance_loss_mlp": 1.02259445,
      "epoch": 0.695244393675224,
      "flos": 14647568244480.0,
      "grad_norm": 1.9915587309888167,
      "language_loss": 0.88834333,
      "learning_rate": 8.975374420856872e-07,
      "loss": 0.91091329,
      "num_input_tokens_seen": 124429690,
      "step": 5782,
      "time_per_iteration": 2.587526321411133
    },
    {
      "auxiliary_loss_clip": 0.01314925,
      "auxiliary_loss_mlp": 0.01022709,
      "balance_loss_clip": 1.04124033,
      "balance_loss_mlp": 1.01567602,
      "epoch": 0.695364636565863,
      "flos": 16873203778560.0,
      "grad_norm": 2.3017899439013396,
      "language_loss": 0.72313994,
      "learning_rate": 8.968875887901157e-07,
      "loss": 0.74651635,
      "num_input_tokens_seen": 124447070,
      "step": 5783,
      "time_per_iteration": 2.697949171066284
    },
    {
      "auxiliary_loss_clip": 0.01275088,
      "auxiliary_loss_mlp": 0.01025149,
      "balance_loss_clip": 1.04474878,
      "balance_loss_mlp": 1.01793945,
      "epoch": 0.6954848794565022,
      "flos": 19354523299200.0,
      "grad_norm": 2.1781912979946365,
      "language_loss": 0.62822652,
      "learning_rate": 8.9623790283582e-07,
      "loss": 0.6512289,
      "num_input_tokens_seen": 124464950,
      "step": 5784,
      "time_per_iteration": 3.5774013996124268
    },
    {
      "auxiliary_loss_clip": 0.01324705,
      "auxiliary_loss_mlp": 0.01029802,
      "balance_loss_clip": 1.04553533,
      "balance_loss_mlp": 1.02238691,
      "epoch": 0.6956051223471412,
      "flos": 18990209606400.0,
      "grad_norm": 2.3854345495506752,
      "language_loss": 0.76477683,
      "learning_rate": 8.955883843213561e-07,
      "loss": 0.78832191,
      "num_input_tokens_seen": 124483965,
      "step": 5785,
      "time_per_iteration": 2.7479870319366455
    },
    {
      "auxiliary_loss_clip": 0.01230392,
      "auxiliary_loss_mlp": 0.01030797,
      "balance_loss_clip": 1.04839182,
      "balance_loss_mlp": 1.02341461,
      "epoch": 0.6957253652377803,
      "flos": 16107229226880.0,
      "grad_norm": 1.7727296439869569,
      "language_loss": 0.86973792,
      "learning_rate": 8.949390333452569e-07,
      "loss": 0.89234984,
      "num_input_tokens_seen": 124501910,
      "step": 5786,
      "time_per_iteration": 2.7972569465637207
    },
    {
      "auxiliary_loss_clip": 0.01176032,
      "auxiliary_loss_mlp": 0.01031304,
      "balance_loss_clip": 1.05188107,
      "balance_loss_mlp": 1.02444363,
      "epoch": 0.6958456081284194,
      "flos": 29388646569600.0,
      "grad_norm": 2.569776164111555,
      "language_loss": 0.67827952,
      "learning_rate": 8.942898500060279e-07,
      "loss": 0.70035291,
      "num_input_tokens_seen": 124521625,
      "step": 5787,
      "time_per_iteration": 2.667105197906494
    },
    {
      "auxiliary_loss_clip": 0.01280622,
      "auxiliary_loss_mlp": 0.01028728,
      "balance_loss_clip": 1.04785264,
      "balance_loss_mlp": 1.0214386,
      "epoch": 0.6959658510190585,
      "flos": 25154850395520.0,
      "grad_norm": 2.6127303585525103,
      "language_loss": 0.721937,
      "learning_rate": 8.936408344021493e-07,
      "loss": 0.74503052,
      "num_input_tokens_seen": 124538540,
      "step": 5788,
      "time_per_iteration": 2.823032855987549
    },
    {
      "auxiliary_loss_clip": 0.01183195,
      "auxiliary_loss_mlp": 0.01030633,
      "balance_loss_clip": 1.05083334,
      "balance_loss_mlp": 1.02269983,
      "epoch": 0.6960860939096976,
      "flos": 42814388759040.0,
      "grad_norm": 2.4493277172156636,
      "language_loss": 0.71133149,
      "learning_rate": 8.929919866320765e-07,
      "loss": 0.73346972,
      "num_input_tokens_seen": 124559355,
      "step": 5789,
      "time_per_iteration": 2.853386640548706
    },
    {
      "auxiliary_loss_clip": 0.01332753,
      "auxiliary_loss_mlp": 0.02568753,
      "balance_loss_clip": 1.0477643,
      "balance_loss_mlp": 0.99986953,
      "epoch": 0.6962063368003367,
      "flos": 17566566986880.0,
      "grad_norm": 2.263201015166706,
      "language_loss": 0.81476259,
      "learning_rate": 8.923433067942385e-07,
      "loss": 0.85377765,
      "num_input_tokens_seen": 124577920,
      "step": 5790,
      "time_per_iteration": 2.811659336090088
    },
    {
      "auxiliary_loss_clip": 0.01232587,
      "auxiliary_loss_mlp": 0.01026697,
      "balance_loss_clip": 1.04846334,
      "balance_loss_mlp": 1.02024174,
      "epoch": 0.6963265796909758,
      "flos": 21251648021760.0,
      "grad_norm": 1.9813255822872426,
      "language_loss": 0.68725955,
      "learning_rate": 8.916947949870417e-07,
      "loss": 0.7098524,
      "num_input_tokens_seen": 124597585,
      "step": 5791,
      "time_per_iteration": 2.7138617038726807
    },
    {
      "auxiliary_loss_clip": 0.01123914,
      "auxiliary_loss_mlp": 0.01003355,
      "balance_loss_clip": 1.01201606,
      "balance_loss_mlp": 1.00238895,
      "epoch": 0.6964468225816148,
      "flos": 68828295801600.0,
      "grad_norm": 0.7356206913647154,
      "language_loss": 0.58054137,
      "learning_rate": 8.910464513088615e-07,
      "loss": 0.60181415,
      "num_input_tokens_seen": 124661625,
      "step": 5792,
      "time_per_iteration": 3.2817447185516357
    },
    {
      "auxiliary_loss_clip": 0.01270675,
      "auxiliary_loss_mlp": 0.0102809,
      "balance_loss_clip": 1.045784,
      "balance_loss_mlp": 1.02093482,
      "epoch": 0.696567065472254,
      "flos": 18950887192320.0,
      "grad_norm": 2.2370995792112995,
      "language_loss": 0.78759843,
      "learning_rate": 8.903982758580542e-07,
      "loss": 0.81058609,
      "num_input_tokens_seen": 124680565,
      "step": 5793,
      "time_per_iteration": 2.6674845218658447
    },
    {
      "auxiliary_loss_clip": 0.01279808,
      "auxiliary_loss_mlp": 0.01024297,
      "balance_loss_clip": 1.04929948,
      "balance_loss_mlp": 1.01688254,
      "epoch": 0.696687308362893,
      "flos": 22856675345280.0,
      "grad_norm": 2.1772907346935186,
      "language_loss": 0.80069852,
      "learning_rate": 8.897502687329457e-07,
      "loss": 0.82373959,
      "num_input_tokens_seen": 124700365,
      "step": 5794,
      "time_per_iteration": 2.6819489002227783
    },
    {
      "auxiliary_loss_clip": 0.01320613,
      "auxiliary_loss_mlp": 0.01024917,
      "balance_loss_clip": 1.04337978,
      "balance_loss_mlp": 1.01780045,
      "epoch": 0.6968075512535321,
      "flos": 24972926987520.0,
      "grad_norm": 1.9145156329558919,
      "language_loss": 0.79746091,
      "learning_rate": 8.891024300318382e-07,
      "loss": 0.82091624,
      "num_input_tokens_seen": 124718935,
      "step": 5795,
      "time_per_iteration": 2.7757420539855957
    },
    {
      "auxiliary_loss_clip": 0.01317049,
      "auxiliary_loss_mlp": 0.01022083,
      "balance_loss_clip": 1.04129076,
      "balance_loss_mlp": 1.01540995,
      "epoch": 0.6969277941441713,
      "flos": 21030438113280.0,
      "grad_norm": 2.7125023247527147,
      "language_loss": 0.7592212,
      "learning_rate": 8.884547598530103e-07,
      "loss": 0.78261256,
      "num_input_tokens_seen": 124739505,
      "step": 5796,
      "time_per_iteration": 2.7578392028808594
    },
    {
      "auxiliary_loss_clip": 0.01465185,
      "auxiliary_loss_mlp": 0.01032793,
      "balance_loss_clip": 1.03893018,
      "balance_loss_mlp": 1.02570033,
      "epoch": 0.6970480370348103,
      "flos": 21579404647680.0,
      "grad_norm": 1.985877231942512,
      "language_loss": 0.75336051,
      "learning_rate": 8.8780725829471e-07,
      "loss": 0.77834034,
      "num_input_tokens_seen": 124757410,
      "step": 5797,
      "time_per_iteration": 3.012160062789917
    },
    {
      "auxiliary_loss_clip": 0.01175217,
      "auxiliary_loss_mlp": 0.0103361,
      "balance_loss_clip": 1.04906058,
      "balance_loss_mlp": 1.02639198,
      "epoch": 0.6971682799254494,
      "flos": 22419175691520.0,
      "grad_norm": 4.9022555669031,
      "language_loss": 0.78180516,
      "learning_rate": 8.87159925455165e-07,
      "loss": 0.80389345,
      "num_input_tokens_seen": 124777240,
      "step": 5798,
      "time_per_iteration": 2.8051111698150635
    },
    {
      "auxiliary_loss_clip": 0.01324944,
      "auxiliary_loss_mlp": 0.01029597,
      "balance_loss_clip": 1.04757321,
      "balance_loss_mlp": 1.02265954,
      "epoch": 0.6972885228160886,
      "flos": 20005834659840.0,
      "grad_norm": 1.797610456467201,
      "language_loss": 0.73438358,
      "learning_rate": 8.865127614325738e-07,
      "loss": 0.75792897,
      "num_input_tokens_seen": 124795670,
      "step": 5799,
      "time_per_iteration": 2.7182815074920654
    },
    {
      "auxiliary_loss_clip": 0.01273132,
      "auxiliary_loss_mlp": 0.0103201,
      "balance_loss_clip": 1.04509115,
      "balance_loss_mlp": 1.02437782,
      "epoch": 0.6974087657067276,
      "flos": 37853437656960.0,
      "grad_norm": 2.1695924492897736,
      "language_loss": 0.67172706,
      "learning_rate": 8.85865766325113e-07,
      "loss": 0.69477844,
      "num_input_tokens_seen": 124819600,
      "step": 5800,
      "time_per_iteration": 2.767507791519165
    },
    {
      "auxiliary_loss_clip": 0.01273929,
      "auxiliary_loss_mlp": 0.01026212,
      "balance_loss_clip": 1.04632258,
      "balance_loss_mlp": 1.01925921,
      "epoch": 0.6975290085973667,
      "flos": 29489267543040.0,
      "grad_norm": 2.40810108641054,
      "language_loss": 0.72541326,
      "learning_rate": 8.852189402309287e-07,
      "loss": 0.74841464,
      "num_input_tokens_seen": 124838785,
      "step": 5801,
      "time_per_iteration": 2.7830047607421875
    },
    {
      "auxiliary_loss_clip": 0.01224848,
      "auxiliary_loss_mlp": 0.01029825,
      "balance_loss_clip": 1.05020213,
      "balance_loss_mlp": 1.02251124,
      "epoch": 0.6976492514880057,
      "flos": 12895630295040.0,
      "grad_norm": 9.111159253241118,
      "language_loss": 0.74489874,
      "learning_rate": 8.845722832481441e-07,
      "loss": 0.76744545,
      "num_input_tokens_seen": 124854215,
      "step": 5802,
      "time_per_iteration": 3.54976487159729
    },
    {
      "auxiliary_loss_clip": 0.01226334,
      "auxiliary_loss_mlp": 0.01027625,
      "balance_loss_clip": 1.05032349,
      "balance_loss_mlp": 1.02032983,
      "epoch": 0.6977694943786449,
      "flos": 24352929308160.0,
      "grad_norm": 2.3041092505730614,
      "language_loss": 0.77724707,
      "learning_rate": 8.83925795474858e-07,
      "loss": 0.79978669,
      "num_input_tokens_seen": 124874340,
      "step": 5803,
      "time_per_iteration": 3.53317928314209
    },
    {
      "auxiliary_loss_clip": 0.01318959,
      "auxiliary_loss_mlp": 0.01030499,
      "balance_loss_clip": 1.04699755,
      "balance_loss_mlp": 1.02247381,
      "epoch": 0.6978897372692839,
      "flos": 29898470257920.0,
      "grad_norm": 2.7402578972927008,
      "language_loss": 0.59354579,
      "learning_rate": 8.832794770091414e-07,
      "loss": 0.6170404,
      "num_input_tokens_seen": 124895175,
      "step": 5804,
      "time_per_iteration": 2.7984459400177
    },
    {
      "auxiliary_loss_clip": 0.01280988,
      "auxiliary_loss_mlp": 0.01025988,
      "balance_loss_clip": 1.04769421,
      "balance_loss_mlp": 1.01897883,
      "epoch": 0.698009980159923,
      "flos": 21761579450880.0,
      "grad_norm": 2.954649139979604,
      "language_loss": 0.82629001,
      "learning_rate": 8.826333279490401e-07,
      "loss": 0.84935975,
      "num_input_tokens_seen": 124915810,
      "step": 5805,
      "time_per_iteration": 2.7883288860321045
    },
    {
      "auxiliary_loss_clip": 0.01278159,
      "auxiliary_loss_mlp": 0.01024242,
      "balance_loss_clip": 1.05097437,
      "balance_loss_mlp": 1.01718795,
      "epoch": 0.6981302230505622,
      "flos": 19857164267520.0,
      "grad_norm": 2.5411725332919697,
      "language_loss": 0.67729867,
      "learning_rate": 8.819873483925748e-07,
      "loss": 0.70032269,
      "num_input_tokens_seen": 124932930,
      "step": 5806,
      "time_per_iteration": 3.462571859359741
    },
    {
      "auxiliary_loss_clip": 0.01330317,
      "auxiliary_loss_mlp": 0.02564827,
      "balance_loss_clip": 1.04842615,
      "balance_loss_mlp": 0.99989778,
      "epoch": 0.6982504659412012,
      "flos": 22198648141440.0,
      "grad_norm": 2.4327074179248878,
      "language_loss": 0.74841928,
      "learning_rate": 8.81341538437739e-07,
      "loss": 0.78737068,
      "num_input_tokens_seen": 124951220,
      "step": 5807,
      "time_per_iteration": 2.7744383811950684
    },
    {
      "auxiliary_loss_clip": 0.012776,
      "auxiliary_loss_mlp": 0.01023597,
      "balance_loss_clip": 1.04509497,
      "balance_loss_mlp": 1.01658416,
      "epoch": 0.6983707088318403,
      "flos": 35588479708800.0,
      "grad_norm": 1.7295664741324293,
      "language_loss": 0.67966688,
      "learning_rate": 8.80695898182503e-07,
      "loss": 0.70267886,
      "num_input_tokens_seen": 124972200,
      "step": 5808,
      "time_per_iteration": 2.7920539379119873
    },
    {
      "auxiliary_loss_clip": 0.01129753,
      "auxiliary_loss_mlp": 0.01003226,
      "balance_loss_clip": 1.02112424,
      "balance_loss_mlp": 1.00223053,
      "epoch": 0.6984909517224794,
      "flos": 65440052760960.0,
      "grad_norm": 0.9095171919704939,
      "language_loss": 0.65056229,
      "learning_rate": 8.800504277248093e-07,
      "loss": 0.67189205,
      "num_input_tokens_seen": 125036950,
      "step": 5809,
      "time_per_iteration": 3.228753089904785
    },
    {
      "auxiliary_loss_clip": 0.01327372,
      "auxiliary_loss_mlp": 0.02565784,
      "balance_loss_clip": 1.05268168,
      "balance_loss_mlp": 0.99991202,
      "epoch": 0.6986111946131185,
      "flos": 18546927863040.0,
      "grad_norm": 2.3173141626109297,
      "language_loss": 0.75180709,
      "learning_rate": 8.794051271625753e-07,
      "loss": 0.79073864,
      "num_input_tokens_seen": 125054585,
      "step": 5810,
      "time_per_iteration": 3.606156349182129
    },
    {
      "auxiliary_loss_clip": 0.01271719,
      "auxiliary_loss_mlp": 0.01027832,
      "balance_loss_clip": 1.04639482,
      "balance_loss_mlp": 1.02078927,
      "epoch": 0.6987314375037575,
      "flos": 23039173370880.0,
      "grad_norm": 3.0905194465099948,
      "language_loss": 0.83319986,
      "learning_rate": 8.787599965936925e-07,
      "loss": 0.85619533,
      "num_input_tokens_seen": 125075515,
      "step": 5811,
      "time_per_iteration": 2.680300235748291
    },
    {
      "auxiliary_loss_clip": 0.01318539,
      "auxiliary_loss_mlp": 0.01032604,
      "balance_loss_clip": 1.0461334,
      "balance_loss_mlp": 1.02550197,
      "epoch": 0.6988516803943967,
      "flos": 38400393029760.0,
      "grad_norm": 1.6725548823844794,
      "language_loss": 0.7232275,
      "learning_rate": 8.781150361160261e-07,
      "loss": 0.74673891,
      "num_input_tokens_seen": 125097425,
      "step": 5812,
      "time_per_iteration": 2.878025531768799
    },
    {
      "auxiliary_loss_clip": 0.0122796,
      "auxiliary_loss_mlp": 0.01033205,
      "balance_loss_clip": 1.04790401,
      "balance_loss_mlp": 1.02590942,
      "epoch": 0.6989719232850358,
      "flos": 24096993926400.0,
      "grad_norm": 2.5772296291025056,
      "language_loss": 0.73710942,
      "learning_rate": 8.774702458274181e-07,
      "loss": 0.75972104,
      "num_input_tokens_seen": 125117830,
      "step": 5813,
      "time_per_iteration": 2.727648973464966
    },
    {
      "auxiliary_loss_clip": 0.01230696,
      "auxiliary_loss_mlp": 0.01024859,
      "balance_loss_clip": 1.0521946,
      "balance_loss_mlp": 1.01742649,
      "epoch": 0.6990921661756748,
      "flos": 14866838818560.0,
      "grad_norm": 2.148759666128042,
      "language_loss": 0.71332014,
      "learning_rate": 8.768256258256799e-07,
      "loss": 0.73587561,
      "num_input_tokens_seen": 125134455,
      "step": 5814,
      "time_per_iteration": 2.6070117950439453
    },
    {
      "auxiliary_loss_clip": 0.01227696,
      "auxiliary_loss_mlp": 0.01026317,
      "balance_loss_clip": 1.04848146,
      "balance_loss_mlp": 1.01946282,
      "epoch": 0.699212409066314,
      "flos": 20193719725440.0,
      "grad_norm": 1.8452299685138527,
      "language_loss": 0.73955524,
      "learning_rate": 8.76181176208602e-07,
      "loss": 0.76209539,
      "num_input_tokens_seen": 125152555,
      "step": 5815,
      "time_per_iteration": 2.6505184173583984
    },
    {
      "auxiliary_loss_clip": 0.01373401,
      "auxiliary_loss_mlp": 0.01031422,
      "balance_loss_clip": 1.04304183,
      "balance_loss_mlp": 1.02309489,
      "epoch": 0.699332651956953,
      "flos": 19427888828160.0,
      "grad_norm": 1.8933104179098255,
      "language_loss": 0.73378682,
      "learning_rate": 8.755368970739461e-07,
      "loss": 0.75783503,
      "num_input_tokens_seen": 125171915,
      "step": 5816,
      "time_per_iteration": 2.7742233276367188
    },
    {
      "auxiliary_loss_clip": 0.01332288,
      "auxiliary_loss_mlp": 0.01025441,
      "balance_loss_clip": 1.0467931,
      "balance_loss_mlp": 1.01793039,
      "epoch": 0.6994528948475921,
      "flos": 16143714466560.0,
      "grad_norm": 2.734759639855676,
      "language_loss": 0.61735755,
      "learning_rate": 8.748927885194479e-07,
      "loss": 0.64093482,
      "num_input_tokens_seen": 125190220,
      "step": 5817,
      "time_per_iteration": 2.697021484375
    },
    {
      "auxiliary_loss_clip": 0.01215241,
      "auxiliary_loss_mlp": 0.01003869,
      "balance_loss_clip": 1.01029813,
      "balance_loss_mlp": 1.00289142,
      "epoch": 0.6995731377382313,
      "flos": 64952420699520.0,
      "grad_norm": 0.7865130506160781,
      "language_loss": 0.57381237,
      "learning_rate": 8.742488506428209e-07,
      "loss": 0.59600341,
      "num_input_tokens_seen": 125249310,
      "step": 5818,
      "time_per_iteration": 3.2413151264190674
    },
    {
      "auxiliary_loss_clip": 0.01277902,
      "auxiliary_loss_mlp": 0.02564764,
      "balance_loss_clip": 1.04789424,
      "balance_loss_mlp": 0.99985957,
      "epoch": 0.6996933806288703,
      "flos": 24900136076160.0,
      "grad_norm": 2.0636774477627196,
      "language_loss": 0.78692579,
      "learning_rate": 8.736050835417466e-07,
      "loss": 0.82535249,
      "num_input_tokens_seen": 125269350,
      "step": 5819,
      "time_per_iteration": 2.7432892322540283
    },
    {
      "auxiliary_loss_clip": 0.01230485,
      "auxiliary_loss_mlp": 0.01029431,
      "balance_loss_clip": 1.04976249,
      "balance_loss_mlp": 1.02270186,
      "epoch": 0.6998136235195094,
      "flos": 20777806782720.0,
      "grad_norm": 1.9379183069207455,
      "language_loss": 0.61553961,
      "learning_rate": 8.729614873138862e-07,
      "loss": 0.63813877,
      "num_input_tokens_seen": 125286985,
      "step": 5820,
      "time_per_iteration": 2.6084043979644775
    },
    {
      "auxiliary_loss_clip": 0.01377412,
      "auxiliary_loss_mlp": 0.01027017,
      "balance_loss_clip": 1.04702473,
      "balance_loss_mlp": 1.01967406,
      "epoch": 0.6999338664101485,
      "flos": 23733470332800.0,
      "grad_norm": 2.756183243150909,
      "language_loss": 0.77806294,
      "learning_rate": 8.723180620568716e-07,
      "loss": 0.80210727,
      "num_input_tokens_seen": 125306240,
      "step": 5821,
      "time_per_iteration": 2.7775988578796387
    },
    {
      "auxiliary_loss_clip": 0.01178605,
      "auxiliary_loss_mlp": 0.01026021,
      "balance_loss_clip": 1.04764771,
      "balance_loss_mlp": 1.01903486,
      "epoch": 0.7000541093007876,
      "flos": 19864598382720.0,
      "grad_norm": 1.8764377377043409,
      "language_loss": 0.84991288,
      "learning_rate": 8.716748078683116e-07,
      "loss": 0.87195909,
      "num_input_tokens_seen": 125323015,
      "step": 5822,
      "time_per_iteration": 2.645879030227661
    },
    {
      "auxiliary_loss_clip": 0.01466331,
      "auxiliary_loss_mlp": 0.01025146,
      "balance_loss_clip": 1.03963256,
      "balance_loss_mlp": 1.01802921,
      "epoch": 0.7001743521914267,
      "flos": 29679056029440.0,
      "grad_norm": 6.126701491483206,
      "language_loss": 0.68969589,
      "learning_rate": 8.710317248457855e-07,
      "loss": 0.71461058,
      "num_input_tokens_seen": 125342630,
      "step": 5823,
      "time_per_iteration": 2.9832546710968018
    },
    {
      "auxiliary_loss_clip": 0.01270291,
      "auxiliary_loss_mlp": 0.01026264,
      "balance_loss_clip": 1.04893577,
      "balance_loss_mlp": 1.0193646,
      "epoch": 0.7002945950820658,
      "flos": 27489762080640.0,
      "grad_norm": 2.03196424882431,
      "language_loss": 0.72150552,
      "learning_rate": 8.703888130868482e-07,
      "loss": 0.74447107,
      "num_input_tokens_seen": 125364480,
      "step": 5824,
      "time_per_iteration": 2.9289066791534424
    },
    {
      "auxiliary_loss_clip": 0.0132414,
      "auxiliary_loss_mlp": 0.0103274,
      "balance_loss_clip": 1.04652596,
      "balance_loss_mlp": 1.02515805,
      "epoch": 0.7004148379727049,
      "flos": 22158463800960.0,
      "grad_norm": 2.912047673445961,
      "language_loss": 0.82068974,
      "learning_rate": 8.697460726890307e-07,
      "loss": 0.84425855,
      "num_input_tokens_seen": 125381625,
      "step": 5825,
      "time_per_iteration": 2.8161470890045166
    },
    {
      "auxiliary_loss_clip": 0.01325322,
      "auxiliary_loss_mlp": 0.02568508,
      "balance_loss_clip": 1.04228449,
      "balance_loss_mlp": 0.99985254,
      "epoch": 0.7005350808633439,
      "flos": 19423758764160.0,
      "grad_norm": 2.8863477176638184,
      "language_loss": 0.9051851,
      "learning_rate": 8.691035037498354e-07,
      "loss": 0.94412339,
      "num_input_tokens_seen": 125397615,
      "step": 5826,
      "time_per_iteration": 2.738332509994507
    },
    {
      "auxiliary_loss_clip": 0.01273697,
      "auxiliary_loss_mlp": 0.01021856,
      "balance_loss_clip": 1.04404545,
      "balance_loss_mlp": 1.01519787,
      "epoch": 0.7006553237539831,
      "flos": 23476708938240.0,
      "grad_norm": 1.6930051593025202,
      "language_loss": 0.72533596,
      "learning_rate": 8.684611063667391e-07,
      "loss": 0.74829149,
      "num_input_tokens_seen": 125418080,
      "step": 5827,
      "time_per_iteration": 2.6811234951019287
    },
    {
      "auxiliary_loss_clip": 0.01225578,
      "auxiliary_loss_mlp": 0.01022821,
      "balance_loss_clip": 1.04703009,
      "balance_loss_mlp": 1.0158329,
      "epoch": 0.7007755666446221,
      "flos": 31212872640000.0,
      "grad_norm": 8.924215925267575,
      "language_loss": 0.76759487,
      "learning_rate": 8.678188806371935e-07,
      "loss": 0.79007882,
      "num_input_tokens_seen": 125440115,
      "step": 5828,
      "time_per_iteration": 2.7661795616149902
    },
    {
      "auxiliary_loss_clip": 0.01222941,
      "auxiliary_loss_mlp": 0.01022918,
      "balance_loss_clip": 1.04635191,
      "balance_loss_mlp": 1.0161469,
      "epoch": 0.7008958095352612,
      "flos": 18149899858560.0,
      "grad_norm": 4.644818167254538,
      "language_loss": 0.85371614,
      "learning_rate": 8.671768266586228e-07,
      "loss": 0.87617475,
      "num_input_tokens_seen": 125458240,
      "step": 5829,
      "time_per_iteration": 4.507591485977173
    },
    {
      "auxiliary_loss_clip": 0.01322516,
      "auxiliary_loss_mlp": 0.01029864,
      "balance_loss_clip": 1.04527092,
      "balance_loss_mlp": 1.02315867,
      "epoch": 0.7010160524259004,
      "flos": 27452307173760.0,
      "grad_norm": 1.7334660006742766,
      "language_loss": 0.78183752,
      "learning_rate": 8.665349445284275e-07,
      "loss": 0.80536127,
      "num_input_tokens_seen": 125477980,
      "step": 5830,
      "time_per_iteration": 2.771275520324707
    },
    {
      "auxiliary_loss_clip": 0.01326066,
      "auxiliary_loss_mlp": 0.01027281,
      "balance_loss_clip": 1.04833698,
      "balance_loss_mlp": 1.01993823,
      "epoch": 0.7011362953165394,
      "flos": 23842064125440.0,
      "grad_norm": 1.5061386227351443,
      "language_loss": 0.80914295,
      "learning_rate": 8.658932343439799e-07,
      "loss": 0.83267647,
      "num_input_tokens_seen": 125497765,
      "step": 5831,
      "time_per_iteration": 3.6157174110412598
    },
    {
      "auxiliary_loss_clip": 0.01178661,
      "auxiliary_loss_mlp": 0.0102918,
      "balance_loss_clip": 1.05232799,
      "balance_loss_mlp": 1.021837,
      "epoch": 0.7012565382071785,
      "flos": 24823430582400.0,
      "grad_norm": 2.091485899682168,
      "language_loss": 0.77733493,
      "learning_rate": 8.65251696202627e-07,
      "loss": 0.79941332,
      "num_input_tokens_seen": 125514145,
      "step": 5832,
      "time_per_iteration": 2.6113028526306152
    },
    {
      "auxiliary_loss_clip": 0.01224487,
      "auxiliary_loss_mlp": 0.0102882,
      "balance_loss_clip": 1.04647875,
      "balance_loss_mlp": 1.02137589,
      "epoch": 0.7013767810978175,
      "flos": 21397445326080.0,
      "grad_norm": 2.2499485188242576,
      "language_loss": 0.87401867,
      "learning_rate": 8.646103302016896e-07,
      "loss": 0.89655173,
      "num_input_tokens_seen": 125533115,
      "step": 5833,
      "time_per_iteration": 2.6925222873687744
    },
    {
      "auxiliary_loss_clip": 0.01383361,
      "auxiliary_loss_mlp": 0.01026562,
      "balance_loss_clip": 1.04424441,
      "balance_loss_mlp": 1.01885509,
      "epoch": 0.7014970239884567,
      "flos": 16687150306560.0,
      "grad_norm": 2.7300553500435187,
      "language_loss": 0.88217622,
      "learning_rate": 8.639691364384614e-07,
      "loss": 0.90627539,
      "num_input_tokens_seen": 125550740,
      "step": 5834,
      "time_per_iteration": 2.673872470855713
    },
    {
      "auxiliary_loss_clip": 0.0127722,
      "auxiliary_loss_mlp": 0.01030574,
      "balance_loss_clip": 1.04841197,
      "balance_loss_mlp": 1.023368,
      "epoch": 0.7016172668790958,
      "flos": 12568268718720.0,
      "grad_norm": 2.515700738086435,
      "language_loss": 0.72301596,
      "learning_rate": 8.633281150102136e-07,
      "loss": 0.74609387,
      "num_input_tokens_seen": 125567590,
      "step": 5835,
      "time_per_iteration": 2.7368955612182617
    },
    {
      "auxiliary_loss_clip": 0.01271882,
      "auxiliary_loss_mlp": 0.01021388,
      "balance_loss_clip": 1.04716158,
      "balance_loss_mlp": 1.01435518,
      "epoch": 0.7017375097697348,
      "flos": 17452729808640.0,
      "grad_norm": 3.3486200728699425,
      "language_loss": 0.67551512,
      "learning_rate": 8.626872660141855e-07,
      "loss": 0.69844782,
      "num_input_tokens_seen": 125585500,
      "step": 5836,
      "time_per_iteration": 3.5676157474517822
    },
    {
      "auxiliary_loss_clip": 0.01372814,
      "auxiliary_loss_mlp": 0.01033274,
      "balance_loss_clip": 1.04675174,
      "balance_loss_mlp": 1.02611542,
      "epoch": 0.701857752660374,
      "flos": 18513028402560.0,
      "grad_norm": 3.5928782627405673,
      "language_loss": 0.74956286,
      "learning_rate": 8.620465895475957e-07,
      "loss": 0.7736237,
      "num_input_tokens_seen": 125603720,
      "step": 5837,
      "time_per_iteration": 2.645437479019165
    },
    {
      "auxiliary_loss_clip": 0.013704,
      "auxiliary_loss_mlp": 0.01031762,
      "balance_loss_clip": 1.04538417,
      "balance_loss_mlp": 1.02466893,
      "epoch": 0.701977995551013,
      "flos": 24425971614720.0,
      "grad_norm": 1.4458550645792665,
      "language_loss": 0.7539798,
      "learning_rate": 8.614060857076333e-07,
      "loss": 0.77800143,
      "num_input_tokens_seen": 125624390,
      "step": 5838,
      "time_per_iteration": 2.665581226348877
    },
    {
      "auxiliary_loss_clip": 0.01270931,
      "auxiliary_loss_mlp": 0.01028675,
      "balance_loss_clip": 1.04628468,
      "balance_loss_mlp": 1.02142119,
      "epoch": 0.7020982384416521,
      "flos": 23002759958400.0,
      "grad_norm": 2.537891675864813,
      "language_loss": 0.75265181,
      "learning_rate": 8.60765754591462e-07,
      "loss": 0.77564788,
      "num_input_tokens_seen": 125644085,
      "step": 5839,
      "time_per_iteration": 2.5448806285858154
    },
    {
      "auxiliary_loss_clip": 0.01175151,
      "auxiliary_loss_mlp": 0.01021948,
      "balance_loss_clip": 1.04925156,
      "balance_loss_mlp": 1.01445544,
      "epoch": 0.7022184813322913,
      "flos": 20449080489600.0,
      "grad_norm": 2.3361447180321306,
      "language_loss": 0.72705787,
      "learning_rate": 8.601255962962211e-07,
      "loss": 0.7490288,
      "num_input_tokens_seen": 125663095,
      "step": 5840,
      "time_per_iteration": 2.5022740364074707
    },
    {
      "auxiliary_loss_clip": 0.01235639,
      "auxiliary_loss_mlp": 0.01030932,
      "balance_loss_clip": 1.05236316,
      "balance_loss_mlp": 1.02279603,
      "epoch": 0.7023387242229303,
      "flos": 19790514581760.0,
      "grad_norm": 2.342792630063301,
      "language_loss": 0.72028518,
      "learning_rate": 8.594856109190194e-07,
      "loss": 0.74295092,
      "num_input_tokens_seen": 125680125,
      "step": 5841,
      "time_per_iteration": 2.6082193851470947
    },
    {
      "auxiliary_loss_clip": 0.01175711,
      "auxiliary_loss_mlp": 0.01022886,
      "balance_loss_clip": 1.05000961,
      "balance_loss_mlp": 1.01583481,
      "epoch": 0.7024589671135694,
      "flos": 33259278286080.0,
      "grad_norm": 1.7383305081462501,
      "language_loss": 0.69289207,
      "learning_rate": 8.588457985569446e-07,
      "loss": 0.71487808,
      "num_input_tokens_seen": 125703035,
      "step": 5842,
      "time_per_iteration": 2.672624111175537
    },
    {
      "auxiliary_loss_clip": 0.0117785,
      "auxiliary_loss_mlp": 0.01022049,
      "balance_loss_clip": 1.05060935,
      "balance_loss_mlp": 1.01484299,
      "epoch": 0.7025792100042085,
      "flos": 19098982967040.0,
      "grad_norm": 2.0507915880196377,
      "language_loss": 0.71899569,
      "learning_rate": 8.582061593070542e-07,
      "loss": 0.74099463,
      "num_input_tokens_seen": 125723765,
      "step": 5843,
      "time_per_iteration": 2.6070573329925537
    },
    {
      "auxiliary_loss_clip": 0.01178074,
      "auxiliary_loss_mlp": 0.02568107,
      "balance_loss_clip": 1.05160713,
      "balance_loss_mlp": 0.99989367,
      "epoch": 0.7026994528948476,
      "flos": 18952611045120.0,
      "grad_norm": 2.9068587662098295,
      "language_loss": 0.77310216,
      "learning_rate": 8.57566693266383e-07,
      "loss": 0.81056398,
      "num_input_tokens_seen": 125741455,
      "step": 5844,
      "time_per_iteration": 2.590712785720825
    },
    {
      "auxiliary_loss_clip": 0.01280028,
      "auxiliary_loss_mlp": 0.02567463,
      "balance_loss_clip": 1.0468905,
      "balance_loss_mlp": 0.99987864,
      "epoch": 0.7028196957854866,
      "flos": 19536662188800.0,
      "grad_norm": 2.2356630541363756,
      "language_loss": 0.69568866,
      "learning_rate": 8.569274005319354e-07,
      "loss": 0.73416358,
      "num_input_tokens_seen": 125759855,
      "step": 5845,
      "time_per_iteration": 2.6465060710906982
    },
    {
      "auxiliary_loss_clip": 0.01223054,
      "auxiliary_loss_mlp": 0.01029243,
      "balance_loss_clip": 1.04733467,
      "balance_loss_mlp": 1.02223361,
      "epoch": 0.7029399386761258,
      "flos": 20845318394880.0,
      "grad_norm": 1.766331721677987,
      "language_loss": 0.7950592,
      "learning_rate": 8.562882812006913e-07,
      "loss": 0.81758213,
      "num_input_tokens_seen": 125777345,
      "step": 5846,
      "time_per_iteration": 2.6434860229492188
    },
    {
      "auxiliary_loss_clip": 0.01174772,
      "auxiliary_loss_mlp": 0.01025516,
      "balance_loss_clip": 1.04964089,
      "balance_loss_mlp": 1.01831865,
      "epoch": 0.7030601815667649,
      "flos": 22055005653120.0,
      "grad_norm": 2.1766117444183997,
      "language_loss": 0.77882189,
      "learning_rate": 8.556493353696066e-07,
      "loss": 0.80082476,
      "num_input_tokens_seen": 125796345,
      "step": 5847,
      "time_per_iteration": 2.575765609741211
    },
    {
      "auxiliary_loss_clip": 0.01229444,
      "auxiliary_loss_mlp": 0.02571439,
      "balance_loss_clip": 1.05065453,
      "balance_loss_mlp": 0.99989539,
      "epoch": 0.7031804244574039,
      "flos": 27198742089600.0,
      "grad_norm": 57.7171915874426,
      "language_loss": 0.68232363,
      "learning_rate": 8.550105631356077e-07,
      "loss": 0.72033238,
      "num_input_tokens_seen": 125816070,
      "step": 5848,
      "time_per_iteration": 2.6934595108032227
    },
    {
      "auxiliary_loss_clip": 0.01320834,
      "auxiliary_loss_mlp": 0.01027999,
      "balance_loss_clip": 1.04310489,
      "balance_loss_mlp": 1.02038169,
      "epoch": 0.7033006673480431,
      "flos": 22379853277440.0,
      "grad_norm": 3.079164227919468,
      "language_loss": 0.77285826,
      "learning_rate": 8.543719645955961e-07,
      "loss": 0.79634666,
      "num_input_tokens_seen": 125834400,
      "step": 5849,
      "time_per_iteration": 2.714679718017578
    },
    {
      "auxiliary_loss_clip": 0.01276106,
      "auxiliary_loss_mlp": 0.01024861,
      "balance_loss_clip": 1.04695451,
      "balance_loss_mlp": 1.01817656,
      "epoch": 0.7034209102386821,
      "flos": 24715986024960.0,
      "grad_norm": 1.701145044297368,
      "language_loss": 0.74486208,
      "learning_rate": 8.537335398464467e-07,
      "loss": 0.76787174,
      "num_input_tokens_seen": 125854720,
      "step": 5850,
      "time_per_iteration": 2.6793060302734375
    },
    {
      "auxiliary_loss_clip": 0.01276289,
      "auxiliary_loss_mlp": 0.0102752,
      "balance_loss_clip": 1.04513073,
      "balance_loss_mlp": 1.02022147,
      "epoch": 0.7035411531293212,
      "flos": 22556174163840.0,
      "grad_norm": 2.7951797320621354,
      "language_loss": 0.85315126,
      "learning_rate": 8.53095288985007e-07,
      "loss": 0.87618935,
      "num_input_tokens_seen": 125868455,
      "step": 5851,
      "time_per_iteration": 2.693242073059082
    },
    {
      "auxiliary_loss_clip": 0.01173585,
      "auxiliary_loss_mlp": 0.01021946,
      "balance_loss_clip": 1.05021632,
      "balance_loss_mlp": 1.01555955,
      "epoch": 0.7036613960199604,
      "flos": 22674967418880.0,
      "grad_norm": 2.3081902934757017,
      "language_loss": 0.82448292,
      "learning_rate": 8.524572121081009e-07,
      "loss": 0.84643829,
      "num_input_tokens_seen": 125888555,
      "step": 5852,
      "time_per_iteration": 2.592787742614746
    },
    {
      "auxiliary_loss_clip": 0.01228457,
      "auxiliary_loss_mlp": 0.01025121,
      "balance_loss_clip": 1.04767096,
      "balance_loss_mlp": 1.01793325,
      "epoch": 0.7037816389105994,
      "flos": 22492146170880.0,
      "grad_norm": 4.074057104036073,
      "language_loss": 0.62265337,
      "learning_rate": 8.518193093125232e-07,
      "loss": 0.64518917,
      "num_input_tokens_seen": 125907610,
      "step": 5853,
      "time_per_iteration": 2.6831910610198975
    },
    {
      "auxiliary_loss_clip": 0.01280946,
      "auxiliary_loss_mlp": 0.01026854,
      "balance_loss_clip": 1.04924107,
      "balance_loss_mlp": 1.02014804,
      "epoch": 0.7039018818012385,
      "flos": 27087490690560.0,
      "grad_norm": 2.1546275681548996,
      "language_loss": 0.8091526,
      "learning_rate": 8.511815806950436e-07,
      "loss": 0.83223057,
      "num_input_tokens_seen": 125928640,
      "step": 5854,
      "time_per_iteration": 2.683828115463257
    },
    {
      "auxiliary_loss_clip": 0.01219837,
      "auxiliary_loss_mlp": 0.01024687,
      "balance_loss_clip": 1.0453577,
      "balance_loss_mlp": 1.01766872,
      "epoch": 0.7040221246918776,
      "flos": 17749819198080.0,
      "grad_norm": 2.2492706098012127,
      "language_loss": 0.77795136,
      "learning_rate": 8.505440263524044e-07,
      "loss": 0.80039656,
      "num_input_tokens_seen": 125947485,
      "step": 5855,
      "time_per_iteration": 4.396458864212036
    },
    {
      "auxiliary_loss_clip": 0.01227481,
      "auxiliary_loss_mlp": 0.01030333,
      "balance_loss_clip": 1.04681945,
      "balance_loss_mlp": 1.02315938,
      "epoch": 0.7041423675825167,
      "flos": 16279851012480.0,
      "grad_norm": 3.834735511742285,
      "language_loss": 0.88285607,
      "learning_rate": 8.49906646381322e-07,
      "loss": 0.90543419,
      "num_input_tokens_seen": 125960320,
      "step": 5856,
      "time_per_iteration": 2.604480266571045
    },
    {
      "auxiliary_loss_clip": 0.01224927,
      "auxiliary_loss_mlp": 0.01026677,
      "balance_loss_clip": 1.04611909,
      "balance_loss_mlp": 1.0193758,
      "epoch": 0.7042626104731557,
      "flos": 25483181639040.0,
      "grad_norm": 2.30531638708349,
      "language_loss": 0.72750664,
      "learning_rate": 8.492694408784884e-07,
      "loss": 0.75002265,
      "num_input_tokens_seen": 125980575,
      "step": 5857,
      "time_per_iteration": 3.790470838546753
    },
    {
      "auxiliary_loss_clip": 0.01231006,
      "auxiliary_loss_mlp": 0.01027191,
      "balance_loss_clip": 1.04977155,
      "balance_loss_mlp": 1.02017248,
      "epoch": 0.7043828533637949,
      "flos": 17857622891520.0,
      "grad_norm": 2.9448212336801287,
      "language_loss": 0.62701344,
      "learning_rate": 8.486324099405642e-07,
      "loss": 0.64959538,
      "num_input_tokens_seen": 125997420,
      "step": 5858,
      "time_per_iteration": 2.621645927429199
    },
    {
      "auxiliary_loss_clip": 0.01223374,
      "auxiliary_loss_mlp": 0.01020538,
      "balance_loss_clip": 1.04770017,
      "balance_loss_mlp": 1.01385331,
      "epoch": 0.704503096254434,
      "flos": 29494259533440.0,
      "grad_norm": 1.7128199747647768,
      "language_loss": 0.74780846,
      "learning_rate": 8.479955536641887e-07,
      "loss": 0.77024764,
      "num_input_tokens_seen": 126018915,
      "step": 5859,
      "time_per_iteration": 2.6963984966278076
    },
    {
      "auxiliary_loss_clip": 0.01267917,
      "auxiliary_loss_mlp": 0.01030276,
      "balance_loss_clip": 1.04182363,
      "balance_loss_mlp": 1.02308452,
      "epoch": 0.704623339145073,
      "flos": 30920739327360.0,
      "grad_norm": 1.9732411190955865,
      "language_loss": 0.66182995,
      "learning_rate": 8.473588721459716e-07,
      "loss": 0.68481183,
      "num_input_tokens_seen": 126038825,
      "step": 5860,
      "time_per_iteration": 2.753777027130127
    },
    {
      "auxiliary_loss_clip": 0.0122683,
      "auxiliary_loss_mlp": 0.01028046,
      "balance_loss_clip": 1.05186796,
      "balance_loss_mlp": 1.01961207,
      "epoch": 0.7047435820357122,
      "flos": 23914747296000.0,
      "grad_norm": 2.2992100514095672,
      "language_loss": 0.70615333,
      "learning_rate": 8.467223654824967e-07,
      "loss": 0.72870207,
      "num_input_tokens_seen": 126058280,
      "step": 5861,
      "time_per_iteration": 2.8392271995544434
    },
    {
      "auxiliary_loss_clip": 0.01223292,
      "auxiliary_loss_mlp": 0.01023924,
      "balance_loss_clip": 1.04760635,
      "balance_loss_mlp": 1.01674128,
      "epoch": 0.7048638249263512,
      "flos": 46494010926720.0,
      "grad_norm": 2.84416271939699,
      "language_loss": 0.62650132,
      "learning_rate": 8.460860337703233e-07,
      "loss": 0.64897346,
      "num_input_tokens_seen": 126078885,
      "step": 5862,
      "time_per_iteration": 3.730315685272217
    },
    {
      "auxiliary_loss_clip": 0.01323844,
      "auxiliary_loss_mlp": 0.01026049,
      "balance_loss_clip": 1.04630995,
      "balance_loss_mlp": 1.01817203,
      "epoch": 0.7049840678169903,
      "flos": 21689219502720.0,
      "grad_norm": 2.446307049655094,
      "language_loss": 0.70672852,
      "learning_rate": 8.454498771059797e-07,
      "loss": 0.73022747,
      "num_input_tokens_seen": 126098260,
      "step": 5863,
      "time_per_iteration": 2.7182745933532715
    },
    {
      "auxiliary_loss_clip": 0.01367319,
      "auxiliary_loss_mlp": 0.01027917,
      "balance_loss_clip": 1.04310918,
      "balance_loss_mlp": 1.02109194,
      "epoch": 0.7051043107076294,
      "flos": 18405081054720.0,
      "grad_norm": 2.041046344920763,
      "language_loss": 0.83455539,
      "learning_rate": 8.448138955859725e-07,
      "loss": 0.85850775,
      "num_input_tokens_seen": 126114845,
      "step": 5864,
      "time_per_iteration": 2.6875553131103516
    },
    {
      "auxiliary_loss_clip": 0.01274644,
      "auxiliary_loss_mlp": 0.01022743,
      "balance_loss_clip": 1.04812896,
      "balance_loss_mlp": 1.01557839,
      "epoch": 0.7052245535982685,
      "flos": 19319043640320.0,
      "grad_norm": 1.9993237736629212,
      "language_loss": 0.90248072,
      "learning_rate": 8.44178089306778e-07,
      "loss": 0.92545462,
      "num_input_tokens_seen": 126132780,
      "step": 5865,
      "time_per_iteration": 2.703164577484131
    },
    {
      "auxiliary_loss_clip": 0.01175714,
      "auxiliary_loss_mlp": 0.01023868,
      "balance_loss_clip": 1.05058491,
      "balance_loss_mlp": 1.01682258,
      "epoch": 0.7053447964889076,
      "flos": 19062138591360.0,
      "grad_norm": 1.8382054431002621,
      "language_loss": 0.76951301,
      "learning_rate": 8.4354245836485e-07,
      "loss": 0.79150879,
      "num_input_tokens_seen": 126151225,
      "step": 5866,
      "time_per_iteration": 2.5712578296661377
    },
    {
      "auxiliary_loss_clip": 0.01326026,
      "auxiliary_loss_mlp": 0.01029526,
      "balance_loss_clip": 1.0467732,
      "balance_loss_mlp": 1.02187026,
      "epoch": 0.7054650393795466,
      "flos": 27379228953600.0,
      "grad_norm": 1.5918280349820018,
      "language_loss": 0.72974515,
      "learning_rate": 8.429070028566108e-07,
      "loss": 0.75330067,
      "num_input_tokens_seen": 126172535,
      "step": 5867,
      "time_per_iteration": 2.743687152862549
    },
    {
      "auxiliary_loss_clip": 0.0122314,
      "auxiliary_loss_mlp": 0.01030092,
      "balance_loss_clip": 1.04869914,
      "balance_loss_mlp": 1.02332389,
      "epoch": 0.7055852822701858,
      "flos": 16102201322880.0,
      "grad_norm": 2.086282197646638,
      "language_loss": 0.75142312,
      "learning_rate": 8.422717228784586e-07,
      "loss": 0.77395546,
      "num_input_tokens_seen": 126189410,
      "step": 5868,
      "time_per_iteration": 2.631183624267578
    },
    {
      "auxiliary_loss_clip": 0.01371949,
      "auxiliary_loss_mlp": 0.01024167,
      "balance_loss_clip": 1.04703808,
      "balance_loss_mlp": 1.01665688,
      "epoch": 0.7057055251608249,
      "flos": 11692299744000.0,
      "grad_norm": 5.189896558309082,
      "language_loss": 0.69133031,
      "learning_rate": 8.416366185267663e-07,
      "loss": 0.71529144,
      "num_input_tokens_seen": 126206910,
      "step": 5869,
      "time_per_iteration": 2.6890931129455566
    },
    {
      "auxiliary_loss_clip": 0.01223921,
      "auxiliary_loss_mlp": 0.01028378,
      "balance_loss_clip": 1.04742956,
      "balance_loss_mlp": 1.02152395,
      "epoch": 0.7058257680514639,
      "flos": 22711560399360.0,
      "grad_norm": 7.449210107110314,
      "language_loss": 0.77938664,
      "learning_rate": 8.410016898978778e-07,
      "loss": 0.80190969,
      "num_input_tokens_seen": 126224385,
      "step": 5870,
      "time_per_iteration": 2.6699061393737793
    },
    {
      "auxiliary_loss_clip": 0.01273191,
      "auxiliary_loss_mlp": 0.01026559,
      "balance_loss_clip": 1.04776239,
      "balance_loss_mlp": 1.01990986,
      "epoch": 0.7059460109421031,
      "flos": 17529543043200.0,
      "grad_norm": 2.274506071059542,
      "language_loss": 0.79139251,
      "learning_rate": 8.403669370881115e-07,
      "loss": 0.81439006,
      "num_input_tokens_seen": 126243120,
      "step": 5871,
      "time_per_iteration": 2.7218527793884277
    },
    {
      "auxiliary_loss_clip": 0.01175611,
      "auxiliary_loss_mlp": 0.01022894,
      "balance_loss_clip": 1.05041575,
      "balance_loss_mlp": 1.01616812,
      "epoch": 0.7060662538327421,
      "flos": 23544687427200.0,
      "grad_norm": 1.5785796481828511,
      "language_loss": 0.7837435,
      "learning_rate": 8.397323601937587e-07,
      "loss": 0.80572855,
      "num_input_tokens_seen": 126263020,
      "step": 5872,
      "time_per_iteration": 2.636770486831665
    },
    {
      "auxiliary_loss_clip": 0.01318464,
      "auxiliary_loss_mlp": 0.01024114,
      "balance_loss_clip": 1.04476452,
      "balance_loss_mlp": 1.0172627,
      "epoch": 0.7061864967233812,
      "flos": 30260736875520.0,
      "grad_norm": 2.1171987976223634,
      "language_loss": 0.77131003,
      "learning_rate": 8.390979593110838e-07,
      "loss": 0.79473579,
      "num_input_tokens_seen": 126285150,
      "step": 5873,
      "time_per_iteration": 2.7573959827423096
    },
    {
      "auxiliary_loss_clip": 0.01278955,
      "auxiliary_loss_mlp": 0.01028027,
      "balance_loss_clip": 1.05010748,
      "balance_loss_mlp": 1.0204128,
      "epoch": 0.7063067396140204,
      "flos": 20701460424960.0,
      "grad_norm": 1.8432152140169413,
      "language_loss": 0.81607658,
      "learning_rate": 8.384637345363262e-07,
      "loss": 0.83914638,
      "num_input_tokens_seen": 126304340,
      "step": 5874,
      "time_per_iteration": 2.715832233428955
    },
    {
      "auxiliary_loss_clip": 0.01272082,
      "auxiliary_loss_mlp": 0.01029259,
      "balance_loss_clip": 1.04439831,
      "balance_loss_mlp": 1.02211809,
      "epoch": 0.7064269825046594,
      "flos": 32266168081920.0,
      "grad_norm": 2.511799582199191,
      "language_loss": 0.76527077,
      "learning_rate": 8.378296859656964e-07,
      "loss": 0.78828418,
      "num_input_tokens_seen": 126325495,
      "step": 5875,
      "time_per_iteration": 2.721820116043091
    },
    {
      "auxiliary_loss_clip": 0.01274683,
      "auxiliary_loss_mlp": 0.01027814,
      "balance_loss_clip": 1.04686618,
      "balance_loss_mlp": 1.02093577,
      "epoch": 0.7065472253952985,
      "flos": 30227124723840.0,
      "grad_norm": 2.733046752509775,
      "language_loss": 0.68440831,
      "learning_rate": 8.371958136953792e-07,
      "loss": 0.70743328,
      "num_input_tokens_seen": 126345525,
      "step": 5876,
      "time_per_iteration": 2.7167656421661377
    },
    {
      "auxiliary_loss_clip": 0.0132718,
      "auxiliary_loss_mlp": 0.01028547,
      "balance_loss_clip": 1.0450573,
      "balance_loss_mlp": 1.02032161,
      "epoch": 0.7066674682859376,
      "flos": 16216720859520.0,
      "grad_norm": 2.994247269083327,
      "language_loss": 0.65748644,
      "learning_rate": 8.365621178215326e-07,
      "loss": 0.68104374,
      "num_input_tokens_seen": 126361995,
      "step": 5877,
      "time_per_iteration": 2.669482707977295
    },
    {
      "auxiliary_loss_clip": 0.01220962,
      "auxiliary_loss_mlp": 0.01028303,
      "balance_loss_clip": 1.04704142,
      "balance_loss_mlp": 1.02087605,
      "epoch": 0.7067877111765767,
      "flos": 14830461319680.0,
      "grad_norm": 2.1957228564987,
      "language_loss": 0.75391889,
      "learning_rate": 8.359285984402871e-07,
      "loss": 0.77641153,
      "num_input_tokens_seen": 126379260,
      "step": 5878,
      "time_per_iteration": 2.6058285236358643
    },
    {
      "auxiliary_loss_clip": 0.01268244,
      "auxiliary_loss_mlp": 0.01029884,
      "balance_loss_clip": 1.04709935,
      "balance_loss_mlp": 1.0231396,
      "epoch": 0.7069079540672157,
      "flos": 25440196037760.0,
      "grad_norm": 1.9607184593279743,
      "language_loss": 0.74008048,
      "learning_rate": 8.352952556477489e-07,
      "loss": 0.76306176,
      "num_input_tokens_seen": 126397170,
      "step": 5879,
      "time_per_iteration": 2.6542465686798096
    },
    {
      "auxiliary_loss_clip": 0.01221751,
      "auxiliary_loss_mlp": 0.01032044,
      "balance_loss_clip": 1.04806232,
      "balance_loss_mlp": 1.02465892,
      "epoch": 0.7070281969578549,
      "flos": 24607751368320.0,
      "grad_norm": 2.495515794061675,
      "language_loss": 0.76503319,
      "learning_rate": 8.34662089539993e-07,
      "loss": 0.78757107,
      "num_input_tokens_seen": 126416680,
      "step": 5880,
      "time_per_iteration": 3.6216371059417725
    },
    {
      "auxiliary_loss_clip": 0.01175538,
      "auxiliary_loss_mlp": 0.01028813,
      "balance_loss_clip": 1.05124855,
      "balance_loss_mlp": 1.02275753,
      "epoch": 0.707148439848494,
      "flos": 26724469887360.0,
      "grad_norm": 1.9170247842489587,
      "language_loss": 0.79557258,
      "learning_rate": 8.340291002130722e-07,
      "loss": 0.81761611,
      "num_input_tokens_seen": 126435870,
      "step": 5881,
      "time_per_iteration": 3.557610511779785
    },
    {
      "auxiliary_loss_clip": 0.01180563,
      "auxiliary_loss_mlp": 0.01025388,
      "balance_loss_clip": 1.05248594,
      "balance_loss_mlp": 1.01821804,
      "epoch": 0.707268682739133,
      "flos": 15085750256640.0,
      "grad_norm": 5.4960986216603676,
      "language_loss": 0.79354644,
      "learning_rate": 8.3339628776301e-07,
      "loss": 0.81560594,
      "num_input_tokens_seen": 126454010,
      "step": 5882,
      "time_per_iteration": 2.569021224975586
    },
    {
      "auxiliary_loss_clip": 0.01176654,
      "auxiliary_loss_mlp": 0.01024067,
      "balance_loss_clip": 1.05007148,
      "balance_loss_mlp": 1.01716816,
      "epoch": 0.7073889256297722,
      "flos": 34313148345600.0,
      "grad_norm": 1.9074211640316827,
      "language_loss": 0.57167542,
      "learning_rate": 8.327636522858033e-07,
      "loss": 0.59368265,
      "num_input_tokens_seen": 126473615,
      "step": 5883,
      "time_per_iteration": 3.603419303894043
    },
    {
      "auxiliary_loss_clip": 0.01273633,
      "auxiliary_loss_mlp": 0.01025753,
      "balance_loss_clip": 1.04752493,
      "balance_loss_mlp": 1.01898539,
      "epoch": 0.7075091685204112,
      "flos": 20083940784000.0,
      "grad_norm": 1.7886341529385117,
      "language_loss": 0.77226913,
      "learning_rate": 8.321311938774225e-07,
      "loss": 0.79526299,
      "num_input_tokens_seen": 126492705,
      "step": 5884,
      "time_per_iteration": 2.7550671100616455
    },
    {
      "auxiliary_loss_clip": 0.01178825,
      "auxiliary_loss_mlp": 0.01033714,
      "balance_loss_clip": 1.05036688,
      "balance_loss_mlp": 1.02613842,
      "epoch": 0.7076294114110503,
      "flos": 20777124424320.0,
      "grad_norm": 2.1234646838317515,
      "language_loss": 0.7906186,
      "learning_rate": 8.314989126338104e-07,
      "loss": 0.81274402,
      "num_input_tokens_seen": 126512715,
      "step": 5885,
      "time_per_iteration": 2.645000696182251
    },
    {
      "auxiliary_loss_clip": 0.01225524,
      "auxiliary_loss_mlp": 0.01029877,
      "balance_loss_clip": 1.0466609,
      "balance_loss_mlp": 1.02273655,
      "epoch": 0.7077496543016895,
      "flos": 17967689141760.0,
      "grad_norm": 2.461786785762009,
      "language_loss": 0.84605008,
      "learning_rate": 8.308668086508847e-07,
      "loss": 0.86860406,
      "num_input_tokens_seen": 126530795,
      "step": 5886,
      "time_per_iteration": 2.6028425693511963
    },
    {
      "auxiliary_loss_clip": 0.01326716,
      "auxiliary_loss_mlp": 0.01024855,
      "balance_loss_clip": 1.0421418,
      "balance_loss_mlp": 1.01815224,
      "epoch": 0.7078698971923285,
      "flos": 45478098564480.0,
      "grad_norm": 2.147887423885243,
      "language_loss": 0.7381261,
      "learning_rate": 8.302348820245342e-07,
      "loss": 0.7616418,
      "num_input_tokens_seen": 126553360,
      "step": 5887,
      "time_per_iteration": 2.8925933837890625
    },
    {
      "auxiliary_loss_clip": 0.01229053,
      "auxiliary_loss_mlp": 0.01027962,
      "balance_loss_clip": 1.04652786,
      "balance_loss_mlp": 1.01994538,
      "epoch": 0.7079901400829676,
      "flos": 26943704547840.0,
      "grad_norm": 6.1058056464656225,
      "language_loss": 0.69578564,
      "learning_rate": 8.296031328506232e-07,
      "loss": 0.71835577,
      "num_input_tokens_seen": 126573110,
      "step": 5888,
      "time_per_iteration": 3.645230293273926
    },
    {
      "auxiliary_loss_clip": 0.01273987,
      "auxiliary_loss_mlp": 0.01024281,
      "balance_loss_clip": 1.04616189,
      "balance_loss_mlp": 1.01713407,
      "epoch": 0.7081103829736067,
      "flos": 24423206267520.0,
      "grad_norm": 2.4432586919608883,
      "language_loss": 0.75670993,
      "learning_rate": 8.289715612249857e-07,
      "loss": 0.77969265,
      "num_input_tokens_seen": 126593725,
      "step": 5889,
      "time_per_iteration": 2.7276535034179688
    },
    {
      "auxiliary_loss_clip": 0.01273725,
      "auxiliary_loss_mlp": 0.0102477,
      "balance_loss_clip": 1.04748881,
      "balance_loss_mlp": 1.01736689,
      "epoch": 0.7082306258642458,
      "flos": 18543300589440.0,
      "grad_norm": 2.5833044574865274,
      "language_loss": 0.77817464,
      "learning_rate": 8.283401672434305e-07,
      "loss": 0.80115962,
      "num_input_tokens_seen": 126608950,
      "step": 5890,
      "time_per_iteration": 2.6112139225006104
    },
    {
      "auxiliary_loss_clip": 0.01270694,
      "auxiliary_loss_mlp": 0.01023496,
      "balance_loss_clip": 1.04887295,
      "balance_loss_mlp": 1.01700854,
      "epoch": 0.7083508687548848,
      "flos": 23477534951040.0,
      "grad_norm": 2.388482628775998,
      "language_loss": 0.70531482,
      "learning_rate": 8.277089510017412e-07,
      "loss": 0.7282567,
      "num_input_tokens_seen": 126629755,
      "step": 5891,
      "time_per_iteration": 2.712289810180664
    },
    {
      "auxiliary_loss_clip": 0.01273063,
      "auxiliary_loss_mlp": 0.01026024,
      "balance_loss_clip": 1.0498575,
      "balance_loss_mlp": 1.01950622,
      "epoch": 0.708471111645524,
      "flos": 22419463000320.0,
      "grad_norm": 1.8967734549157547,
      "language_loss": 0.82041168,
      "learning_rate": 8.270779125956719e-07,
      "loss": 0.84340256,
      "num_input_tokens_seen": 126650135,
      "step": 5892,
      "time_per_iteration": 2.673396110534668
    },
    {
      "auxiliary_loss_clip": 0.01367978,
      "auxiliary_loss_mlp": 0.0102359,
      "balance_loss_clip": 1.04325545,
      "balance_loss_mlp": 1.01644325,
      "epoch": 0.7085913545361631,
      "flos": 20922885815040.0,
      "grad_norm": 2.0027457090004575,
      "language_loss": 0.80273306,
      "learning_rate": 8.264470521209505e-07,
      "loss": 0.82664871,
      "num_input_tokens_seen": 126668500,
      "step": 5893,
      "time_per_iteration": 2.7325599193573
    },
    {
      "auxiliary_loss_clip": 0.01221528,
      "auxiliary_loss_mlp": 0.01027714,
      "balance_loss_clip": 1.0463903,
      "balance_loss_mlp": 1.0203408,
      "epoch": 0.7087115974268021,
      "flos": 15012384727680.0,
      "grad_norm": 5.798270690099815,
      "language_loss": 0.76736474,
      "learning_rate": 8.258163696732785e-07,
      "loss": 0.78985715,
      "num_input_tokens_seen": 126686090,
      "step": 5894,
      "time_per_iteration": 2.5857813358306885
    },
    {
      "auxiliary_loss_clip": 0.0122371,
      "auxiliary_loss_mlp": 0.01027614,
      "balance_loss_clip": 1.04723239,
      "balance_loss_mlp": 1.020172,
      "epoch": 0.7088318403174413,
      "flos": 21539040739200.0,
      "grad_norm": 2.2768191306622976,
      "language_loss": 0.76868582,
      "learning_rate": 8.251858653483288e-07,
      "loss": 0.79119909,
      "num_input_tokens_seen": 126704255,
      "step": 5895,
      "time_per_iteration": 2.6750447750091553
    },
    {
      "auxiliary_loss_clip": 0.01226464,
      "auxiliary_loss_mlp": 0.01029117,
      "balance_loss_clip": 1.05092287,
      "balance_loss_mlp": 1.02195835,
      "epoch": 0.7089520832080803,
      "flos": 15516785462400.0,
      "grad_norm": 2.8884882926127857,
      "language_loss": 0.85776639,
      "learning_rate": 8.245555392417501e-07,
      "loss": 0.88032222,
      "num_input_tokens_seen": 126718910,
      "step": 5896,
      "time_per_iteration": 2.549610137939453
    },
    {
      "auxiliary_loss_clip": 0.01362454,
      "auxiliary_loss_mlp": 0.01023924,
      "balance_loss_clip": 1.04020262,
      "balance_loss_mlp": 1.01706934,
      "epoch": 0.7090723260987194,
      "flos": 20412667077120.0,
      "grad_norm": 1.7035248770415534,
      "language_loss": 0.78921044,
      "learning_rate": 8.239253914491613e-07,
      "loss": 0.81307423,
      "num_input_tokens_seen": 126737235,
      "step": 5897,
      "time_per_iteration": 2.748521327972412
    },
    {
      "auxiliary_loss_clip": 0.01321131,
      "auxiliary_loss_mlp": 0.01028659,
      "balance_loss_clip": 1.04676545,
      "balance_loss_mlp": 1.02150393,
      "epoch": 0.7091925689893585,
      "flos": 25668337271040.0,
      "grad_norm": 2.4760273439137537,
      "language_loss": 0.75181961,
      "learning_rate": 8.232954220661556e-07,
      "loss": 0.77531755,
      "num_input_tokens_seen": 126759970,
      "step": 5898,
      "time_per_iteration": 2.7554194927215576
    },
    {
      "auxiliary_loss_clip": 0.01175608,
      "auxiliary_loss_mlp": 0.01026709,
      "balance_loss_clip": 1.05211294,
      "balance_loss_mlp": 1.02007484,
      "epoch": 0.7093128118799976,
      "flos": 24206629213440.0,
      "grad_norm": 2.6680784225991703,
      "language_loss": 0.70345795,
      "learning_rate": 8.226656311882989e-07,
      "loss": 0.72548115,
      "num_input_tokens_seen": 126779280,
      "step": 5899,
      "time_per_iteration": 2.631424903869629
    },
    {
      "auxiliary_loss_clip": 0.01222183,
      "auxiliary_loss_mlp": 0.01023305,
      "balance_loss_clip": 1.04944372,
      "balance_loss_mlp": 1.01686752,
      "epoch": 0.7094330547706367,
      "flos": 16646786398080.0,
      "grad_norm": 2.16701423217405,
      "language_loss": 0.76651001,
      "learning_rate": 8.22036018911129e-07,
      "loss": 0.78896493,
      "num_input_tokens_seen": 126797310,
      "step": 5900,
      "time_per_iteration": 2.5704567432403564
    },
    {
      "auxiliary_loss_clip": 0.01178927,
      "auxiliary_loss_mlp": 0.01027286,
      "balance_loss_clip": 1.05063879,
      "balance_loss_mlp": 1.01991308,
      "epoch": 0.7095532976612757,
      "flos": 16283370545280.0,
      "grad_norm": 2.1959944450809328,
      "language_loss": 0.80742484,
      "learning_rate": 8.214065853301599e-07,
      "loss": 0.82948697,
      "num_input_tokens_seen": 126812840,
      "step": 5901,
      "time_per_iteration": 2.639282703399658
    },
    {
      "auxiliary_loss_clip": 0.01125995,
      "auxiliary_loss_mlp": 0.0100049,
      "balance_loss_clip": 1.01273787,
      "balance_loss_mlp": 0.99947053,
      "epoch": 0.7096735405519149,
      "flos": 70722080559360.0,
      "grad_norm": 0.8148449386016225,
      "language_loss": 0.58199197,
      "learning_rate": 8.207773305408734e-07,
      "loss": 0.60325682,
      "num_input_tokens_seen": 126880060,
      "step": 5902,
      "time_per_iteration": 3.3316452503204346
    },
    {
      "auxiliary_loss_clip": 0.01379356,
      "auxiliary_loss_mlp": 0.01025292,
      "balance_loss_clip": 1.04325664,
      "balance_loss_mlp": 1.01788342,
      "epoch": 0.709793783442554,
      "flos": 23621500661760.0,
      "grad_norm": 3.0499334291563676,
      "language_loss": 0.80142498,
      "learning_rate": 8.201482546387288e-07,
      "loss": 0.82547146,
      "num_input_tokens_seen": 126899535,
      "step": 5903,
      "time_per_iteration": 2.7531797885894775
    },
    {
      "auxiliary_loss_clip": 0.01222996,
      "auxiliary_loss_mlp": 0.01030552,
      "balance_loss_clip": 1.04856241,
      "balance_loss_mlp": 1.02336979,
      "epoch": 0.709914026333193,
      "flos": 25993472204160.0,
      "grad_norm": 1.6403406910434282,
      "language_loss": 0.91694701,
      "learning_rate": 8.195193577191553e-07,
      "loss": 0.93948245,
      "num_input_tokens_seen": 126921365,
      "step": 5904,
      "time_per_iteration": 2.697624683380127
    },
    {
      "auxiliary_loss_clip": 0.0118215,
      "auxiliary_loss_mlp": 0.02566059,
      "balance_loss_clip": 1.04808342,
      "balance_loss_mlp": 0.99988061,
      "epoch": 0.7100342692238322,
      "flos": 24861531934080.0,
      "grad_norm": 1.7941677500031643,
      "language_loss": 0.84521824,
      "learning_rate": 8.188906398775579e-07,
      "loss": 0.88270032,
      "num_input_tokens_seen": 126941910,
      "step": 5905,
      "time_per_iteration": 2.6936471462249756
    },
    {
      "auxiliary_loss_clip": 0.0117527,
      "auxiliary_loss_mlp": 0.0256729,
      "balance_loss_clip": 1.04900765,
      "balance_loss_mlp": 0.99988139,
      "epoch": 0.7101545121144712,
      "flos": 24932203943040.0,
      "grad_norm": 2.1381997046122954,
      "language_loss": 0.68752319,
      "learning_rate": 8.18262101209311e-07,
      "loss": 0.72494876,
      "num_input_tokens_seen": 126961120,
      "step": 5906,
      "time_per_iteration": 3.6303818225860596
    },
    {
      "auxiliary_loss_clip": 0.01126573,
      "auxiliary_loss_mlp": 0.01020979,
      "balance_loss_clip": 1.04737425,
      "balance_loss_mlp": 1.01419616,
      "epoch": 0.7102747550051103,
      "flos": 23768842250880.0,
      "grad_norm": 1.9899901738622872,
      "language_loss": 0.7042762,
      "learning_rate": 8.176337418097626e-07,
      "loss": 0.72575176,
      "num_input_tokens_seen": 126981590,
      "step": 5907,
      "time_per_iteration": 3.547966957092285
    },
    {
      "auxiliary_loss_clip": 0.01226282,
      "auxiliary_loss_mlp": 0.02565341,
      "balance_loss_clip": 1.05224133,
      "balance_loss_mlp": 0.9998672,
      "epoch": 0.7103949978957494,
      "flos": 15303907509120.0,
      "grad_norm": 5.25090864534407,
      "language_loss": 0.79998672,
      "learning_rate": 8.170055617742364e-07,
      "loss": 0.83790296,
      "num_input_tokens_seen": 126998870,
      "step": 5908,
      "time_per_iteration": 2.618321418762207
    },
    {
      "auxiliary_loss_clip": 0.01268442,
      "auxiliary_loss_mlp": 0.01031833,
      "balance_loss_clip": 1.04528034,
      "balance_loss_mlp": 1.02428675,
      "epoch": 0.7105152407863885,
      "flos": 22638805401600.0,
      "grad_norm": 1.8235677249896851,
      "language_loss": 0.70527565,
      "learning_rate": 8.163775611980252e-07,
      "loss": 0.72827834,
      "num_input_tokens_seen": 127017980,
      "step": 5909,
      "time_per_iteration": 3.5883419513702393
    },
    {
      "auxiliary_loss_clip": 0.01273073,
      "auxiliary_loss_mlp": 0.01031095,
      "balance_loss_clip": 1.04906571,
      "balance_loss_mlp": 1.0242672,
      "epoch": 0.7106354836770276,
      "flos": 17238594879360.0,
      "grad_norm": 1.7990477578416386,
      "language_loss": 0.78999615,
      "learning_rate": 8.157497401763982e-07,
      "loss": 0.81303775,
      "num_input_tokens_seen": 127035645,
      "step": 5910,
      "time_per_iteration": 2.632392644882202
    },
    {
      "auxiliary_loss_clip": 0.01224995,
      "auxiliary_loss_mlp": 0.01029815,
      "balance_loss_clip": 1.05040765,
      "balance_loss_mlp": 1.02298141,
      "epoch": 0.7107557265676667,
      "flos": 20193647898240.0,
      "grad_norm": 1.8613478662150384,
      "language_loss": 0.77923644,
      "learning_rate": 8.151220988045935e-07,
      "loss": 0.80178452,
      "num_input_tokens_seen": 127054900,
      "step": 5911,
      "time_per_iteration": 2.6504716873168945
    },
    {
      "auxiliary_loss_clip": 0.01221208,
      "auxiliary_loss_mlp": 0.01028424,
      "balance_loss_clip": 1.04804432,
      "balance_loss_mlp": 1.02152193,
      "epoch": 0.7108759694583058,
      "flos": 21507080613120.0,
      "grad_norm": 2.6379825550999363,
      "language_loss": 0.82656986,
      "learning_rate": 8.144946371778234e-07,
      "loss": 0.8490662,
      "num_input_tokens_seen": 127075010,
      "step": 5912,
      "time_per_iteration": 2.660689353942871
    },
    {
      "auxiliary_loss_clip": 0.01273767,
      "auxiliary_loss_mlp": 0.02568841,
      "balance_loss_clip": 1.04858577,
      "balance_loss_mlp": 0.99990314,
      "epoch": 0.7109962123489448,
      "flos": 24061909317120.0,
      "grad_norm": 2.154621122456865,
      "language_loss": 0.78737092,
      "learning_rate": 8.138673553912751e-07,
      "loss": 0.82579702,
      "num_input_tokens_seen": 127095570,
      "step": 5913,
      "time_per_iteration": 2.658308982849121
    },
    {
      "auxiliary_loss_clip": 0.01274583,
      "auxiliary_loss_mlp": 0.01026522,
      "balance_loss_clip": 1.04372883,
      "balance_loss_mlp": 1.01913667,
      "epoch": 0.711116455239584,
      "flos": 30480474326400.0,
      "grad_norm": 7.315498804949607,
      "language_loss": 0.56623459,
      "learning_rate": 8.132402535401059e-07,
      "loss": 0.58924556,
      "num_input_tokens_seen": 127116825,
      "step": 5914,
      "time_per_iteration": 3.7115705013275146
    },
    {
      "auxiliary_loss_clip": 0.01230048,
      "auxiliary_loss_mlp": 0.01023406,
      "balance_loss_clip": 1.05254102,
      "balance_loss_mlp": 1.01690912,
      "epoch": 0.711236698130223,
      "flos": 25045610158080.0,
      "grad_norm": 1.9145228280617346,
      "language_loss": 0.74254203,
      "learning_rate": 8.126133317194465e-07,
      "loss": 0.76507658,
      "num_input_tokens_seen": 127137015,
      "step": 5915,
      "time_per_iteration": 2.6091763973236084
    },
    {
      "auxiliary_loss_clip": 0.0133845,
      "auxiliary_loss_mlp": 0.0102819,
      "balance_loss_clip": 1.04357541,
      "balance_loss_mlp": 1.02029204,
      "epoch": 0.7113569410208621,
      "flos": 24206701040640.0,
      "grad_norm": 4.569618945962764,
      "language_loss": 0.74755365,
      "learning_rate": 8.11986590024401e-07,
      "loss": 0.77122009,
      "num_input_tokens_seen": 127156755,
      "step": 5916,
      "time_per_iteration": 2.9392287731170654
    },
    {
      "auxiliary_loss_clip": 0.01283215,
      "auxiliary_loss_mlp": 0.0102648,
      "balance_loss_clip": 1.05006301,
      "balance_loss_mlp": 1.01854038,
      "epoch": 0.7114771839115013,
      "flos": 35439306526080.0,
      "grad_norm": 1.9407709516981124,
      "language_loss": 0.68813109,
      "learning_rate": 8.113600285500442e-07,
      "loss": 0.71122801,
      "num_input_tokens_seen": 127176965,
      "step": 5917,
      "time_per_iteration": 2.961409568786621
    },
    {
      "auxiliary_loss_clip": 0.01174812,
      "auxiliary_loss_mlp": 0.01025056,
      "balance_loss_clip": 1.04866672,
      "balance_loss_mlp": 1.01853275,
      "epoch": 0.7115974268021403,
      "flos": 21099458096640.0,
      "grad_norm": 1.9417182611035864,
      "language_loss": 0.74660933,
      "learning_rate": 8.107336473914268e-07,
      "loss": 0.76860809,
      "num_input_tokens_seen": 127195595,
      "step": 5918,
      "time_per_iteration": 2.6059730052948
    },
    {
      "auxiliary_loss_clip": 0.01176667,
      "auxiliary_loss_mlp": 0.01001252,
      "balance_loss_clip": 1.01200366,
      "balance_loss_mlp": 1.00016689,
      "epoch": 0.7117176696927794,
      "flos": 56752866616320.0,
      "grad_norm": 0.7722970597643075,
      "language_loss": 0.55763364,
      "learning_rate": 8.101074466435694e-07,
      "loss": 0.57941282,
      "num_input_tokens_seen": 127255070,
      "step": 5919,
      "time_per_iteration": 3.177396059036255
    },
    {
      "auxiliary_loss_clip": 0.01220526,
      "auxiliary_loss_mlp": 0.01028536,
      "balance_loss_clip": 1.04651916,
      "balance_loss_mlp": 1.02131152,
      "epoch": 0.7118379125834186,
      "flos": 15925269905280.0,
      "grad_norm": 4.256589370798534,
      "language_loss": 0.67999923,
      "learning_rate": 8.094814264014662e-07,
      "loss": 0.70248979,
      "num_input_tokens_seen": 127273825,
      "step": 5920,
      "time_per_iteration": 2.5858206748962402
    },
    {
      "auxiliary_loss_clip": 0.01178132,
      "auxiliary_loss_mlp": 0.01026883,
      "balance_loss_clip": 1.04989433,
      "balance_loss_mlp": 1.01943254,
      "epoch": 0.7119581554740576,
      "flos": 20193360589440.0,
      "grad_norm": 3.9833627989596208,
      "language_loss": 0.81369948,
      "learning_rate": 8.088555867600844e-07,
      "loss": 0.83574963,
      "num_input_tokens_seen": 127289990,
      "step": 5921,
      "time_per_iteration": 2.554027557373047
    },
    {
      "auxiliary_loss_clip": 0.01322248,
      "auxiliary_loss_mlp": 0.01025149,
      "balance_loss_clip": 1.04514027,
      "balance_loss_mlp": 1.01838994,
      "epoch": 0.7120783983646967,
      "flos": 34715383822080.0,
      "grad_norm": 1.8675214754439786,
      "language_loss": 0.60127515,
      "learning_rate": 8.08229927814362e-07,
      "loss": 0.62474906,
      "num_input_tokens_seen": 127312880,
      "step": 5922,
      "time_per_iteration": 2.793527841567993
    },
    {
      "auxiliary_loss_clip": 0.01320509,
      "auxiliary_loss_mlp": 0.01028457,
      "balance_loss_clip": 1.0450412,
      "balance_loss_mlp": 1.02128959,
      "epoch": 0.7121986412553358,
      "flos": 26359114700160.0,
      "grad_norm": 1.9603917490672167,
      "language_loss": 0.65150213,
      "learning_rate": 8.076044496592134e-07,
      "loss": 0.67499185,
      "num_input_tokens_seen": 127334730,
      "step": 5923,
      "time_per_iteration": 2.7760169506073
    },
    {
      "auxiliary_loss_clip": 0.01269474,
      "auxiliary_loss_mlp": 0.01027901,
      "balance_loss_clip": 1.04820657,
      "balance_loss_mlp": 1.02071881,
      "epoch": 0.7123188841459749,
      "flos": 11145344371200.0,
      "grad_norm": 2.2401373840914354,
      "language_loss": 0.77907121,
      "learning_rate": 8.069791523895204e-07,
      "loss": 0.80204493,
      "num_input_tokens_seen": 127351180,
      "step": 5924,
      "time_per_iteration": 2.631726026535034
    },
    {
      "auxiliary_loss_clip": 0.01317402,
      "auxiliary_loss_mlp": 0.01021681,
      "balance_loss_clip": 1.04248452,
      "balance_loss_mlp": 1.01441252,
      "epoch": 0.7124391270366139,
      "flos": 20811670329600.0,
      "grad_norm": 3.2375368002383085,
      "language_loss": 0.77423716,
      "learning_rate": 8.063540361001422e-07,
      "loss": 0.79762793,
      "num_input_tokens_seen": 127369750,
      "step": 5925,
      "time_per_iteration": 2.699613571166992
    },
    {
      "auxiliary_loss_clip": 0.01322594,
      "auxiliary_loss_mlp": 0.01029894,
      "balance_loss_clip": 1.04606938,
      "balance_loss_mlp": 1.02278566,
      "epoch": 0.7125593699272531,
      "flos": 17603734584960.0,
      "grad_norm": 2.2425855591580346,
      "language_loss": 0.79385531,
      "learning_rate": 8.057291008859069e-07,
      "loss": 0.81738019,
      "num_input_tokens_seen": 127387910,
      "step": 5926,
      "time_per_iteration": 2.678907632827759
    },
    {
      "auxiliary_loss_clip": 0.01222615,
      "auxiliary_loss_mlp": 0.0102522,
      "balance_loss_clip": 1.04708195,
      "balance_loss_mlp": 1.01852298,
      "epoch": 0.7126796128178922,
      "flos": 28654057526400.0,
      "grad_norm": 2.125759287721424,
      "language_loss": 0.67951298,
      "learning_rate": 8.051043468416187e-07,
      "loss": 0.70199132,
      "num_input_tokens_seen": 127409160,
      "step": 5927,
      "time_per_iteration": 2.702498197555542
    },
    {
      "auxiliary_loss_clip": 0.01174179,
      "auxiliary_loss_mlp": 0.01025043,
      "balance_loss_clip": 1.0511322,
      "balance_loss_mlp": 1.0185883,
      "epoch": 0.7127998557085312,
      "flos": 16034438315520.0,
      "grad_norm": 2.107843607315672,
      "language_loss": 0.82290876,
      "learning_rate": 8.044797740620506e-07,
      "loss": 0.84490097,
      "num_input_tokens_seen": 127427765,
      "step": 5928,
      "time_per_iteration": 2.580397605895996
    },
    {
      "auxiliary_loss_clip": 0.01366863,
      "auxiliary_loss_mlp": 0.0102363,
      "balance_loss_clip": 1.04454243,
      "balance_loss_mlp": 1.01689208,
      "epoch": 0.7129200985991703,
      "flos": 23403271582080.0,
      "grad_norm": 2.8427138173251563,
      "language_loss": 0.7874676,
      "learning_rate": 8.038553826419494e-07,
      "loss": 0.81137258,
      "num_input_tokens_seen": 127446475,
      "step": 5929,
      "time_per_iteration": 2.8006014823913574
    },
    {
      "auxiliary_loss_clip": 0.01172723,
      "auxiliary_loss_mlp": 0.0102589,
      "balance_loss_clip": 1.04830813,
      "balance_loss_mlp": 1.01907134,
      "epoch": 0.7130403414898094,
      "flos": 21397445326080.0,
      "grad_norm": 1.6287573381659903,
      "language_loss": 0.81183672,
      "learning_rate": 8.032311726760364e-07,
      "loss": 0.83382285,
      "num_input_tokens_seen": 127467695,
      "step": 5930,
      "time_per_iteration": 2.6343538761138916
    },
    {
      "auxiliary_loss_clip": 0.01317516,
      "auxiliary_loss_mlp": 0.01027348,
      "balance_loss_clip": 1.0440352,
      "balance_loss_mlp": 1.02027631,
      "epoch": 0.7131605843804485,
      "flos": 74739045306240.0,
      "grad_norm": 2.3890919377286144,
      "language_loss": 0.69454497,
      "learning_rate": 8.026071442590022e-07,
      "loss": 0.71799362,
      "num_input_tokens_seen": 127494590,
      "step": 5931,
      "time_per_iteration": 3.1084680557250977
    },
    {
      "auxiliary_loss_clip": 0.01224561,
      "auxiliary_loss_mlp": 0.0102357,
      "balance_loss_clip": 1.05068874,
      "balance_loss_mlp": 1.01695347,
      "epoch": 0.7132808272710875,
      "flos": 18368739469440.0,
      "grad_norm": 1.9680143893574589,
      "language_loss": 0.80761534,
      "learning_rate": 8.019832974855134e-07,
      "loss": 0.83009666,
      "num_input_tokens_seen": 127512550,
      "step": 5932,
      "time_per_iteration": 3.658043384552002
    },
    {
      "auxiliary_loss_clip": 0.01321111,
      "auxiliary_loss_mlp": 0.01024156,
      "balance_loss_clip": 1.04479337,
      "balance_loss_mlp": 1.01709855,
      "epoch": 0.7134010701617267,
      "flos": 23253380127360.0,
      "grad_norm": 2.4775047421512704,
      "language_loss": 0.82444888,
      "learning_rate": 8.013596324502052e-07,
      "loss": 0.84790158,
      "num_input_tokens_seen": 127531015,
      "step": 5933,
      "time_per_iteration": 3.603930950164795
    },
    {
      "auxiliary_loss_clip": 0.01218262,
      "auxiliary_loss_mlp": 0.01024499,
      "balance_loss_clip": 1.04994452,
      "balance_loss_mlp": 1.01798987,
      "epoch": 0.7135213130523658,
      "flos": 23653137565440.0,
      "grad_norm": 1.8347346256581167,
      "language_loss": 0.78748327,
      "learning_rate": 8.007361492476872e-07,
      "loss": 0.80991089,
      "num_input_tokens_seen": 127550340,
      "step": 5934,
      "time_per_iteration": 2.631948947906494
    },
    {
      "auxiliary_loss_clip": 0.01331773,
      "auxiliary_loss_mlp": 0.01030467,
      "balance_loss_clip": 1.04617083,
      "balance_loss_mlp": 1.02304339,
      "epoch": 0.7136415559430048,
      "flos": 24790644443520.0,
      "grad_norm": 1.597879719582589,
      "language_loss": 0.78872013,
      "learning_rate": 8.001128479725426e-07,
      "loss": 0.81234258,
      "num_input_tokens_seen": 127572245,
      "step": 5935,
      "time_per_iteration": 3.6226654052734375
    },
    {
      "auxiliary_loss_clip": 0.01367498,
      "auxiliary_loss_mlp": 0.0102685,
      "balance_loss_clip": 1.04169726,
      "balance_loss_mlp": 1.01960802,
      "epoch": 0.713761798833644,
      "flos": 18296954138880.0,
      "grad_norm": 1.650509980079547,
      "language_loss": 0.80912769,
      "learning_rate": 7.994897287193248e-07,
      "loss": 0.83307111,
      "num_input_tokens_seen": 127591625,
      "step": 5936,
      "time_per_iteration": 2.732728958129883
    },
    {
      "auxiliary_loss_clip": 0.01227211,
      "auxiliary_loss_mlp": 0.01023762,
      "balance_loss_clip": 1.04800332,
      "balance_loss_mlp": 1.01657391,
      "epoch": 0.713882041724283,
      "flos": 15558262692480.0,
      "grad_norm": 2.6454628759522256,
      "language_loss": 0.83761537,
      "learning_rate": 7.988667915825605e-07,
      "loss": 0.86012506,
      "num_input_tokens_seen": 127608690,
      "step": 5937,
      "time_per_iteration": 2.610414981842041
    },
    {
      "auxiliary_loss_clip": 0.01276656,
      "auxiliary_loss_mlp": 0.01021173,
      "balance_loss_clip": 1.04766917,
      "balance_loss_mlp": 1.01397276,
      "epoch": 0.7140022846149221,
      "flos": 24061011477120.0,
      "grad_norm": 2.0748590171738988,
      "language_loss": 0.7551223,
      "learning_rate": 7.982440366567491e-07,
      "loss": 0.77810055,
      "num_input_tokens_seen": 127627180,
      "step": 5938,
      "time_per_iteration": 2.7322845458984375
    },
    {
      "auxiliary_loss_clip": 0.01218808,
      "auxiliary_loss_mlp": 0.01022556,
      "balance_loss_clip": 1.04523039,
      "balance_loss_mlp": 1.01621675,
      "epoch": 0.7141225275055613,
      "flos": 27891710248320.0,
      "grad_norm": 1.5944601146856607,
      "language_loss": 0.7532748,
      "learning_rate": 7.97621464036361e-07,
      "loss": 0.77568847,
      "num_input_tokens_seen": 127648940,
      "step": 5939,
      "time_per_iteration": 2.694451093673706
    },
    {
      "auxiliary_loss_clip": 0.01225487,
      "auxiliary_loss_mlp": 0.0102502,
      "balance_loss_clip": 1.04733813,
      "balance_loss_mlp": 1.01823735,
      "epoch": 0.7142427703962003,
      "flos": 19682603147520.0,
      "grad_norm": 2.416100178911936,
      "language_loss": 0.68209934,
      "learning_rate": 7.969990738158417e-07,
      "loss": 0.70460451,
      "num_input_tokens_seen": 127667350,
      "step": 5940,
      "time_per_iteration": 3.570748805999756
    },
    {
      "auxiliary_loss_clip": 0.01228686,
      "auxiliary_loss_mlp": 0.01022474,
      "balance_loss_clip": 1.05257058,
      "balance_loss_mlp": 1.01564074,
      "epoch": 0.7143630132868394,
      "flos": 21032377447680.0,
      "grad_norm": 2.274189338541866,
      "language_loss": 0.85477501,
      "learning_rate": 7.963768660896062e-07,
      "loss": 0.87728661,
      "num_input_tokens_seen": 127685760,
      "step": 5941,
      "time_per_iteration": 2.610940456390381
    },
    {
      "auxiliary_loss_clip": 0.01227419,
      "auxiliary_loss_mlp": 0.01023453,
      "balance_loss_clip": 1.04749537,
      "balance_loss_mlp": 1.01642513,
      "epoch": 0.7144832561774785,
      "flos": 24129923719680.0,
      "grad_norm": 1.9399512176288194,
      "language_loss": 0.82316935,
      "learning_rate": 7.957548409520432e-07,
      "loss": 0.84567809,
      "num_input_tokens_seen": 127704985,
      "step": 5942,
      "time_per_iteration": 2.6751492023468018
    },
    {
      "auxiliary_loss_clip": 0.01320948,
      "auxiliary_loss_mlp": 0.01029083,
      "balance_loss_clip": 1.04495943,
      "balance_loss_mlp": 1.02237487,
      "epoch": 0.7146034990681176,
      "flos": 16325817442560.0,
      "grad_norm": 2.079056778379464,
      "language_loss": 0.84201419,
      "learning_rate": 7.951329984975135e-07,
      "loss": 0.86551452,
      "num_input_tokens_seen": 127721925,
      "step": 5943,
      "time_per_iteration": 2.6579058170318604
    },
    {
      "auxiliary_loss_clip": 0.01134359,
      "auxiliary_loss_mlp": 0.00999721,
      "balance_loss_clip": 1.01277077,
      "balance_loss_mlp": 0.9988147,
      "epoch": 0.7147237419587567,
      "flos": 69627164232960.0,
      "grad_norm": 0.717422328040634,
      "language_loss": 0.54193568,
      "learning_rate": 7.94511338820349e-07,
      "loss": 0.56327653,
      "num_input_tokens_seen": 127784230,
      "step": 5944,
      "time_per_iteration": 3.234156370162964
    },
    {
      "auxiliary_loss_clip": 0.01272179,
      "auxiliary_loss_mlp": 0.02569328,
      "balance_loss_clip": 1.04697108,
      "balance_loss_mlp": 0.99989027,
      "epoch": 0.7148439848493958,
      "flos": 22266806198400.0,
      "grad_norm": 2.7422190069208683,
      "language_loss": 0.78537619,
      "learning_rate": 7.938898620148575e-07,
      "loss": 0.82379121,
      "num_input_tokens_seen": 127801990,
      "step": 5945,
      "time_per_iteration": 2.749943733215332
    },
    {
      "auxiliary_loss_clip": 0.01271604,
      "auxiliary_loss_mlp": 0.01028664,
      "balance_loss_clip": 1.04662585,
      "balance_loss_mlp": 1.02141011,
      "epoch": 0.7149642277400349,
      "flos": 17931383470080.0,
      "grad_norm": 2.000003783717063,
      "language_loss": 0.70726603,
      "learning_rate": 7.932685681753135e-07,
      "loss": 0.73026872,
      "num_input_tokens_seen": 127819270,
      "step": 5946,
      "time_per_iteration": 2.6436381340026855
    },
    {
      "auxiliary_loss_clip": 0.01170381,
      "auxiliary_loss_mlp": 0.01021628,
      "balance_loss_clip": 1.0485363,
      "balance_loss_mlp": 1.01484156,
      "epoch": 0.7150844706306739,
      "flos": 31681937370240.0,
      "grad_norm": 5.647116235347149,
      "language_loss": 0.62561107,
      "learning_rate": 7.92647457395969e-07,
      "loss": 0.64753115,
      "num_input_tokens_seen": 127841095,
      "step": 5947,
      "time_per_iteration": 2.7194204330444336
    },
    {
      "auxiliary_loss_clip": 0.0143181,
      "auxiliary_loss_mlp": 0.01028328,
      "balance_loss_clip": 1.03915191,
      "balance_loss_mlp": 1.02127671,
      "epoch": 0.7152047135213131,
      "flos": 10926217451520.0,
      "grad_norm": 2.9939644464040103,
      "language_loss": 0.74414921,
      "learning_rate": 7.920265297710444e-07,
      "loss": 0.76875067,
      "num_input_tokens_seen": 127858485,
      "step": 5948,
      "time_per_iteration": 2.7600209712982178
    },
    {
      "auxiliary_loss_clip": 0.01227786,
      "auxiliary_loss_mlp": 0.01023669,
      "balance_loss_clip": 1.05048299,
      "balance_loss_mlp": 1.01753879,
      "epoch": 0.7153249564119522,
      "flos": 20995640812800.0,
      "grad_norm": 1.8328910417134827,
      "language_loss": 0.73509967,
      "learning_rate": 7.914057853947363e-07,
      "loss": 0.75761425,
      "num_input_tokens_seen": 127877665,
      "step": 5949,
      "time_per_iteration": 2.6229407787323
    },
    {
      "auxiliary_loss_clip": 0.01329563,
      "auxiliary_loss_mlp": 0.01031669,
      "balance_loss_clip": 1.04683292,
      "balance_loss_mlp": 1.02397084,
      "epoch": 0.7154451993025912,
      "flos": 24243114453120.0,
      "grad_norm": 2.127765499753154,
      "language_loss": 0.62666494,
      "learning_rate": 7.907852243612089e-07,
      "loss": 0.65027726,
      "num_input_tokens_seen": 127898070,
      "step": 5950,
      "time_per_iteration": 2.7219955921173096
    },
    {
      "auxiliary_loss_clip": 0.01270592,
      "auxiliary_loss_mlp": 0.01024027,
      "balance_loss_clip": 1.0448885,
      "balance_loss_mlp": 1.01705027,
      "epoch": 0.7155654421932304,
      "flos": 23330947547520.0,
      "grad_norm": 2.27453548970221,
      "language_loss": 0.7218321,
      "learning_rate": 7.901648467646009e-07,
      "loss": 0.74477828,
      "num_input_tokens_seen": 127917010,
      "step": 5951,
      "time_per_iteration": 2.675102949142456
    },
    {
      "auxiliary_loss_clip": 0.01177647,
      "auxiliary_loss_mlp": 0.01021786,
      "balance_loss_clip": 1.0509057,
      "balance_loss_mlp": 1.01514268,
      "epoch": 0.7156856850838694,
      "flos": 22711883621760.0,
      "grad_norm": 2.818612924499078,
      "language_loss": 0.724944,
      "learning_rate": 7.895446526990244e-07,
      "loss": 0.74693835,
      "num_input_tokens_seen": 127937025,
      "step": 5952,
      "time_per_iteration": 2.5880191326141357
    },
    {
      "auxiliary_loss_clip": 0.01380695,
      "auxiliary_loss_mlp": 0.01024068,
      "balance_loss_clip": 1.04362071,
      "balance_loss_mlp": 1.01715994,
      "epoch": 0.7158059279745085,
      "flos": 19865424395520.0,
      "grad_norm": 1.9441009464373546,
      "language_loss": 0.75626659,
      "learning_rate": 7.889246422585609e-07,
      "loss": 0.78031421,
      "num_input_tokens_seen": 127956410,
      "step": 5953,
      "time_per_iteration": 2.7732906341552734
    },
    {
      "auxiliary_loss_clip": 0.01176891,
      "auxiliary_loss_mlp": 0.01021799,
      "balance_loss_clip": 1.05130053,
      "balance_loss_mlp": 1.01493573,
      "epoch": 0.7159261708651476,
      "flos": 24134772055680.0,
      "grad_norm": 1.7459553495804907,
      "language_loss": 0.73586494,
      "learning_rate": 7.883048155372675e-07,
      "loss": 0.75785184,
      "num_input_tokens_seen": 127974925,
      "step": 5954,
      "time_per_iteration": 2.577468156814575
    },
    {
      "auxiliary_loss_clip": 0.01280607,
      "auxiliary_loss_mlp": 0.01025438,
      "balance_loss_clip": 1.04929709,
      "balance_loss_mlp": 1.01852083,
      "epoch": 0.7160464137557867,
      "flos": 16983198201600.0,
      "grad_norm": 2.5687882504027466,
      "language_loss": 0.70974058,
      "learning_rate": 7.876851726291698e-07,
      "loss": 0.73280108,
      "num_input_tokens_seen": 127993225,
      "step": 5955,
      "time_per_iteration": 2.675692319869995
    },
    {
      "auxiliary_loss_clip": 0.01327052,
      "auxiliary_loss_mlp": 0.01021276,
      "balance_loss_clip": 1.04401278,
      "balance_loss_mlp": 1.01423073,
      "epoch": 0.7161666566464258,
      "flos": 25228251838080.0,
      "grad_norm": 2.4573865318422836,
      "language_loss": 0.78457737,
      "learning_rate": 7.870657136282666e-07,
      "loss": 0.80806065,
      "num_input_tokens_seen": 128012085,
      "step": 5956,
      "time_per_iteration": 2.6881115436553955
    },
    {
      "auxiliary_loss_clip": 0.01219711,
      "auxiliary_loss_mlp": 0.0103229,
      "balance_loss_clip": 1.04607081,
      "balance_loss_mlp": 1.02546823,
      "epoch": 0.7162868995370649,
      "flos": 26468390851200.0,
      "grad_norm": 1.9421507389846253,
      "language_loss": 0.82052428,
      "learning_rate": 7.86446438628531e-07,
      "loss": 0.84304428,
      "num_input_tokens_seen": 128033155,
      "step": 5957,
      "time_per_iteration": 2.663004159927368
    },
    {
      "auxiliary_loss_clip": 0.01068074,
      "auxiliary_loss_mlp": 0.009995,
      "balance_loss_clip": 1.01181316,
      "balance_loss_mlp": 0.99860042,
      "epoch": 0.716407142427704,
      "flos": 69998912040960.0,
      "grad_norm": 0.7594398194094182,
      "language_loss": 0.56817383,
      "learning_rate": 7.858273477239059e-07,
      "loss": 0.58884954,
      "num_input_tokens_seen": 128101575,
      "step": 5958,
      "time_per_iteration": 4.166310548782349
    },
    {
      "auxiliary_loss_clip": 0.01363642,
      "auxiliary_loss_mlp": 0.01026713,
      "balance_loss_clip": 1.04090095,
      "balance_loss_mlp": 1.01991534,
      "epoch": 0.716527385318343,
      "flos": 20740459616640.0,
      "grad_norm": 1.7392622404063502,
      "language_loss": 0.71447396,
      "learning_rate": 7.852084410083067e-07,
      "loss": 0.73837757,
      "num_input_tokens_seen": 128120395,
      "step": 5959,
      "time_per_iteration": 3.636399269104004
    },
    {
      "auxiliary_loss_clip": 0.01267221,
      "auxiliary_loss_mlp": 0.01025144,
      "balance_loss_clip": 1.04709852,
      "balance_loss_mlp": 1.01870108,
      "epoch": 0.7166476282089821,
      "flos": 25371966153600.0,
      "grad_norm": 1.6261757017706762,
      "language_loss": 0.63558489,
      "learning_rate": 7.84589718575621e-07,
      "loss": 0.65850854,
      "num_input_tokens_seen": 128140840,
      "step": 5960,
      "time_per_iteration": 3.6102259159088135
    },
    {
      "auxiliary_loss_clip": 0.01274228,
      "auxiliary_loss_mlp": 0.01023984,
      "balance_loss_clip": 1.04291427,
      "balance_loss_mlp": 1.01705503,
      "epoch": 0.7167678710996213,
      "flos": 24133730561280.0,
      "grad_norm": 2.2149782862988254,
      "language_loss": 0.69107431,
      "learning_rate": 7.83971180519708e-07,
      "loss": 0.71405637,
      "num_input_tokens_seen": 128159695,
      "step": 5961,
      "time_per_iteration": 2.7111892700195312
    },
    {
      "auxiliary_loss_clip": 0.01176694,
      "auxiliary_loss_mlp": 0.01023589,
      "balance_loss_clip": 1.05069208,
      "balance_loss_mlp": 1.01636744,
      "epoch": 0.7168881139902603,
      "flos": 30226586019840.0,
      "grad_norm": 2.0389951113073392,
      "language_loss": 0.75469208,
      "learning_rate": 7.833528269344008e-07,
      "loss": 0.77669489,
      "num_input_tokens_seen": 128179600,
      "step": 5962,
      "time_per_iteration": 2.6802830696105957
    },
    {
      "auxiliary_loss_clip": 0.01329029,
      "auxiliary_loss_mlp": 0.01030978,
      "balance_loss_clip": 1.04858136,
      "balance_loss_mlp": 1.02349138,
      "epoch": 0.7170083568808994,
      "flos": 14606414236800.0,
      "grad_norm": 2.2426438022568203,
      "language_loss": 0.77649409,
      "learning_rate": 7.827346579135023e-07,
      "loss": 0.80009419,
      "num_input_tokens_seen": 128196940,
      "step": 5963,
      "time_per_iteration": 2.667869806289673
    },
    {
      "auxiliary_loss_clip": 0.01271297,
      "auxiliary_loss_mlp": 0.01022303,
      "balance_loss_clip": 1.04659581,
      "balance_loss_mlp": 1.01481962,
      "epoch": 0.7171285997715385,
      "flos": 23331091201920.0,
      "grad_norm": 2.0002164004919387,
      "language_loss": 0.8365016,
      "learning_rate": 7.821166735507885e-07,
      "loss": 0.85943758,
      "num_input_tokens_seen": 128215970,
      "step": 5964,
      "time_per_iteration": 2.6534228324890137
    },
    {
      "auxiliary_loss_clip": 0.01173329,
      "auxiliary_loss_mlp": 0.01025969,
      "balance_loss_clip": 1.05089998,
      "balance_loss_mlp": 1.01967788,
      "epoch": 0.7172488426621776,
      "flos": 16543543731840.0,
      "grad_norm": 4.303914518065851,
      "language_loss": 0.68709868,
      "learning_rate": 7.81498873940007e-07,
      "loss": 0.70909166,
      "num_input_tokens_seen": 128233185,
      "step": 5965,
      "time_per_iteration": 2.5598297119140625
    },
    {
      "auxiliary_loss_clip": 0.01230347,
      "auxiliary_loss_mlp": 0.01027479,
      "balance_loss_clip": 1.04748404,
      "balance_loss_mlp": 1.02030015,
      "epoch": 0.7173690855528166,
      "flos": 26541612725760.0,
      "grad_norm": 2.2033915052461066,
      "language_loss": 0.77378464,
      "learning_rate": 7.808812591748768e-07,
      "loss": 0.79636288,
      "num_input_tokens_seen": 128253565,
      "step": 5966,
      "time_per_iteration": 3.6012301445007324
    },
    {
      "auxiliary_loss_clip": 0.01322007,
      "auxiliary_loss_mlp": 0.01033613,
      "balance_loss_clip": 1.04530609,
      "balance_loss_mlp": 1.02600098,
      "epoch": 0.7174893284434558,
      "flos": 22784099915520.0,
      "grad_norm": 2.418817509813332,
      "language_loss": 0.64861202,
      "learning_rate": 7.802638293490915e-07,
      "loss": 0.6721682,
      "num_input_tokens_seen": 128273210,
      "step": 5967,
      "time_per_iteration": 2.6807761192321777
    },
    {
      "auxiliary_loss_clip": 0.01278496,
      "auxiliary_loss_mlp": 0.01024808,
      "balance_loss_clip": 1.04761791,
      "balance_loss_mlp": 1.017658,
      "epoch": 0.7176095713340949,
      "flos": 23293564467840.0,
      "grad_norm": 1.6252609366954602,
      "language_loss": 0.76927984,
      "learning_rate": 7.796465845563123e-07,
      "loss": 0.79231286,
      "num_input_tokens_seen": 128292085,
      "step": 5968,
      "time_per_iteration": 2.7058422565460205
    },
    {
      "auxiliary_loss_clip": 0.01269733,
      "auxiliary_loss_mlp": 0.02564834,
      "balance_loss_clip": 1.04584503,
      "balance_loss_mlp": 0.99988723,
      "epoch": 0.7177298142247339,
      "flos": 25591631777280.0,
      "grad_norm": 3.209063205505246,
      "language_loss": 0.79618001,
      "learning_rate": 7.790295248901766e-07,
      "loss": 0.83452564,
      "num_input_tokens_seen": 128313215,
      "step": 5969,
      "time_per_iteration": 2.6915714740753174
    },
    {
      "auxiliary_loss_clip": 0.01226691,
      "auxiliary_loss_mlp": 0.0102845,
      "balance_loss_clip": 1.04946518,
      "balance_loss_mlp": 1.02152109,
      "epoch": 0.7178500571153731,
      "flos": 31652778504960.0,
      "grad_norm": 1.635887264156226,
      "language_loss": 0.62343508,
      "learning_rate": 7.784126504442902e-07,
      "loss": 0.6459865,
      "num_input_tokens_seen": 128336445,
      "step": 5970,
      "time_per_iteration": 2.7545413970947266
    },
    {
      "auxiliary_loss_clip": 0.01318576,
      "auxiliary_loss_mlp": 0.01023689,
      "balance_loss_clip": 1.04623652,
      "balance_loss_mlp": 1.01661706,
      "epoch": 0.7179703000060121,
      "flos": 19427242383360.0,
      "grad_norm": 1.4323684682182403,
      "language_loss": 0.67919981,
      "learning_rate": 7.777959613122351e-07,
      "loss": 0.70262241,
      "num_input_tokens_seen": 128356270,
      "step": 5971,
      "time_per_iteration": 2.6851065158843994
    },
    {
      "auxiliary_loss_clip": 0.01269065,
      "auxiliary_loss_mlp": 0.01026389,
      "balance_loss_clip": 1.04899311,
      "balance_loss_mlp": 1.0194478,
      "epoch": 0.7180905428966512,
      "flos": 28839249072000.0,
      "grad_norm": 1.873458220411954,
      "language_loss": 0.77786887,
      "learning_rate": 7.771794575875604e-07,
      "loss": 0.80082345,
      "num_input_tokens_seen": 128378140,
      "step": 5972,
      "time_per_iteration": 2.728510618209839
    },
    {
      "auxiliary_loss_clip": 0.01232679,
      "auxiliary_loss_mlp": 0.01024181,
      "balance_loss_clip": 1.05356789,
      "balance_loss_mlp": 1.01699543,
      "epoch": 0.7182107857872904,
      "flos": 20047563285120.0,
      "grad_norm": 2.4366105762533796,
      "language_loss": 0.7756511,
      "learning_rate": 7.765631393637888e-07,
      "loss": 0.79821968,
      "num_input_tokens_seen": 128396335,
      "step": 5973,
      "time_per_iteration": 2.729074239730835
    },
    {
      "auxiliary_loss_clip": 0.01225077,
      "auxiliary_loss_mlp": 0.01027087,
      "balance_loss_clip": 1.04607749,
      "balance_loss_mlp": 1.01949608,
      "epoch": 0.7183310286779294,
      "flos": 22747686503040.0,
      "grad_norm": 3.1222308665172536,
      "language_loss": 0.48938826,
      "learning_rate": 7.75947006734417e-07,
      "loss": 0.5119099,
      "num_input_tokens_seen": 128414115,
      "step": 5974,
      "time_per_iteration": 2.6543447971343994
    },
    {
      "auxiliary_loss_clip": 0.0117362,
      "auxiliary_loss_mlp": 0.01027333,
      "balance_loss_clip": 1.04781795,
      "balance_loss_mlp": 1.02039206,
      "epoch": 0.7184512715685685,
      "flos": 17158262112000.0,
      "grad_norm": 2.125103450868867,
      "language_loss": 0.83101332,
      "learning_rate": 7.753310597929101e-07,
      "loss": 0.85302281,
      "num_input_tokens_seen": 128430755,
      "step": 5975,
      "time_per_iteration": 2.4853689670562744
    },
    {
      "auxiliary_loss_clip": 0.01068327,
      "auxiliary_loss_mlp": 0.01001124,
      "balance_loss_clip": 1.01197541,
      "balance_loss_mlp": 1.00019383,
      "epoch": 0.7185715144592076,
      "flos": 65509611448320.0,
      "grad_norm": 0.7511514293808181,
      "language_loss": 0.55066073,
      "learning_rate": 7.747152986327095e-07,
      "loss": 0.57135528,
      "num_input_tokens_seen": 128491300,
      "step": 5976,
      "time_per_iteration": 3.073885440826416
    },
    {
      "auxiliary_loss_clip": 0.01375974,
      "auxiliary_loss_mlp": 0.01026834,
      "balance_loss_clip": 1.04368091,
      "balance_loss_mlp": 1.02061725,
      "epoch": 0.7186917573498467,
      "flos": 16180522928640.0,
      "grad_norm": 2.5194239381993757,
      "language_loss": 0.68383974,
      "learning_rate": 7.740997233472228e-07,
      "loss": 0.7078678,
      "num_input_tokens_seen": 128508920,
      "step": 5977,
      "time_per_iteration": 2.6963229179382324
    },
    {
      "auxiliary_loss_clip": 0.01273082,
      "auxiliary_loss_mlp": 0.0102926,
      "balance_loss_clip": 1.04577327,
      "balance_loss_mlp": 1.02241182,
      "epoch": 0.7188120002404857,
      "flos": 29242274647680.0,
      "grad_norm": 3.119503678804544,
      "language_loss": 0.70388269,
      "learning_rate": 7.734843340298329e-07,
      "loss": 0.72690618,
      "num_input_tokens_seen": 128528745,
      "step": 5978,
      "time_per_iteration": 2.694007396697998
    },
    {
      "auxiliary_loss_clip": 0.01278156,
      "auxiliary_loss_mlp": 0.01026387,
      "balance_loss_clip": 1.04594862,
      "balance_loss_mlp": 1.0187459,
      "epoch": 0.7189322431311249,
      "flos": 33401161008000.0,
      "grad_norm": 2.686605695577765,
      "language_loss": 0.75281823,
      "learning_rate": 7.72869130773895e-07,
      "loss": 0.77586371,
      "num_input_tokens_seen": 128549345,
      "step": 5979,
      "time_per_iteration": 2.7841813564300537
    },
    {
      "auxiliary_loss_clip": 0.0112408,
      "auxiliary_loss_mlp": 0.00999788,
      "balance_loss_clip": 1.01234317,
      "balance_loss_mlp": 0.99887657,
      "epoch": 0.719052486021764,
      "flos": 61351263792000.0,
      "grad_norm": 0.8025263532660742,
      "language_loss": 0.59317547,
      "learning_rate": 7.722541136727343e-07,
      "loss": 0.61441416,
      "num_input_tokens_seen": 128605360,
      "step": 5980,
      "time_per_iteration": 3.0865020751953125
    },
    {
      "auxiliary_loss_clip": 0.01223695,
      "auxiliary_loss_mlp": 0.01023719,
      "balance_loss_clip": 1.04882455,
      "balance_loss_mlp": 1.01680827,
      "epoch": 0.719172728912403,
      "flos": 15596795007360.0,
      "grad_norm": 2.0197958005920498,
      "language_loss": 0.80981112,
      "learning_rate": 7.716392828196483e-07,
      "loss": 0.83228528,
      "num_input_tokens_seen": 128623160,
      "step": 5981,
      "time_per_iteration": 2.595790147781372
    },
    {
      "auxiliary_loss_clip": 0.01227018,
      "auxiliary_loss_mlp": 0.01027411,
      "balance_loss_clip": 1.05071926,
      "balance_loss_mlp": 1.02009773,
      "epoch": 0.7192929718030422,
      "flos": 15553162961280.0,
      "grad_norm": 3.4109628413949395,
      "language_loss": 0.77507067,
      "learning_rate": 7.710246383079064e-07,
      "loss": 0.79761493,
      "num_input_tokens_seen": 128638545,
      "step": 5982,
      "time_per_iteration": 2.560617685317993
    },
    {
      "auxiliary_loss_clip": 0.0127563,
      "auxiliary_loss_mlp": 0.01023361,
      "balance_loss_clip": 1.0441258,
      "balance_loss_mlp": 1.01679802,
      "epoch": 0.7194132146936812,
      "flos": 21862487733120.0,
      "grad_norm": 2.7817768370409546,
      "language_loss": 0.91923529,
      "learning_rate": 7.704101802307492e-07,
      "loss": 0.94222522,
      "num_input_tokens_seen": 128650845,
      "step": 5983,
      "time_per_iteration": 3.6022841930389404
    },
    {
      "auxiliary_loss_clip": 0.01319316,
      "auxiliary_loss_mlp": 0.01024316,
      "balance_loss_clip": 1.04485703,
      "balance_loss_mlp": 1.01715744,
      "epoch": 0.7195334575843203,
      "flos": 27338900958720.0,
      "grad_norm": 2.588311236508762,
      "language_loss": 0.86954236,
      "learning_rate": 7.697959086813912e-07,
      "loss": 0.89297867,
      "num_input_tokens_seen": 128667010,
      "step": 5984,
      "time_per_iteration": 3.655559778213501
    },
    {
      "auxiliary_loss_clip": 0.01318985,
      "auxiliary_loss_mlp": 0.01025872,
      "balance_loss_clip": 1.04271889,
      "balance_loss_mlp": 1.01845074,
      "epoch": 0.7196537004749595,
      "flos": 18770615809920.0,
      "grad_norm": 1.612382373294819,
      "language_loss": 0.80168939,
      "learning_rate": 7.691818237530145e-07,
      "loss": 0.82513797,
      "num_input_tokens_seen": 128685870,
      "step": 5985,
      "time_per_iteration": 2.7146339416503906
    },
    {
      "auxiliary_loss_clip": 0.01384317,
      "auxiliary_loss_mlp": 0.01025723,
      "balance_loss_clip": 1.04402483,
      "balance_loss_mlp": 1.01862121,
      "epoch": 0.7197739433655985,
      "flos": 24531009960960.0,
      "grad_norm": 2.0522967178307403,
      "language_loss": 0.77399039,
      "learning_rate": 7.685679255387774e-07,
      "loss": 0.79809082,
      "num_input_tokens_seen": 128704185,
      "step": 5986,
      "time_per_iteration": 3.6164064407348633
    },
    {
      "auxiliary_loss_clip": 0.01271968,
      "auxiliary_loss_mlp": 0.01023247,
      "balance_loss_clip": 1.04774714,
      "balance_loss_mlp": 1.01636243,
      "epoch": 0.7198941862562376,
      "flos": 18040587793920.0,
      "grad_norm": 2.0546021639441916,
      "language_loss": 0.76674616,
      "learning_rate": 7.679542141318065e-07,
      "loss": 0.78969836,
      "num_input_tokens_seen": 128721290,
      "step": 5987,
      "time_per_iteration": 2.681095600128174
    },
    {
      "auxiliary_loss_clip": 0.01266916,
      "auxiliary_loss_mlp": 0.01030718,
      "balance_loss_clip": 1.04370379,
      "balance_loss_mlp": 1.023983,
      "epoch": 0.7200144291468767,
      "flos": 29022393542400.0,
      "grad_norm": 2.040784169741203,
      "language_loss": 0.75908172,
      "learning_rate": 7.673406896252013e-07,
      "loss": 0.78205812,
      "num_input_tokens_seen": 128742665,
      "step": 5988,
      "time_per_iteration": 2.6941769123077393
    },
    {
      "auxiliary_loss_clip": 0.01321212,
      "auxiliary_loss_mlp": 0.01028832,
      "balance_loss_clip": 1.04246581,
      "balance_loss_mlp": 1.02165854,
      "epoch": 0.7201346720375158,
      "flos": 25374264624000.0,
      "grad_norm": 1.71505792754764,
      "language_loss": 0.7861672,
      "learning_rate": 7.667273521120347e-07,
      "loss": 0.80966765,
      "num_input_tokens_seen": 128762225,
      "step": 5989,
      "time_per_iteration": 2.8818881511688232
    },
    {
      "auxiliary_loss_clip": 0.01322396,
      "auxiliary_loss_mlp": 0.01028611,
      "balance_loss_clip": 1.04449081,
      "balance_loss_mlp": 1.02152395,
      "epoch": 0.7202549149281549,
      "flos": 14355614499840.0,
      "grad_norm": 1.9483856827888342,
      "language_loss": 0.79475021,
      "learning_rate": 7.661142016853468e-07,
      "loss": 0.81826031,
      "num_input_tokens_seen": 128779585,
      "step": 5990,
      "time_per_iteration": 2.6534714698791504
    },
    {
      "auxiliary_loss_clip": 0.013717,
      "auxiliary_loss_mlp": 0.01026877,
      "balance_loss_clip": 1.04172254,
      "balance_loss_mlp": 1.01973653,
      "epoch": 0.7203751578187939,
      "flos": 23001682550400.0,
      "grad_norm": 1.7545426527856822,
      "language_loss": 0.74638903,
      "learning_rate": 7.655012384381543e-07,
      "loss": 0.77037477,
      "num_input_tokens_seen": 128799070,
      "step": 5991,
      "time_per_iteration": 2.736276865005493
    },
    {
      "auxiliary_loss_clip": 0.01271345,
      "auxiliary_loss_mlp": 0.01032041,
      "balance_loss_clip": 1.04905128,
      "balance_loss_mlp": 1.02491796,
      "epoch": 0.7204954007094331,
      "flos": 23692424065920.0,
      "grad_norm": 3.018069927766298,
      "language_loss": 0.81884313,
      "learning_rate": 7.648884624634415e-07,
      "loss": 0.84187692,
      "num_input_tokens_seen": 128817620,
      "step": 5992,
      "time_per_iteration": 3.615183115005493
    },
    {
      "auxiliary_loss_clip": 0.01220923,
      "auxiliary_loss_mlp": 0.01024041,
      "balance_loss_clip": 1.04861903,
      "balance_loss_mlp": 1.01735616,
      "epoch": 0.7206156436000721,
      "flos": 16253026531200.0,
      "grad_norm": 2.18952584536538,
      "language_loss": 0.89065409,
      "learning_rate": 7.642758738541683e-07,
      "loss": 0.9131037,
      "num_input_tokens_seen": 128834200,
      "step": 5993,
      "time_per_iteration": 2.601533889770508
    },
    {
      "auxiliary_loss_clip": 0.01122425,
      "auxiliary_loss_mlp": 0.0100395,
      "balance_loss_clip": 1.01312172,
      "balance_loss_mlp": 1.00302052,
      "epoch": 0.7207358864907112,
      "flos": 54377806504320.0,
      "grad_norm": 0.7683450160840272,
      "language_loss": 0.60732484,
      "learning_rate": 7.636634727032621e-07,
      "loss": 0.62858862,
      "num_input_tokens_seen": 128891305,
      "step": 5994,
      "time_per_iteration": 3.0593743324279785
    },
    {
      "auxiliary_loss_clip": 0.01329825,
      "auxiliary_loss_mlp": 0.01024858,
      "balance_loss_clip": 1.04311728,
      "balance_loss_mlp": 1.01724052,
      "epoch": 0.7208561293813504,
      "flos": 19135540033920.0,
      "grad_norm": 2.182834546857687,
      "language_loss": 0.78824651,
      "learning_rate": 7.630512591036231e-07,
      "loss": 0.81179339,
      "num_input_tokens_seen": 128910615,
      "step": 5995,
      "time_per_iteration": 2.7113571166992188
    },
    {
      "auxiliary_loss_clip": 0.01226195,
      "auxiliary_loss_mlp": 0.01023453,
      "balance_loss_clip": 1.05057502,
      "balance_loss_mlp": 1.0163331,
      "epoch": 0.7209763722719894,
      "flos": 17748526308480.0,
      "grad_norm": 2.788572976245571,
      "language_loss": 0.64824045,
      "learning_rate": 7.624392331481255e-07,
      "loss": 0.67073697,
      "num_input_tokens_seen": 128928270,
      "step": 5996,
      "time_per_iteration": 2.6183338165283203
    },
    {
      "auxiliary_loss_clip": 0.01119019,
      "auxiliary_loss_mlp": 0.0100283,
      "balance_loss_clip": 1.0115447,
      "balance_loss_mlp": 1.00183511,
      "epoch": 0.7210966151626285,
      "flos": 66819488716800.0,
      "grad_norm": 0.8185116723051613,
      "language_loss": 0.51847219,
      "learning_rate": 7.618273949296115e-07,
      "loss": 0.53969067,
      "num_input_tokens_seen": 128987780,
      "step": 5997,
      "time_per_iteration": 3.102205514907837
    },
    {
      "auxiliary_loss_clip": 0.01268307,
      "auxiliary_loss_mlp": 0.01025312,
      "balance_loss_clip": 1.04410684,
      "balance_loss_mlp": 1.01812053,
      "epoch": 0.7212168580532676,
      "flos": 21141869080320.0,
      "grad_norm": 3.406882180543949,
      "language_loss": 0.68548411,
      "learning_rate": 7.612157445408987e-07,
      "loss": 0.70842028,
      "num_input_tokens_seen": 129005590,
      "step": 5998,
      "time_per_iteration": 2.658825635910034
    },
    {
      "auxiliary_loss_clip": 0.01284839,
      "auxiliary_loss_mlp": 0.01031131,
      "balance_loss_clip": 1.05201733,
      "balance_loss_mlp": 1.02347803,
      "epoch": 0.7213371009439067,
      "flos": 22345738335360.0,
      "grad_norm": 2.083570441025999,
      "language_loss": 0.74188197,
      "learning_rate": 7.606042820747716e-07,
      "loss": 0.76504165,
      "num_input_tokens_seen": 129021995,
      "step": 5999,
      "time_per_iteration": 2.6026697158813477
    },
    {
      "auxiliary_loss_clip": 0.01284498,
      "auxiliary_loss_mlp": 0.01025318,
      "balance_loss_clip": 1.05320549,
      "balance_loss_mlp": 1.01795053,
      "epoch": 0.7214573438345457,
      "flos": 18515901490560.0,
      "grad_norm": 1.932398921831613,
      "language_loss": 0.85328257,
      "learning_rate": 7.599930076239889e-07,
      "loss": 0.87638068,
      "num_input_tokens_seen": 129039280,
      "step": 6000,
      "time_per_iteration": 2.6824817657470703
    },
    {
      "auxiliary_loss_clip": 0.01379413,
      "auxiliary_loss_mlp": 0.02566404,
      "balance_loss_clip": 1.04788232,
      "balance_loss_mlp": 0.99989331,
      "epoch": 0.7215775867251849,
      "flos": 35736108606720.0,
      "grad_norm": 1.9424135754922212,
      "language_loss": 0.70870423,
      "learning_rate": 7.593819212812818e-07,
      "loss": 0.74816239,
      "num_input_tokens_seen": 129060860,
      "step": 6001,
      "time_per_iteration": 2.86793851852417
    },
    {
      "auxiliary_loss_clip": 0.01222765,
      "auxiliary_loss_mlp": 0.01028587,
      "balance_loss_clip": 1.04885435,
      "balance_loss_mlp": 1.02157736,
      "epoch": 0.721697829615824,
      "flos": 20372410909440.0,
      "grad_norm": 2.0600584061111133,
      "language_loss": 0.71810937,
      "learning_rate": 7.587710231393508e-07,
      "loss": 0.74062288,
      "num_input_tokens_seen": 129079215,
      "step": 6002,
      "time_per_iteration": 2.6317508220672607
    },
    {
      "auxiliary_loss_clip": 0.01465584,
      "auxiliary_loss_mlp": 0.01026829,
      "balance_loss_clip": 1.04039013,
      "balance_loss_mlp": 1.0198729,
      "epoch": 0.721818072506463,
      "flos": 20229809915520.0,
      "grad_norm": 2.105942035367066,
      "language_loss": 0.8374818,
      "learning_rate": 7.581603132908685e-07,
      "loss": 0.8624059,
      "num_input_tokens_seen": 129097185,
      "step": 6003,
      "time_per_iteration": 2.8666701316833496
    },
    {
      "auxiliary_loss_clip": 0.01315322,
      "auxiliary_loss_mlp": 0.01025533,
      "balance_loss_clip": 1.0434618,
      "balance_loss_mlp": 1.01835704,
      "epoch": 0.7219383153971022,
      "flos": 18186887888640.0,
      "grad_norm": 2.018201871612926,
      "language_loss": 0.78438532,
      "learning_rate": 7.575497918284795e-07,
      "loss": 0.80779386,
      "num_input_tokens_seen": 129114730,
      "step": 6004,
      "time_per_iteration": 2.901970386505127
    },
    {
      "auxiliary_loss_clip": 0.01179523,
      "auxiliary_loss_mlp": 0.01032478,
      "balance_loss_clip": 1.04993534,
      "balance_loss_mlp": 1.0253253,
      "epoch": 0.7220585582877412,
      "flos": 17342124854400.0,
      "grad_norm": 2.3742287445896864,
      "language_loss": 0.74533558,
      "learning_rate": 7.569394588447984e-07,
      "loss": 0.76745558,
      "num_input_tokens_seen": 129131745,
      "step": 6005,
      "time_per_iteration": 2.5633316040039062
    },
    {
      "auxiliary_loss_clip": 0.01218745,
      "auxiliary_loss_mlp": 0.01024781,
      "balance_loss_clip": 1.04548168,
      "balance_loss_mlp": 1.01794767,
      "epoch": 0.7221788011783803,
      "flos": 16976338704000.0,
      "grad_norm": 2.5642413737107606,
      "language_loss": 0.77805853,
      "learning_rate": 7.563293144324146e-07,
      "loss": 0.80049372,
      "num_input_tokens_seen": 129147295,
      "step": 6006,
      "time_per_iteration": 2.5446665287017822
    },
    {
      "auxiliary_loss_clip": 0.01173818,
      "auxiliary_loss_mlp": 0.01025625,
      "balance_loss_clip": 1.05016828,
      "balance_loss_mlp": 1.01859772,
      "epoch": 0.7222990440690195,
      "flos": 26286359702400.0,
      "grad_norm": 2.5671355103698255,
      "language_loss": 0.80264354,
      "learning_rate": 7.557193586838834e-07,
      "loss": 0.82463795,
      "num_input_tokens_seen": 129162660,
      "step": 6007,
      "time_per_iteration": 2.6515095233917236
    },
    {
      "auxiliary_loss_clip": 0.01175493,
      "auxiliary_loss_mlp": 0.01024028,
      "balance_loss_clip": 1.04585195,
      "balance_loss_mlp": 1.0167737,
      "epoch": 0.7224192869596585,
      "flos": 17601687509760.0,
      "grad_norm": 2.2895862739696797,
      "language_loss": 0.70753241,
      "learning_rate": 7.551095916917371e-07,
      "loss": 0.72952753,
      "num_input_tokens_seen": 129179990,
      "step": 6008,
      "time_per_iteration": 2.640380620956421
    },
    {
      "auxiliary_loss_clip": 0.01233493,
      "auxiliary_loss_mlp": 0.01026905,
      "balance_loss_clip": 1.0453763,
      "balance_loss_mlp": 1.01903152,
      "epoch": 0.7225395298502976,
      "flos": 12932331016320.0,
      "grad_norm": 3.589529598952646,
      "language_loss": 0.66543555,
      "learning_rate": 7.545000135484758e-07,
      "loss": 0.68803954,
      "num_input_tokens_seen": 129197425,
      "step": 6009,
      "time_per_iteration": 3.741142749786377
    },
    {
      "auxiliary_loss_clip": 0.01176298,
      "auxiliary_loss_mlp": 0.02568139,
      "balance_loss_clip": 1.04978848,
      "balance_loss_mlp": 0.99988008,
      "epoch": 0.7226597727409367,
      "flos": 29643899592960.0,
      "grad_norm": 4.216731882571455,
      "language_loss": 0.6276564,
      "learning_rate": 7.538906243465714e-07,
      "loss": 0.66510081,
      "num_input_tokens_seen": 129217560,
      "step": 6010,
      "time_per_iteration": 3.5479612350463867
    },
    {
      "auxiliary_loss_clip": 0.011763,
      "auxiliary_loss_mlp": 0.01024516,
      "balance_loss_clip": 1.05130959,
      "balance_loss_mlp": 1.01737881,
      "epoch": 0.7227800156315758,
      "flos": 13771635183360.0,
      "grad_norm": 3.3152192774045237,
      "language_loss": 0.7844466,
      "learning_rate": 7.5328142417847e-07,
      "loss": 0.80645472,
      "num_input_tokens_seen": 129234325,
      "step": 6011,
      "time_per_iteration": 2.519859790802002
    },
    {
      "auxiliary_loss_clip": 0.012222,
      "auxiliary_loss_mlp": 0.01026774,
      "balance_loss_clip": 1.04552698,
      "balance_loss_mlp": 1.01996684,
      "epoch": 0.7229002585222148,
      "flos": 20301882554880.0,
      "grad_norm": 1.743762986854525,
      "language_loss": 0.69180226,
      "learning_rate": 7.526724131365838e-07,
      "loss": 0.71429193,
      "num_input_tokens_seen": 129255280,
      "step": 6012,
      "time_per_iteration": 3.4501099586486816
    },
    {
      "auxiliary_loss_clip": 0.01273776,
      "auxiliary_loss_mlp": 0.01027784,
      "balance_loss_clip": 1.0502224,
      "balance_loss_mlp": 1.02027988,
      "epoch": 0.723020501412854,
      "flos": 16581250033920.0,
      "grad_norm": 2.08814031914499,
      "language_loss": 0.70636976,
      "learning_rate": 7.520635913133017e-07,
      "loss": 0.72938538,
      "num_input_tokens_seen": 129273910,
      "step": 6013,
      "time_per_iteration": 2.6019370555877686
    },
    {
      "auxiliary_loss_clip": 0.01230729,
      "auxiliary_loss_mlp": 0.01022611,
      "balance_loss_clip": 1.04996765,
      "balance_loss_mlp": 1.01517189,
      "epoch": 0.7231407443034931,
      "flos": 28548300908160.0,
      "grad_norm": 1.9240050333631207,
      "language_loss": 0.82361829,
      "learning_rate": 7.514549588009798e-07,
      "loss": 0.84615165,
      "num_input_tokens_seen": 129294785,
      "step": 6014,
      "time_per_iteration": 2.6869900226593018
    },
    {
      "auxiliary_loss_clip": 0.01277929,
      "auxiliary_loss_mlp": 0.01025342,
      "balance_loss_clip": 1.04975665,
      "balance_loss_mlp": 1.01755762,
      "epoch": 0.7232609871941321,
      "flos": 30008536508160.0,
      "grad_norm": 2.228321915394622,
      "language_loss": 0.70842957,
      "learning_rate": 7.508465156919492e-07,
      "loss": 0.73146236,
      "num_input_tokens_seen": 129318295,
      "step": 6015,
      "time_per_iteration": 2.7669787406921387
    },
    {
      "auxiliary_loss_clip": 0.01272888,
      "auxiliary_loss_mlp": 0.01030041,
      "balance_loss_clip": 1.04453194,
      "balance_loss_mlp": 1.02278697,
      "epoch": 0.7233812300847713,
      "flos": 16654005031680.0,
      "grad_norm": 3.6572921041609017,
      "language_loss": 0.61109471,
      "learning_rate": 7.502382620785083e-07,
      "loss": 0.63412404,
      "num_input_tokens_seen": 129334845,
      "step": 6016,
      "time_per_iteration": 2.5856759548187256
    },
    {
      "auxiliary_loss_clip": 0.01219768,
      "auxiliary_loss_mlp": 0.01005721,
      "balance_loss_clip": 1.01037312,
      "balance_loss_mlp": 1.00475502,
      "epoch": 0.7235014729754103,
      "flos": 67258784050560.0,
      "grad_norm": 0.8076483368732894,
      "language_loss": 0.62521088,
      "learning_rate": 7.496301980529289e-07,
      "loss": 0.64746571,
      "num_input_tokens_seen": 129398055,
      "step": 6017,
      "time_per_iteration": 3.274460554122925
    },
    {
      "auxiliary_loss_clip": 0.01178152,
      "auxiliary_loss_mlp": 0.01027558,
      "balance_loss_clip": 1.0507828,
      "balance_loss_mlp": 1.02026832,
      "epoch": 0.7236217158660494,
      "flos": 26943237671040.0,
      "grad_norm": 2.8296425511068293,
      "language_loss": 0.74300975,
      "learning_rate": 7.490223237074547e-07,
      "loss": 0.76506686,
      "num_input_tokens_seen": 129417765,
      "step": 6018,
      "time_per_iteration": 3.7193727493286133
    },
    {
      "auxiliary_loss_clip": 0.01324859,
      "auxiliary_loss_mlp": 0.01023664,
      "balance_loss_clip": 1.04272532,
      "balance_loss_mlp": 1.01655316,
      "epoch": 0.7237419587566886,
      "flos": 29423372042880.0,
      "grad_norm": 1.8841978972129265,
      "language_loss": 0.66067874,
      "learning_rate": 7.484146391342989e-07,
      "loss": 0.68416399,
      "num_input_tokens_seen": 129437560,
      "step": 6019,
      "time_per_iteration": 2.8049747943878174
    },
    {
      "auxiliary_loss_clip": 0.01268938,
      "auxiliary_loss_mlp": 0.0102513,
      "balance_loss_clip": 1.04500711,
      "balance_loss_mlp": 1.01851726,
      "epoch": 0.7238622016473276,
      "flos": 17821496787840.0,
      "grad_norm": 2.203243027028387,
      "language_loss": 0.5709548,
      "learning_rate": 7.478071444256484e-07,
      "loss": 0.59389544,
      "num_input_tokens_seen": 129455320,
      "step": 6020,
      "time_per_iteration": 2.6695635318756104
    },
    {
      "auxiliary_loss_clip": 0.01234216,
      "auxiliary_loss_mlp": 0.01028021,
      "balance_loss_clip": 1.04678428,
      "balance_loss_mlp": 1.02076685,
      "epoch": 0.7239824445379667,
      "flos": 25739117020800.0,
      "grad_norm": 2.0552200859045913,
      "language_loss": 0.79293263,
      "learning_rate": 7.471998396736579e-07,
      "loss": 0.81555498,
      "num_input_tokens_seen": 129475700,
      "step": 6021,
      "time_per_iteration": 2.776374340057373
    },
    {
      "auxiliary_loss_clip": 0.01323269,
      "auxiliary_loss_mlp": 0.01029329,
      "balance_loss_clip": 1.04617357,
      "balance_loss_mlp": 1.02239156,
      "epoch": 0.7241026874286057,
      "flos": 23148916398720.0,
      "grad_norm": 1.8393289774064632,
      "language_loss": 0.7626003,
      "learning_rate": 7.465927249704549e-07,
      "loss": 0.78612626,
      "num_input_tokens_seen": 129493585,
      "step": 6022,
      "time_per_iteration": 2.8894524574279785
    },
    {
      "auxiliary_loss_clip": 0.01222066,
      "auxiliary_loss_mlp": 0.01024331,
      "balance_loss_clip": 1.04796731,
      "balance_loss_mlp": 1.01711321,
      "epoch": 0.7242229303192449,
      "flos": 20266905686400.0,
      "grad_norm": 1.969001359642246,
      "language_loss": 0.77522135,
      "learning_rate": 7.459858004081398e-07,
      "loss": 0.79768533,
      "num_input_tokens_seen": 129511555,
      "step": 6023,
      "time_per_iteration": 2.6393566131591797
    },
    {
      "auxiliary_loss_clip": 0.01214257,
      "auxiliary_loss_mlp": 0.01003755,
      "balance_loss_clip": 1.01071763,
      "balance_loss_mlp": 1.00283706,
      "epoch": 0.724343173209884,
      "flos": 62311659684480.0,
      "grad_norm": 0.7100737493187583,
      "language_loss": 0.57990015,
      "learning_rate": 7.453790660787815e-07,
      "loss": 0.60208023,
      "num_input_tokens_seen": 129579650,
      "step": 6024,
      "time_per_iteration": 3.3846049308776855
    },
    {
      "auxiliary_loss_clip": 0.01277013,
      "auxiliary_loss_mlp": 0.01029589,
      "balance_loss_clip": 1.04818344,
      "balance_loss_mlp": 1.02229369,
      "epoch": 0.724463416100523,
      "flos": 35006403813120.0,
      "grad_norm": 4.645017133926816,
      "language_loss": 0.63636297,
      "learning_rate": 7.447725220744214e-07,
      "loss": 0.65942895,
      "num_input_tokens_seen": 129601895,
      "step": 6025,
      "time_per_iteration": 2.7954699993133545
    },
    {
      "auxiliary_loss_clip": 0.01176822,
      "auxiliary_loss_mlp": 0.01024098,
      "balance_loss_clip": 1.0497613,
      "balance_loss_mlp": 1.01681995,
      "epoch": 0.7245836589911622,
      "flos": 21871968923520.0,
      "grad_norm": 2.7617900828654802,
      "language_loss": 0.770872,
      "learning_rate": 7.441661684870717e-07,
      "loss": 0.79288119,
      "num_input_tokens_seen": 129622150,
      "step": 6026,
      "time_per_iteration": 2.6312646865844727
    },
    {
      "auxiliary_loss_clip": 0.0117683,
      "auxiliary_loss_mlp": 0.01023243,
      "balance_loss_clip": 1.05002451,
      "balance_loss_mlp": 1.01619172,
      "epoch": 0.7247039018818012,
      "flos": 23006494972800.0,
      "grad_norm": 1.776870145160391,
      "language_loss": 0.81805676,
      "learning_rate": 7.435600054087152e-07,
      "loss": 0.84005749,
      "num_input_tokens_seen": 129644315,
      "step": 6027,
      "time_per_iteration": 2.673257827758789
    },
    {
      "auxiliary_loss_clip": 0.01180004,
      "auxiliary_loss_mlp": 0.01030249,
      "balance_loss_clip": 1.05203664,
      "balance_loss_mlp": 1.02282834,
      "epoch": 0.7248241447724403,
      "flos": 31722588587520.0,
      "grad_norm": 2.2890636628303245,
      "language_loss": 0.74550152,
      "learning_rate": 7.42954032931308e-07,
      "loss": 0.76760399,
      "num_input_tokens_seen": 129665355,
      "step": 6028,
      "time_per_iteration": 2.6606762409210205
    },
    {
      "auxiliary_loss_clip": 0.01274393,
      "auxiliary_loss_mlp": 0.01029746,
      "balance_loss_clip": 1.04751992,
      "balance_loss_mlp": 1.02263784,
      "epoch": 0.7249443876630794,
      "flos": 34896984007680.0,
      "grad_norm": 1.9485468480348205,
      "language_loss": 0.74740916,
      "learning_rate": 7.423482511467733e-07,
      "loss": 0.77045059,
      "num_input_tokens_seen": 129686125,
      "step": 6029,
      "time_per_iteration": 2.7783994674682617
    },
    {
      "auxiliary_loss_clip": 0.01414465,
      "auxiliary_loss_mlp": 0.01028055,
      "balance_loss_clip": 1.0415225,
      "balance_loss_mlp": 1.02125168,
      "epoch": 0.7250646305537185,
      "flos": 26359294268160.0,
      "grad_norm": 2.8948561856401813,
      "language_loss": 0.64608759,
      "learning_rate": 7.417426601470099e-07,
      "loss": 0.6705128,
      "num_input_tokens_seen": 129706485,
      "step": 6030,
      "time_per_iteration": 2.8125898838043213
    },
    {
      "auxiliary_loss_clip": 0.01227646,
      "auxiliary_loss_mlp": 0.01028766,
      "balance_loss_clip": 1.0490936,
      "balance_loss_mlp": 1.02114904,
      "epoch": 0.7251848734443576,
      "flos": 30081614728320.0,
      "grad_norm": 2.367959889933403,
      "language_loss": 0.78555286,
      "learning_rate": 7.411372600238841e-07,
      "loss": 0.80811691,
      "num_input_tokens_seen": 129727100,
      "step": 6031,
      "time_per_iteration": 2.6708152294158936
    },
    {
      "auxiliary_loss_clip": 0.01177006,
      "auxiliary_loss_mlp": 0.01023671,
      "balance_loss_clip": 1.05052638,
      "balance_loss_mlp": 1.0163343,
      "epoch": 0.7253051163349967,
      "flos": 17785262943360.0,
      "grad_norm": 2.3394326850840343,
      "language_loss": 0.74282992,
      "learning_rate": 7.405320508692346e-07,
      "loss": 0.76483673,
      "num_input_tokens_seen": 129745840,
      "step": 6032,
      "time_per_iteration": 2.5553367137908936
    },
    {
      "auxiliary_loss_clip": 0.01172131,
      "auxiliary_loss_mlp": 0.01024861,
      "balance_loss_clip": 1.04982066,
      "balance_loss_mlp": 1.01875174,
      "epoch": 0.7254253592256358,
      "flos": 12641346938880.0,
      "grad_norm": 2.3085958953795482,
      "language_loss": 0.75458527,
      "learning_rate": 7.399270327748727e-07,
      "loss": 0.77655524,
      "num_input_tokens_seen": 129763500,
      "step": 6033,
      "time_per_iteration": 2.5766327381134033
    },
    {
      "auxiliary_loss_clip": 0.01326428,
      "auxiliary_loss_mlp": 0.02561479,
      "balance_loss_clip": 1.04559541,
      "balance_loss_mlp": 0.99990588,
      "epoch": 0.7255456021162748,
      "flos": 27199208966400.0,
      "grad_norm": 1.9469647934556997,
      "language_loss": 0.74356413,
      "learning_rate": 7.39322205832577e-07,
      "loss": 0.78244328,
      "num_input_tokens_seen": 129784390,
      "step": 6034,
      "time_per_iteration": 2.756756544113159
    },
    {
      "auxiliary_loss_clip": 0.01270306,
      "auxiliary_loss_mlp": 0.01025541,
      "balance_loss_clip": 1.04632676,
      "balance_loss_mlp": 1.01801264,
      "epoch": 0.725665845006914,
      "flos": 21288205088640.0,
      "grad_norm": 1.9013777018159936,
      "language_loss": 0.81249762,
      "learning_rate": 7.387175701341009e-07,
      "loss": 0.83545607,
      "num_input_tokens_seen": 129803060,
      "step": 6035,
      "time_per_iteration": 3.646077871322632
    },
    {
      "auxiliary_loss_clip": 0.01222947,
      "auxiliary_loss_mlp": 0.01026751,
      "balance_loss_clip": 1.04731011,
      "balance_loss_mlp": 1.01921153,
      "epoch": 0.7257860878975531,
      "flos": 16033684129920.0,
      "grad_norm": 2.497714497225054,
      "language_loss": 0.72009808,
      "learning_rate": 7.381131257711659e-07,
      "loss": 0.74259508,
      "num_input_tokens_seen": 129820165,
      "step": 6036,
      "time_per_iteration": 3.536708354949951
    },
    {
      "auxiliary_loss_clip": 0.01274358,
      "auxiliary_loss_mlp": 0.01026182,
      "balance_loss_clip": 1.0527879,
      "balance_loss_mlp": 1.01954508,
      "epoch": 0.7259063307881921,
      "flos": 12129943052160.0,
      "grad_norm": 1.9999774317361363,
      "language_loss": 0.83673066,
      "learning_rate": 7.375088728354677e-07,
      "loss": 0.85973603,
      "num_input_tokens_seen": 129835195,
      "step": 6037,
      "time_per_iteration": 2.6322882175445557
    },
    {
      "auxiliary_loss_clip": 0.01325233,
      "auxiliary_loss_mlp": 0.01027665,
      "balance_loss_clip": 1.04525828,
      "balance_loss_mlp": 1.02110517,
      "epoch": 0.7260265736788313,
      "flos": 30443845432320.0,
      "grad_norm": 2.1291199128661042,
      "language_loss": 0.67728299,
      "learning_rate": 7.369048114186691e-07,
      "loss": 0.70081198,
      "num_input_tokens_seen": 129856240,
      "step": 6038,
      "time_per_iteration": 3.8121414184570312
    },
    {
      "auxiliary_loss_clip": 0.01231447,
      "auxiliary_loss_mlp": 0.02562033,
      "balance_loss_clip": 1.04790294,
      "balance_loss_mlp": 0.9998973,
      "epoch": 0.7261468165694703,
      "flos": 21142264129920.0,
      "grad_norm": 1.6935255311534207,
      "language_loss": 0.82821482,
      "learning_rate": 7.363009416124055e-07,
      "loss": 0.8661496,
      "num_input_tokens_seen": 129875565,
      "step": 6039,
      "time_per_iteration": 2.7420573234558105
    },
    {
      "auxiliary_loss_clip": 0.01324836,
      "auxiliary_loss_mlp": 0.0102896,
      "balance_loss_clip": 1.04666328,
      "balance_loss_mlp": 1.02119982,
      "epoch": 0.7262670594601094,
      "flos": 22306308180480.0,
      "grad_norm": 5.160769033336852,
      "language_loss": 0.62841344,
      "learning_rate": 7.356972635082852e-07,
      "loss": 0.65195137,
      "num_input_tokens_seen": 129894420,
      "step": 6040,
      "time_per_iteration": 2.695060968399048
    },
    {
      "auxiliary_loss_clip": 0.01374301,
      "auxiliary_loss_mlp": 0.01027574,
      "balance_loss_clip": 1.04854059,
      "balance_loss_mlp": 1.02003396,
      "epoch": 0.7263873023507486,
      "flos": 25335049950720.0,
      "grad_norm": 2.1063358523534204,
      "language_loss": 0.75660747,
      "learning_rate": 7.35093777197884e-07,
      "loss": 0.78062618,
      "num_input_tokens_seen": 129914490,
      "step": 6041,
      "time_per_iteration": 2.8083248138427734
    },
    {
      "auxiliary_loss_clip": 0.01269719,
      "auxiliary_loss_mlp": 0.01023565,
      "balance_loss_clip": 1.04561424,
      "balance_loss_mlp": 1.01702082,
      "epoch": 0.7265075452413876,
      "flos": 23878621192320.0,
      "grad_norm": 2.374038268639117,
      "language_loss": 0.85221308,
      "learning_rate": 7.344904827727525e-07,
      "loss": 0.87514591,
      "num_input_tokens_seen": 129931670,
      "step": 6042,
      "time_per_iteration": 2.8112728595733643
    },
    {
      "auxiliary_loss_clip": 0.01326096,
      "auxiliary_loss_mlp": 0.01030345,
      "balance_loss_clip": 1.04388356,
      "balance_loss_mlp": 1.02391088,
      "epoch": 0.7266277881320267,
      "flos": 28724549967360.0,
      "grad_norm": 3.285100262430146,
      "language_loss": 0.73433149,
      "learning_rate": 7.338873803244076e-07,
      "loss": 0.75789589,
      "num_input_tokens_seen": 129946905,
      "step": 6043,
      "time_per_iteration": 2.778938055038452
    },
    {
      "auxiliary_loss_clip": 0.01269828,
      "auxiliary_loss_mlp": 0.01023842,
      "balance_loss_clip": 1.04657543,
      "balance_loss_mlp": 1.01750946,
      "epoch": 0.7267480310226658,
      "flos": 24863507182080.0,
      "grad_norm": 1.860551875855276,
      "language_loss": 0.80908394,
      "learning_rate": 7.332844699443401e-07,
      "loss": 0.83202058,
      "num_input_tokens_seen": 129965505,
      "step": 6044,
      "time_per_iteration": 3.6199100017547607
    },
    {
      "auxiliary_loss_clip": 0.01365885,
      "auxiliary_loss_mlp": 0.01028275,
      "balance_loss_clip": 1.0419488,
      "balance_loss_mlp": 1.02186108,
      "epoch": 0.7268682739133049,
      "flos": 27198490694400.0,
      "grad_norm": 1.888853830692175,
      "language_loss": 0.75661415,
      "learning_rate": 7.326817517240121e-07,
      "loss": 0.78055573,
      "num_input_tokens_seen": 129987210,
      "step": 6045,
      "time_per_iteration": 2.7752795219421387
    },
    {
      "auxiliary_loss_clip": 0.01225033,
      "auxiliary_loss_mlp": 0.02560579,
      "balance_loss_clip": 1.04745162,
      "balance_loss_mlp": 0.99993235,
      "epoch": 0.7269885168039439,
      "flos": 33508138688640.0,
      "grad_norm": 2.175521404754996,
      "language_loss": 0.83389729,
      "learning_rate": 7.320792257548545e-07,
      "loss": 0.87175333,
      "num_input_tokens_seen": 130008385,
      "step": 6046,
      "time_per_iteration": 2.761828660964966
    },
    {
      "auxiliary_loss_clip": 0.01280014,
      "auxiliary_loss_mlp": 0.01026806,
      "balance_loss_clip": 1.04905832,
      "balance_loss_mlp": 1.01921844,
      "epoch": 0.7271087596945831,
      "flos": 24313750548480.0,
      "grad_norm": 2.100964466446376,
      "language_loss": 0.76432472,
      "learning_rate": 7.314768921282704e-07,
      "loss": 0.78739297,
      "num_input_tokens_seen": 130029040,
      "step": 6047,
      "time_per_iteration": 2.7057058811187744
    },
    {
      "auxiliary_loss_clip": 0.01227317,
      "auxiliary_loss_mlp": 0.01025928,
      "balance_loss_clip": 1.04792595,
      "balance_loss_mlp": 1.01884103,
      "epoch": 0.7272290025852222,
      "flos": 23805147922560.0,
      "grad_norm": 2.547771716483514,
      "language_loss": 0.71906888,
      "learning_rate": 7.30874750935633e-07,
      "loss": 0.74160135,
      "num_input_tokens_seen": 130048725,
      "step": 6048,
      "time_per_iteration": 2.6897242069244385
    },
    {
      "auxiliary_loss_clip": 0.01319828,
      "auxiliary_loss_mlp": 0.0102482,
      "balance_loss_clip": 1.04632413,
      "balance_loss_mlp": 1.01842105,
      "epoch": 0.7273492454758612,
      "flos": 16720367408640.0,
      "grad_norm": 2.1382923134064886,
      "language_loss": 0.79569548,
      "learning_rate": 7.30272802268286e-07,
      "loss": 0.81914198,
      "num_input_tokens_seen": 130065720,
      "step": 6049,
      "time_per_iteration": 2.6750850677490234
    },
    {
      "auxiliary_loss_clip": 0.01402176,
      "auxiliary_loss_mlp": 0.01022459,
      "balance_loss_clip": 1.03946352,
      "balance_loss_mlp": 1.01582456,
      "epoch": 0.7274694883665004,
      "flos": 28031330413440.0,
      "grad_norm": 2.1637325769644358,
      "language_loss": 0.76336777,
      "learning_rate": 7.29671046217547e-07,
      "loss": 0.78761411,
      "num_input_tokens_seen": 130084830,
      "step": 6050,
      "time_per_iteration": 2.8447535037994385
    },
    {
      "auxiliary_loss_clip": 0.01324602,
      "auxiliary_loss_mlp": 0.01025042,
      "balance_loss_clip": 1.04561853,
      "balance_loss_mlp": 1.01866436,
      "epoch": 0.7275897312571394,
      "flos": 30372706546560.0,
      "grad_norm": 1.8869149947985737,
      "language_loss": 0.82212669,
      "learning_rate": 7.290694828746988e-07,
      "loss": 0.84562314,
      "num_input_tokens_seen": 130104495,
      "step": 6051,
      "time_per_iteration": 2.7554259300231934
    },
    {
      "auxiliary_loss_clip": 0.01329418,
      "auxiliary_loss_mlp": 0.01023508,
      "balance_loss_clip": 1.04505849,
      "balance_loss_mlp": 1.01681781,
      "epoch": 0.7277099741477785,
      "flos": 19204775498880.0,
      "grad_norm": 1.8887329862158206,
      "language_loss": 0.85652339,
      "learning_rate": 7.284681123310004e-07,
      "loss": 0.88005269,
      "num_input_tokens_seen": 130123210,
      "step": 6052,
      "time_per_iteration": 2.72348690032959
    },
    {
      "auxiliary_loss_clip": 0.01223805,
      "auxiliary_loss_mlp": 0.01027125,
      "balance_loss_clip": 1.04851794,
      "balance_loss_mlp": 1.02003825,
      "epoch": 0.7278302170384175,
      "flos": 20667884186880.0,
      "grad_norm": 1.8365552841508,
      "language_loss": 0.7941044,
      "learning_rate": 7.27866934677678e-07,
      "loss": 0.81661367,
      "num_input_tokens_seen": 130142880,
      "step": 6053,
      "time_per_iteration": 2.669830322265625
    },
    {
      "auxiliary_loss_clip": 0.01368854,
      "auxiliary_loss_mlp": 0.0102488,
      "balance_loss_clip": 1.04311633,
      "balance_loss_mlp": 1.01790023,
      "epoch": 0.7279504599290567,
      "flos": 19093200877440.0,
      "grad_norm": 1.6213937548599522,
      "language_loss": 0.78908932,
      "learning_rate": 7.272659500059297e-07,
      "loss": 0.81302667,
      "num_input_tokens_seen": 130160220,
      "step": 6054,
      "time_per_iteration": 2.7290589809417725
    },
    {
      "auxiliary_loss_clip": 0.01223166,
      "auxiliary_loss_mlp": 0.01027572,
      "balance_loss_clip": 1.04865456,
      "balance_loss_mlp": 1.02065206,
      "epoch": 0.7280707028196958,
      "flos": 19062174504960.0,
      "grad_norm": 3.0731355216415124,
      "language_loss": 0.80492854,
      "learning_rate": 7.266651584069264e-07,
      "loss": 0.82743585,
      "num_input_tokens_seen": 130177885,
      "step": 6055,
      "time_per_iteration": 2.6831986904144287
    },
    {
      "auxiliary_loss_clip": 0.01229132,
      "auxiliary_loss_mlp": 0.01028047,
      "balance_loss_clip": 1.05161881,
      "balance_loss_mlp": 1.02157354,
      "epoch": 0.7281909457103348,
      "flos": 37196308293120.0,
      "grad_norm": 1.6289412538524368,
      "language_loss": 0.56646335,
      "learning_rate": 7.260645599718045e-07,
      "loss": 0.58903515,
      "num_input_tokens_seen": 130204240,
      "step": 6056,
      "time_per_iteration": 2.732125997543335
    },
    {
      "auxiliary_loss_clip": 0.01275372,
      "auxiliary_loss_mlp": 0.01027451,
      "balance_loss_clip": 1.04860258,
      "balance_loss_mlp": 1.02026868,
      "epoch": 0.728311188600974,
      "flos": 20667094087680.0,
      "grad_norm": 6.29477714699912,
      "language_loss": 0.67384863,
      "learning_rate": 7.254641547916767e-07,
      "loss": 0.69687688,
      "num_input_tokens_seen": 130221735,
      "step": 6057,
      "time_per_iteration": 2.688161849975586
    },
    {
      "auxiliary_loss_clip": 0.01176766,
      "auxiliary_loss_mlp": 0.01026846,
      "balance_loss_clip": 1.05136514,
      "balance_loss_mlp": 1.01979232,
      "epoch": 0.728431431491613,
      "flos": 28840685616000.0,
      "grad_norm": 1.9558641211295063,
      "language_loss": 0.6934371,
      "learning_rate": 7.248639429576226e-07,
      "loss": 0.71547323,
      "num_input_tokens_seen": 130241190,
      "step": 6058,
      "time_per_iteration": 2.630911350250244
    },
    {
      "auxiliary_loss_clip": 0.0122862,
      "auxiliary_loss_mlp": 0.01026436,
      "balance_loss_clip": 1.05123734,
      "balance_loss_mlp": 1.01940513,
      "epoch": 0.7285516743822521,
      "flos": 25991856092160.0,
      "grad_norm": 1.7234552126907416,
      "language_loss": 0.72158515,
      "learning_rate": 7.242639245606959e-07,
      "loss": 0.74413568,
      "num_input_tokens_seen": 130260980,
      "step": 6059,
      "time_per_iteration": 2.688734531402588
    },
    {
      "auxiliary_loss_clip": 0.01278964,
      "auxiliary_loss_mlp": 0.01028396,
      "balance_loss_clip": 1.04653096,
      "balance_loss_mlp": 1.02140749,
      "epoch": 0.7286719172728913,
      "flos": 16399721675520.0,
      "grad_norm": 2.0887998958915763,
      "language_loss": 0.82269895,
      "learning_rate": 7.236640996919168e-07,
      "loss": 0.84577262,
      "num_input_tokens_seen": 130280025,
      "step": 6060,
      "time_per_iteration": 2.642817497253418
    },
    {
      "auxiliary_loss_clip": 0.01228416,
      "auxiliary_loss_mlp": 0.01026582,
      "balance_loss_clip": 1.05040717,
      "balance_loss_mlp": 1.02017164,
      "epoch": 0.7287921601635303,
      "flos": 22018161277440.0,
      "grad_norm": 1.8166457875773183,
      "language_loss": 0.70732379,
      "learning_rate": 7.230644684422782e-07,
      "loss": 0.72987384,
      "num_input_tokens_seen": 130300255,
      "step": 6061,
      "time_per_iteration": 3.5984373092651367
    },
    {
      "auxiliary_loss_clip": 0.01319558,
      "auxiliary_loss_mlp": 0.01022363,
      "balance_loss_clip": 1.04512572,
      "balance_loss_mlp": 1.01501369,
      "epoch": 0.7289124030541694,
      "flos": 24600927784320.0,
      "grad_norm": 1.9019326046748992,
      "language_loss": 0.81377208,
      "learning_rate": 7.224650309027451e-07,
      "loss": 0.83719122,
      "num_input_tokens_seen": 130320005,
      "step": 6062,
      "time_per_iteration": 3.599729061126709
    },
    {
      "auxiliary_loss_clip": 0.01231772,
      "auxiliary_loss_mlp": 0.01022018,
      "balance_loss_clip": 1.05296063,
      "balance_loss_mlp": 1.01525331,
      "epoch": 0.7290326459448085,
      "flos": 21393638484480.0,
      "grad_norm": 2.3157866561199065,
      "language_loss": 0.68881989,
      "learning_rate": 7.218657871642506e-07,
      "loss": 0.71135777,
      "num_input_tokens_seen": 130338810,
      "step": 6063,
      "time_per_iteration": 3.503795623779297
    },
    {
      "auxiliary_loss_clip": 0.01180925,
      "auxiliary_loss_mlp": 0.01023012,
      "balance_loss_clip": 1.05302691,
      "balance_loss_mlp": 1.01583004,
      "epoch": 0.7291528888354476,
      "flos": 18587686821120.0,
      "grad_norm": 4.389339114217605,
      "language_loss": 0.62300622,
      "learning_rate": 7.212667373177012e-07,
      "loss": 0.64504564,
      "num_input_tokens_seen": 130353805,
      "step": 6064,
      "time_per_iteration": 2.670264959335327
    },
    {
      "auxiliary_loss_clip": 0.01325612,
      "auxiliary_loss_mlp": 0.01027656,
      "balance_loss_clip": 1.0440259,
      "balance_loss_mlp": 1.0207715,
      "epoch": 0.7292731317260867,
      "flos": 18951066760320.0,
      "grad_norm": 1.648301898834153,
      "language_loss": 0.75135446,
      "learning_rate": 7.206678814539704e-07,
      "loss": 0.77488714,
      "num_input_tokens_seen": 130372105,
      "step": 6065,
      "time_per_iteration": 2.693706512451172
    },
    {
      "auxiliary_loss_clip": 0.01382571,
      "auxiliary_loss_mlp": 0.01026315,
      "balance_loss_clip": 1.04518259,
      "balance_loss_mlp": 1.01965117,
      "epoch": 0.7293933746167258,
      "flos": 21067569797760.0,
      "grad_norm": 1.4884004947086507,
      "language_loss": 0.72700655,
      "learning_rate": 7.20069219663904e-07,
      "loss": 0.75109541,
      "num_input_tokens_seen": 130391990,
      "step": 6066,
      "time_per_iteration": 2.7409887313842773
    },
    {
      "auxiliary_loss_clip": 0.01228943,
      "auxiliary_loss_mlp": 0.01028938,
      "balance_loss_clip": 1.04799676,
      "balance_loss_mlp": 1.0217433,
      "epoch": 0.7295136175073649,
      "flos": 22453326547200.0,
      "grad_norm": 2.957682660791064,
      "language_loss": 0.8010956,
      "learning_rate": 7.1947075203832e-07,
      "loss": 0.82367438,
      "num_input_tokens_seen": 130411970,
      "step": 6067,
      "time_per_iteration": 2.650104284286499
    },
    {
      "auxiliary_loss_clip": 0.0106684,
      "auxiliary_loss_mlp": 0.01002938,
      "balance_loss_clip": 1.01102829,
      "balance_loss_mlp": 1.00201452,
      "epoch": 0.7296338603980039,
      "flos": 56125506648960.0,
      "grad_norm": 0.8565103802203478,
      "language_loss": 0.60123616,
      "learning_rate": 7.188724786680049e-07,
      "loss": 0.62193388,
      "num_input_tokens_seen": 130472440,
      "step": 6068,
      "time_per_iteration": 3.177157402038574
    },
    {
      "auxiliary_loss_clip": 0.01273964,
      "auxiliary_loss_mlp": 0.01020879,
      "balance_loss_clip": 1.04631984,
      "balance_loss_mlp": 1.01395547,
      "epoch": 0.7297541032886431,
      "flos": 25228287751680.0,
      "grad_norm": 1.6676783057016382,
      "language_loss": 0.75658214,
      "learning_rate": 7.182743996437162e-07,
      "loss": 0.77953064,
      "num_input_tokens_seen": 130491975,
      "step": 6069,
      "time_per_iteration": 2.733860969543457
    },
    {
      "auxiliary_loss_clip": 0.01330014,
      "auxiliary_loss_mlp": 0.01027853,
      "balance_loss_clip": 1.04605508,
      "balance_loss_mlp": 1.01968694,
      "epoch": 0.7298743461792822,
      "flos": 26467600752000.0,
      "grad_norm": 32.662405792892315,
      "language_loss": 0.68684548,
      "learning_rate": 7.176765150561819e-07,
      "loss": 0.71042418,
      "num_input_tokens_seen": 130510580,
      "step": 6070,
      "time_per_iteration": 3.6851398944854736
    },
    {
      "auxiliary_loss_clip": 0.01178449,
      "auxiliary_loss_mlp": 0.01026431,
      "balance_loss_clip": 1.05037045,
      "balance_loss_mlp": 1.01953804,
      "epoch": 0.7299945890699212,
      "flos": 19569053278080.0,
      "grad_norm": 2.308865361980118,
      "language_loss": 0.80080682,
      "learning_rate": 7.170788249961002e-07,
      "loss": 0.82285565,
      "num_input_tokens_seen": 130529090,
      "step": 6071,
      "time_per_iteration": 2.596703052520752
    },
    {
      "auxiliary_loss_clip": 0.01172809,
      "auxiliary_loss_mlp": 0.0102317,
      "balance_loss_clip": 1.04885352,
      "balance_loss_mlp": 1.01605916,
      "epoch": 0.7301148319605604,
      "flos": 22928963466240.0,
      "grad_norm": 2.0132957587283777,
      "language_loss": 0.88146675,
      "learning_rate": 7.164813295541418e-07,
      "loss": 0.90342659,
      "num_input_tokens_seen": 130548655,
      "step": 6072,
      "time_per_iteration": 2.5773067474365234
    },
    {
      "auxiliary_loss_clip": 0.01275283,
      "auxiliary_loss_mlp": 0.01029119,
      "balance_loss_clip": 1.04653287,
      "balance_loss_mlp": 1.02178717,
      "epoch": 0.7302350748511994,
      "flos": 25369703596800.0,
      "grad_norm": 1.9642663707625094,
      "language_loss": 0.70314252,
      "learning_rate": 7.15884028820944e-07,
      "loss": 0.72618651,
      "num_input_tokens_seen": 130567710,
      "step": 6073,
      "time_per_iteration": 2.7155799865722656
    },
    {
      "auxiliary_loss_clip": 0.01322762,
      "auxiliary_loss_mlp": 0.01023001,
      "balance_loss_clip": 1.04403639,
      "balance_loss_mlp": 1.0156126,
      "epoch": 0.7303553177418385,
      "flos": 27819170732160.0,
      "grad_norm": 3.1884172961305746,
      "language_loss": 0.60488945,
      "learning_rate": 7.152869228871185e-07,
      "loss": 0.6283471,
      "num_input_tokens_seen": 130590195,
      "step": 6074,
      "time_per_iteration": 2.7576606273651123
    },
    {
      "auxiliary_loss_clip": 0.01275984,
      "auxiliary_loss_mlp": 0.01028902,
      "balance_loss_clip": 1.04777622,
      "balance_loss_mlp": 1.02135026,
      "epoch": 0.7304755606324776,
      "flos": 24426510318720.0,
      "grad_norm": 1.8341588104007065,
      "language_loss": 0.72652757,
      "learning_rate": 7.146900118432457e-07,
      "loss": 0.74957639,
      "num_input_tokens_seen": 130609940,
      "step": 6075,
      "time_per_iteration": 2.7106711864471436
    },
    {
      "auxiliary_loss_clip": 0.01470868,
      "auxiliary_loss_mlp": 0.01023792,
      "balance_loss_clip": 1.03616476,
      "balance_loss_mlp": 1.01692581,
      "epoch": 0.7305958035231167,
      "flos": 23840483927040.0,
      "grad_norm": 2.0556127086186464,
      "language_loss": 0.86049652,
      "learning_rate": 7.140932957798753e-07,
      "loss": 0.88544309,
      "num_input_tokens_seen": 130628380,
      "step": 6076,
      "time_per_iteration": 2.918816089630127
    },
    {
      "auxiliary_loss_clip": 0.01278701,
      "auxiliary_loss_mlp": 0.01021454,
      "balance_loss_clip": 1.04515505,
      "balance_loss_mlp": 1.01476312,
      "epoch": 0.7307160464137558,
      "flos": 16726939597440.0,
      "grad_norm": 7.246585463338025,
      "language_loss": 0.71078533,
      "learning_rate": 7.134967747875309e-07,
      "loss": 0.73378688,
      "num_input_tokens_seen": 130646590,
      "step": 6077,
      "time_per_iteration": 2.7619755268096924
    },
    {
      "auxiliary_loss_clip": 0.01222633,
      "auxiliary_loss_mlp": 0.01024435,
      "balance_loss_clip": 1.04733551,
      "balance_loss_mlp": 1.01702642,
      "epoch": 0.7308362893043949,
      "flos": 21798280172160.0,
      "grad_norm": 2.4698891656860185,
      "language_loss": 0.81987423,
      "learning_rate": 7.129004489567014e-07,
      "loss": 0.84234494,
      "num_input_tokens_seen": 130664070,
      "step": 6078,
      "time_per_iteration": 2.6458818912506104
    },
    {
      "auxiliary_loss_clip": 0.01323125,
      "auxiliary_loss_mlp": 0.01022317,
      "balance_loss_clip": 1.04402924,
      "balance_loss_mlp": 1.01572812,
      "epoch": 0.730956532195034,
      "flos": 10707377840640.0,
      "grad_norm": 2.722893409409163,
      "language_loss": 0.78036922,
      "learning_rate": 7.123043183778512e-07,
      "loss": 0.80382359,
      "num_input_tokens_seen": 130681400,
      "step": 6079,
      "time_per_iteration": 2.669877052307129
    },
    {
      "auxiliary_loss_clip": 0.01327269,
      "auxiliary_loss_mlp": 0.01027377,
      "balance_loss_clip": 1.04793644,
      "balance_loss_mlp": 1.02027774,
      "epoch": 0.731076775085673,
      "flos": 19791987039360.0,
      "grad_norm": 1.5520759635931385,
      "language_loss": 0.65096492,
      "learning_rate": 7.117083831414114e-07,
      "loss": 0.67451143,
      "num_input_tokens_seen": 130700675,
      "step": 6080,
      "time_per_iteration": 2.673048973083496
    },
    {
      "auxiliary_loss_clip": 0.0117525,
      "auxiliary_loss_mlp": 0.01027401,
      "balance_loss_clip": 1.05097771,
      "balance_loss_mlp": 1.02068996,
      "epoch": 0.7311970179763122,
      "flos": 20447033414400.0,
      "grad_norm": 1.8958256068990107,
      "language_loss": 0.69969064,
      "learning_rate": 7.11112643337787e-07,
      "loss": 0.72171712,
      "num_input_tokens_seen": 130719720,
      "step": 6081,
      "time_per_iteration": 2.58666729927063
    },
    {
      "auxiliary_loss_clip": 0.01282313,
      "auxiliary_loss_mlp": 0.01029219,
      "balance_loss_clip": 1.05257154,
      "balance_loss_mlp": 1.02167308,
      "epoch": 0.7313172608669513,
      "flos": 18513818501760.0,
      "grad_norm": 2.516180017852946,
      "language_loss": 0.77061242,
      "learning_rate": 7.10517099057349e-07,
      "loss": 0.79372776,
      "num_input_tokens_seen": 130736670,
      "step": 6082,
      "time_per_iteration": 2.6469478607177734
    },
    {
      "auxiliary_loss_clip": 0.01278278,
      "auxiliary_loss_mlp": 0.01023195,
      "balance_loss_clip": 1.04827762,
      "balance_loss_mlp": 1.01568449,
      "epoch": 0.7314375037575903,
      "flos": 16180738410240.0,
      "grad_norm": 2.327576560898737,
      "language_loss": 0.61489546,
      "learning_rate": 7.099217503904411e-07,
      "loss": 0.63791013,
      "num_input_tokens_seen": 130754525,
      "step": 6083,
      "time_per_iteration": 2.720355987548828
    },
    {
      "auxiliary_loss_clip": 0.0127907,
      "auxiliary_loss_mlp": 0.01028652,
      "balance_loss_clip": 1.0488534,
      "balance_loss_mlp": 1.02208018,
      "epoch": 0.7315577466482295,
      "flos": 17967940536960.0,
      "grad_norm": 1.8149189464498579,
      "language_loss": 0.90384626,
      "learning_rate": 7.093265974273788e-07,
      "loss": 0.92692345,
      "num_input_tokens_seen": 130772420,
      "step": 6084,
      "time_per_iteration": 2.738978624343872
    },
    {
      "auxiliary_loss_clip": 0.01227413,
      "auxiliary_loss_mlp": 0.01025311,
      "balance_loss_clip": 1.04810977,
      "balance_loss_mlp": 1.01898146,
      "epoch": 0.7316779895388685,
      "flos": 18405440190720.0,
      "grad_norm": 2.1762594004810882,
      "language_loss": 0.71535742,
      "learning_rate": 7.087316402584447e-07,
      "loss": 0.7378847,
      "num_input_tokens_seen": 130791245,
      "step": 6085,
      "time_per_iteration": 2.631948471069336
    },
    {
      "auxiliary_loss_clip": 0.01174775,
      "auxiliary_loss_mlp": 0.01022724,
      "balance_loss_clip": 1.04970574,
      "balance_loss_mlp": 1.01567912,
      "epoch": 0.7317982324295076,
      "flos": 17928294900480.0,
      "grad_norm": 2.1121681794781404,
      "language_loss": 0.86156547,
      "learning_rate": 7.081368789738953e-07,
      "loss": 0.88354051,
      "num_input_tokens_seen": 130808445,
      "step": 6086,
      "time_per_iteration": 2.5991873741149902
    },
    {
      "auxiliary_loss_clip": 0.01271706,
      "auxiliary_loss_mlp": 0.01029749,
      "balance_loss_clip": 1.04373991,
      "balance_loss_mlp": 1.02268589,
      "epoch": 0.7319184753201466,
      "flos": 27229840289280.0,
      "grad_norm": 2.2037119070427376,
      "language_loss": 0.77738649,
      "learning_rate": 7.075423136639537e-07,
      "loss": 0.80040103,
      "num_input_tokens_seen": 130827700,
      "step": 6087,
      "time_per_iteration": 3.6870412826538086
    },
    {
      "auxiliary_loss_clip": 0.01320229,
      "auxiliary_loss_mlp": 0.01024685,
      "balance_loss_clip": 1.04422855,
      "balance_loss_mlp": 1.01733625,
      "epoch": 0.7320387182107858,
      "flos": 37448544574080.0,
      "grad_norm": 1.8700745029310044,
      "language_loss": 0.74674839,
      "learning_rate": 7.069479444188149e-07,
      "loss": 0.77019751,
      "num_input_tokens_seen": 130848290,
      "step": 6088,
      "time_per_iteration": 2.822258472442627
    },
    {
      "auxiliary_loss_clip": 0.01270936,
      "auxiliary_loss_mlp": 0.01026035,
      "balance_loss_clip": 1.04613662,
      "balance_loss_mlp": 1.01809537,
      "epoch": 0.7321589611014249,
      "flos": 17859023521920.0,
      "grad_norm": 2.1323902100371854,
      "language_loss": 0.82255948,
      "learning_rate": 7.063537713286453e-07,
      "loss": 0.8455292,
      "num_input_tokens_seen": 130865970,
      "step": 6089,
      "time_per_iteration": 3.5812249183654785
    },
    {
      "auxiliary_loss_clip": 0.01278367,
      "auxiliary_loss_mlp": 0.0102397,
      "balance_loss_clip": 1.04631495,
      "balance_loss_mlp": 1.01673126,
      "epoch": 0.7322792039920639,
      "flos": 26100593539200.0,
      "grad_norm": 2.027506922174194,
      "language_loss": 0.8074525,
      "learning_rate": 7.057597944835803e-07,
      "loss": 0.83047587,
      "num_input_tokens_seen": 130885245,
      "step": 6090,
      "time_per_iteration": 2.7118425369262695
    },
    {
      "auxiliary_loss_clip": 0.01329436,
      "auxiliary_loss_mlp": 0.01025787,
      "balance_loss_clip": 1.04522896,
      "balance_loss_mlp": 1.01882815,
      "epoch": 0.7323994468827031,
      "flos": 25369093065600.0,
      "grad_norm": 2.3347493402586617,
      "language_loss": 0.74779356,
      "learning_rate": 7.051660139737253e-07,
      "loss": 0.77134579,
      "num_input_tokens_seen": 130903465,
      "step": 6091,
      "time_per_iteration": 2.711240291595459
    },
    {
      "auxiliary_loss_clip": 0.01221996,
      "auxiliary_loss_mlp": 0.02565838,
      "balance_loss_clip": 1.04886866,
      "balance_loss_mlp": 0.99990833,
      "epoch": 0.7325196897733421,
      "flos": 26907075653760.0,
      "grad_norm": 2.8628177884795565,
      "language_loss": 0.76765448,
      "learning_rate": 7.045724298891565e-07,
      "loss": 0.80553281,
      "num_input_tokens_seen": 130922935,
      "step": 6092,
      "time_per_iteration": 2.6162407398223877
    },
    {
      "auxiliary_loss_clip": 0.01223872,
      "auxiliary_loss_mlp": 0.01027557,
      "balance_loss_clip": 1.05012047,
      "balance_loss_mlp": 1.02001715,
      "epoch": 0.7326399326639812,
      "flos": 25775781828480.0,
      "grad_norm": 2.019446308261088,
      "language_loss": 0.69060481,
      "learning_rate": 7.039790423199192e-07,
      "loss": 0.71311903,
      "num_input_tokens_seen": 130942575,
      "step": 6093,
      "time_per_iteration": 2.6447596549987793
    },
    {
      "auxiliary_loss_clip": 0.01278059,
      "auxiliary_loss_mlp": 0.01027068,
      "balance_loss_clip": 1.04885745,
      "balance_loss_mlp": 1.01990104,
      "epoch": 0.7327601755546204,
      "flos": 21032269706880.0,
      "grad_norm": 2.174853608286618,
      "language_loss": 0.78094131,
      "learning_rate": 7.033858513560322e-07,
      "loss": 0.80399263,
      "num_input_tokens_seen": 130958870,
      "step": 6094,
      "time_per_iteration": 2.642805337905884
    },
    {
      "auxiliary_loss_clip": 0.0122768,
      "auxiliary_loss_mlp": 0.01025413,
      "balance_loss_clip": 1.05220413,
      "balance_loss_mlp": 1.01860952,
      "epoch": 0.7328804184452594,
      "flos": 16289224462080.0,
      "grad_norm": 2.787957809094547,
      "language_loss": 0.76285326,
      "learning_rate": 7.027928570874794e-07,
      "loss": 0.78538418,
      "num_input_tokens_seen": 130977060,
      "step": 6095,
      "time_per_iteration": 2.593320608139038
    },
    {
      "auxiliary_loss_clip": 0.01172889,
      "auxiliary_loss_mlp": 0.01026848,
      "balance_loss_clip": 1.04906642,
      "balance_loss_mlp": 1.0197041,
      "epoch": 0.7330006613358985,
      "flos": 17858233422720.0,
      "grad_norm": 1.811052333871359,
      "language_loss": 0.85599881,
      "learning_rate": 7.022000596042194e-07,
      "loss": 0.87799621,
      "num_input_tokens_seen": 130994160,
      "step": 6096,
      "time_per_iteration": 3.5031301975250244
    },
    {
      "auxiliary_loss_clip": 0.01322148,
      "auxiliary_loss_mlp": 0.0102571,
      "balance_loss_clip": 1.04207587,
      "balance_loss_mlp": 1.01894152,
      "epoch": 0.7331209042265376,
      "flos": 22492074343680.0,
      "grad_norm": 2.4535247860022062,
      "language_loss": 0.81815481,
      "learning_rate": 7.016074589961784e-07,
      "loss": 0.84163332,
      "num_input_tokens_seen": 131012725,
      "step": 6097,
      "time_per_iteration": 2.785069465637207
    },
    {
      "auxiliary_loss_clip": 0.01277051,
      "auxiliary_loss_mlp": 0.01027493,
      "balance_loss_clip": 1.05009389,
      "balance_loss_mlp": 1.02079356,
      "epoch": 0.7332411471171767,
      "flos": 33072757937280.0,
      "grad_norm": 2.116362364164424,
      "language_loss": 0.66988432,
      "learning_rate": 7.01015055353253e-07,
      "loss": 0.6929298,
      "num_input_tokens_seen": 131035150,
      "step": 6098,
      "time_per_iteration": 2.7459042072296143
    },
    {
      "auxiliary_loss_clip": 0.01372963,
      "auxiliary_loss_mlp": 0.01024253,
      "balance_loss_clip": 1.04598081,
      "balance_loss_mlp": 1.01686513,
      "epoch": 0.7333613900078157,
      "flos": 22743017735040.0,
      "grad_norm": 1.9279618784701693,
      "language_loss": 0.78223872,
      "learning_rate": 7.004228487653123e-07,
      "loss": 0.80621088,
      "num_input_tokens_seen": 131055955,
      "step": 6099,
      "time_per_iteration": 2.740433931350708
    },
    {
      "auxiliary_loss_clip": 0.01324243,
      "auxiliary_loss_mlp": 0.0102509,
      "balance_loss_clip": 1.04072952,
      "balance_loss_mlp": 1.01789904,
      "epoch": 0.7334816328984549,
      "flos": 22346133384960.0,
      "grad_norm": 2.0401832219985208,
      "language_loss": 0.78049815,
      "learning_rate": 6.998308393221906e-07,
      "loss": 0.80399156,
      "num_input_tokens_seen": 131074360,
      "step": 6100,
      "time_per_iteration": 2.6887688636779785
    },
    {
      "auxiliary_loss_clip": 0.01327203,
      "auxiliary_loss_mlp": 0.01024613,
      "balance_loss_clip": 1.04663336,
      "balance_loss_mlp": 1.01839066,
      "epoch": 0.733601875789094,
      "flos": 20736149984640.0,
      "grad_norm": 2.9770352403663543,
      "language_loss": 0.71003383,
      "learning_rate": 6.992390271136977e-07,
      "loss": 0.73355204,
      "num_input_tokens_seen": 131090070,
      "step": 6101,
      "time_per_iteration": 2.72068190574646
    },
    {
      "auxiliary_loss_clip": 0.01220094,
      "auxiliary_loss_mlp": 0.01023536,
      "balance_loss_clip": 1.04524732,
      "balance_loss_mlp": 1.0166043,
      "epoch": 0.733722118679733,
      "flos": 22564362464640.0,
      "grad_norm": 3.667229653719235,
      "language_loss": 0.85862792,
      "learning_rate": 6.986474122296094e-07,
      "loss": 0.88106424,
      "num_input_tokens_seen": 131109185,
      "step": 6102,
      "time_per_iteration": 2.629157543182373
    },
    {
      "auxiliary_loss_clip": 0.01181607,
      "auxiliary_loss_mlp": 0.01024184,
      "balance_loss_clip": 1.05328441,
      "balance_loss_mlp": 1.01678681,
      "epoch": 0.7338423615703722,
      "flos": 20084192179200.0,
      "grad_norm": 1.8044303836480957,
      "language_loss": 0.72291887,
      "learning_rate": 6.980559947596751e-07,
      "loss": 0.7449767,
      "num_input_tokens_seen": 131127725,
      "step": 6103,
      "time_per_iteration": 2.5883724689483643
    },
    {
      "auxiliary_loss_clip": 0.01372175,
      "auxiliary_loss_mlp": 0.01026448,
      "balance_loss_clip": 1.04364038,
      "balance_loss_mlp": 1.01927781,
      "epoch": 0.7339626044610112,
      "flos": 21687675217920.0,
      "grad_norm": 1.9667315582819909,
      "language_loss": 0.75952697,
      "learning_rate": 6.974647747936109e-07,
      "loss": 0.78351319,
      "num_input_tokens_seen": 131146110,
      "step": 6104,
      "time_per_iteration": 2.695183515548706
    },
    {
      "auxiliary_loss_clip": 0.01177517,
      "auxiliary_loss_mlp": 0.0256486,
      "balance_loss_clip": 1.05141926,
      "balance_loss_mlp": 0.99995828,
      "epoch": 0.7340828473516503,
      "flos": 15268248282240.0,
      "grad_norm": 1.9252103848600302,
      "language_loss": 0.82430947,
      "learning_rate": 6.968737524211039e-07,
      "loss": 0.86173326,
      "num_input_tokens_seen": 131162920,
      "step": 6105,
      "time_per_iteration": 2.6466257572174072
    },
    {
      "auxiliary_loss_clip": 0.01224073,
      "auxiliary_loss_mlp": 0.01022828,
      "balance_loss_clip": 1.04873538,
      "balance_loss_mlp": 1.01584864,
      "epoch": 0.7342030902422895,
      "flos": 22930112701440.0,
      "grad_norm": 2.4481391884213464,
      "language_loss": 0.80456436,
      "learning_rate": 6.962829277318132e-07,
      "loss": 0.8270334,
      "num_input_tokens_seen": 131182515,
      "step": 6106,
      "time_per_iteration": 2.6095941066741943
    },
    {
      "auxiliary_loss_clip": 0.01228356,
      "auxiliary_loss_mlp": 0.01024278,
      "balance_loss_clip": 1.05202174,
      "balance_loss_mlp": 1.01760256,
      "epoch": 0.7343233331329285,
      "flos": 25847890381440.0,
      "grad_norm": 1.8470694917955117,
      "language_loss": 0.83731669,
      "learning_rate": 6.956923008153652e-07,
      "loss": 0.85984302,
      "num_input_tokens_seen": 131202280,
      "step": 6107,
      "time_per_iteration": 2.654435873031616
    },
    {
      "auxiliary_loss_clip": 0.01226942,
      "auxiliary_loss_mlp": 0.0102823,
      "balance_loss_clip": 1.04845238,
      "balance_loss_mlp": 1.02201343,
      "epoch": 0.7344435760235676,
      "flos": 18478985287680.0,
      "grad_norm": 2.3335057693210257,
      "language_loss": 0.84187281,
      "learning_rate": 6.951018717613593e-07,
      "loss": 0.86442459,
      "num_input_tokens_seen": 131221295,
      "step": 6108,
      "time_per_iteration": 2.6262354850769043
    },
    {
      "auxiliary_loss_clip": 0.01223613,
      "auxiliary_loss_mlp": 0.01024023,
      "balance_loss_clip": 1.04918277,
      "balance_loss_mlp": 1.0169481,
      "epoch": 0.7345638189142067,
      "flos": 17640040256640.0,
      "grad_norm": 2.5129570874396028,
      "language_loss": 0.78405905,
      "learning_rate": 6.945116406593614e-07,
      "loss": 0.80653536,
      "num_input_tokens_seen": 131240150,
      "step": 6109,
      "time_per_iteration": 2.615175724029541
    },
    {
      "auxiliary_loss_clip": 0.01376926,
      "auxiliary_loss_mlp": 0.01028621,
      "balance_loss_clip": 1.04672718,
      "balance_loss_mlp": 1.02201653,
      "epoch": 0.7346840618048458,
      "flos": 20260225756800.0,
      "grad_norm": 2.239542761454538,
      "language_loss": 0.74537098,
      "learning_rate": 6.939216075989089e-07,
      "loss": 0.76942647,
      "num_input_tokens_seen": 131258080,
      "step": 6110,
      "time_per_iteration": 2.7062175273895264
    },
    {
      "auxiliary_loss_clip": 0.01271007,
      "auxiliary_loss_mlp": 0.01025705,
      "balance_loss_clip": 1.04669464,
      "balance_loss_mlp": 1.01806927,
      "epoch": 0.7348043046954849,
      "flos": 29023183641600.0,
      "grad_norm": 2.2948383265944914,
      "language_loss": 0.65881139,
      "learning_rate": 6.933317726695109e-07,
      "loss": 0.68177849,
      "num_input_tokens_seen": 131279310,
      "step": 6111,
      "time_per_iteration": 2.7970571517944336
    },
    {
      "auxiliary_loss_clip": 0.01327712,
      "auxiliary_loss_mlp": 0.01027204,
      "balance_loss_clip": 1.05179,
      "balance_loss_mlp": 1.02050722,
      "epoch": 0.734924547586124,
      "flos": 17931203902080.0,
      "grad_norm": 2.5268833536747697,
      "language_loss": 0.80018288,
      "learning_rate": 6.92742135960644e-07,
      "loss": 0.82373202,
      "num_input_tokens_seen": 131297010,
      "step": 6112,
      "time_per_iteration": 2.6629130840301514
    },
    {
      "auxiliary_loss_clip": 0.01119622,
      "auxiliary_loss_mlp": 0.0099942,
      "balance_loss_clip": 1.01078415,
      "balance_loss_mlp": 0.99845451,
      "epoch": 0.7350447904767631,
      "flos": 63588319850880.0,
      "grad_norm": 0.8256064458886008,
      "language_loss": 0.55619931,
      "learning_rate": 6.921526975617556e-07,
      "loss": 0.57738972,
      "num_input_tokens_seen": 131356470,
      "step": 6113,
      "time_per_iteration": 4.128159761428833
    },
    {
      "auxiliary_loss_clip": 0.01178755,
      "auxiliary_loss_mlp": 0.01028502,
      "balance_loss_clip": 1.04647827,
      "balance_loss_mlp": 1.02100945,
      "epoch": 0.7351650333674021,
      "flos": 21580015178880.0,
      "grad_norm": 1.985047746779728,
      "language_loss": 0.75243199,
      "learning_rate": 6.915634575622631e-07,
      "loss": 0.77450454,
      "num_input_tokens_seen": 131374985,
      "step": 6114,
      "time_per_iteration": 3.7653064727783203
    },
    {
      "auxiliary_loss_clip": 0.01174468,
      "auxiliary_loss_mlp": 0.01023339,
      "balance_loss_clip": 1.04912758,
      "balance_loss_mlp": 1.01662791,
      "epoch": 0.7352852762580413,
      "flos": 18186349184640.0,
      "grad_norm": 2.1457197717855427,
      "language_loss": 0.71091366,
      "learning_rate": 6.909744160515532e-07,
      "loss": 0.73289168,
      "num_input_tokens_seen": 131393125,
      "step": 6115,
      "time_per_iteration": 3.530716896057129
    },
    {
      "auxiliary_loss_clip": 0.01269581,
      "auxiliary_loss_mlp": 0.01020017,
      "balance_loss_clip": 1.04787481,
      "balance_loss_mlp": 1.01295733,
      "epoch": 0.7354055191486804,
      "flos": 38910073063680.0,
      "grad_norm": 3.968505615926375,
      "language_loss": 0.69527054,
      "learning_rate": 6.903855731189849e-07,
      "loss": 0.71816653,
      "num_input_tokens_seen": 131415760,
      "step": 6116,
      "time_per_iteration": 2.8107709884643555
    },
    {
      "auxiliary_loss_clip": 0.01283337,
      "auxiliary_loss_mlp": 0.01023916,
      "balance_loss_clip": 1.0494802,
      "balance_loss_mlp": 1.01621521,
      "epoch": 0.7355257620393194,
      "flos": 16289978647680.0,
      "grad_norm": 3.879971792912543,
      "language_loss": 0.81662273,
      "learning_rate": 6.897969288538825e-07,
      "loss": 0.83969527,
      "num_input_tokens_seen": 131433705,
      "step": 6117,
      "time_per_iteration": 2.6271846294403076
    },
    {
      "auxiliary_loss_clip": 0.01265299,
      "auxiliary_loss_mlp": 0.01029147,
      "balance_loss_clip": 1.04555368,
      "balance_loss_mlp": 1.02176213,
      "epoch": 0.7356460049299585,
      "flos": 18114240631680.0,
      "grad_norm": 1.866329697517354,
      "language_loss": 0.81485051,
      "learning_rate": 6.892084833455452e-07,
      "loss": 0.8377949,
      "num_input_tokens_seen": 131453275,
      "step": 6118,
      "time_per_iteration": 2.6525349617004395
    },
    {
      "auxiliary_loss_clip": 0.01221977,
      "auxiliary_loss_mlp": 0.01024791,
      "balance_loss_clip": 1.04922473,
      "balance_loss_mlp": 1.01863956,
      "epoch": 0.7357662478205976,
      "flos": 21325193118720.0,
      "grad_norm": 1.7284763893768438,
      "language_loss": 0.84059083,
      "learning_rate": 6.886202366832384e-07,
      "loss": 0.86305851,
      "num_input_tokens_seen": 131474960,
      "step": 6119,
      "time_per_iteration": 2.6830334663391113
    },
    {
      "auxiliary_loss_clip": 0.01378105,
      "auxiliary_loss_mlp": 0.01026429,
      "balance_loss_clip": 1.04927826,
      "balance_loss_mlp": 1.01900506,
      "epoch": 0.7358864907112367,
      "flos": 14246841139200.0,
      "grad_norm": 1.9383319453275663,
      "language_loss": 0.73681742,
      "learning_rate": 6.880321889561987e-07,
      "loss": 0.76086277,
      "num_input_tokens_seen": 131492935,
      "step": 6120,
      "time_per_iteration": 2.730783700942993
    },
    {
      "auxiliary_loss_clip": 0.01319631,
      "auxiliary_loss_mlp": 0.01029257,
      "balance_loss_clip": 1.04500675,
      "balance_loss_mlp": 1.02147913,
      "epoch": 0.7360067336018757,
      "flos": 22309684058880.0,
      "grad_norm": 2.358475084418294,
      "language_loss": 0.65585488,
      "learning_rate": 6.874443402536338e-07,
      "loss": 0.67934382,
      "num_input_tokens_seen": 131512025,
      "step": 6121,
      "time_per_iteration": 2.7054529190063477
    },
    {
      "auxiliary_loss_clip": 0.01279056,
      "auxiliary_loss_mlp": 0.01026559,
      "balance_loss_clip": 1.05000377,
      "balance_loss_mlp": 1.01945138,
      "epoch": 0.7361269764925149,
      "flos": 25554607833600.0,
      "grad_norm": 25.312146166150775,
      "language_loss": 0.80305982,
      "learning_rate": 6.868566906647177e-07,
      "loss": 0.82611597,
      "num_input_tokens_seen": 131532975,
      "step": 6122,
      "time_per_iteration": 3.609323501586914
    },
    {
      "auxiliary_loss_clip": 0.01226144,
      "auxiliary_loss_mlp": 0.0102318,
      "balance_loss_clip": 1.04821718,
      "balance_loss_mlp": 1.01566935,
      "epoch": 0.736247219383154,
      "flos": 20376505059840.0,
      "grad_norm": 2.7021712317982556,
      "language_loss": 0.83626795,
      "learning_rate": 6.862692402785984e-07,
      "loss": 0.85876119,
      "num_input_tokens_seen": 131553225,
      "step": 6123,
      "time_per_iteration": 2.6267597675323486
    },
    {
      "auxiliary_loss_clip": 0.0123308,
      "auxiliary_loss_mlp": 0.01002204,
      "balance_loss_clip": 1.02036858,
      "balance_loss_mlp": 1.0011847,
      "epoch": 0.736367462273793,
      "flos": 70339525735680.0,
      "grad_norm": 0.6848392950552733,
      "language_loss": 0.49549723,
      "learning_rate": 6.856819891843899e-07,
      "loss": 0.5178501,
      "num_input_tokens_seen": 131617930,
      "step": 6124,
      "time_per_iteration": 3.350579023361206
    },
    {
      "auxiliary_loss_clip": 0.0142323,
      "auxiliary_loss_mlp": 0.01024805,
      "balance_loss_clip": 1.04595232,
      "balance_loss_mlp": 1.0174377,
      "epoch": 0.7364877051644322,
      "flos": 22412711243520.0,
      "grad_norm": 2.095752608200626,
      "language_loss": 0.72533894,
      "learning_rate": 6.8509493747118e-07,
      "loss": 0.74981928,
      "num_input_tokens_seen": 131636740,
      "step": 6125,
      "time_per_iteration": 2.7668466567993164
    },
    {
      "auxiliary_loss_clip": 0.01177003,
      "auxiliary_loss_mlp": 0.0102517,
      "balance_loss_clip": 1.05254745,
      "balance_loss_mlp": 1.01802933,
      "epoch": 0.7366079480550712,
      "flos": 12130266274560.0,
      "grad_norm": 2.954277180450541,
      "language_loss": 0.87989116,
      "learning_rate": 6.845080852280221e-07,
      "loss": 0.90191287,
      "num_input_tokens_seen": 131653810,
      "step": 6126,
      "time_per_iteration": 2.5901620388031006
    },
    {
      "auxiliary_loss_clip": 0.01324265,
      "auxiliary_loss_mlp": 0.01028125,
      "balance_loss_clip": 1.0460248,
      "balance_loss_mlp": 1.02151775,
      "epoch": 0.7367281909457103,
      "flos": 15049336844160.0,
      "grad_norm": 1.5991712429626839,
      "language_loss": 0.74560392,
      "learning_rate": 6.839214325439409e-07,
      "loss": 0.76912785,
      "num_input_tokens_seen": 131671505,
      "step": 6127,
      "time_per_iteration": 2.752765417098999
    },
    {
      "auxiliary_loss_clip": 0.01267799,
      "auxiliary_loss_mlp": 0.01027209,
      "balance_loss_clip": 1.04946923,
      "balance_loss_mlp": 1.02024078,
      "epoch": 0.7368484338363495,
      "flos": 23510752053120.0,
      "grad_norm": 1.632968537987518,
      "language_loss": 0.72088242,
      "learning_rate": 6.833349795079327e-07,
      "loss": 0.74383247,
      "num_input_tokens_seen": 131690615,
      "step": 6128,
      "time_per_iteration": 2.6660990715026855
    },
    {
      "auxiliary_loss_clip": 0.01321052,
      "auxiliary_loss_mlp": 0.01023147,
      "balance_loss_clip": 1.04718876,
      "balance_loss_mlp": 1.01648927,
      "epoch": 0.7369686767269885,
      "flos": 27417833095680.0,
      "grad_norm": 2.20454923405599,
      "language_loss": 0.68388879,
      "learning_rate": 6.827487262089613e-07,
      "loss": 0.70733082,
      "num_input_tokens_seen": 131711120,
      "step": 6129,
      "time_per_iteration": 2.759269952774048
    },
    {
      "auxiliary_loss_clip": 0.01165985,
      "auxiliary_loss_mlp": 0.01003654,
      "balance_loss_clip": 1.0110333,
      "balance_loss_mlp": 1.00266445,
      "epoch": 0.7370889196176276,
      "flos": 70293343824000.0,
      "grad_norm": 0.9394468893660747,
      "language_loss": 0.56737435,
      "learning_rate": 6.821626727359606e-07,
      "loss": 0.5890708,
      "num_input_tokens_seen": 131776680,
      "step": 6130,
      "time_per_iteration": 3.287353277206421
    },
    {
      "auxiliary_loss_clip": 0.01280802,
      "auxiliary_loss_mlp": 0.01026374,
      "balance_loss_clip": 1.05293918,
      "balance_loss_mlp": 1.01822042,
      "epoch": 0.7372091625082667,
      "flos": 18040839189120.0,
      "grad_norm": 2.6969548329094777,
      "language_loss": 0.77351099,
      "learning_rate": 6.815768191778348e-07,
      "loss": 0.79658282,
      "num_input_tokens_seen": 131794760,
      "step": 6131,
      "time_per_iteration": 2.645496368408203
    },
    {
      "auxiliary_loss_clip": 0.01222434,
      "auxiliary_loss_mlp": 0.01028366,
      "balance_loss_clip": 1.04686737,
      "balance_loss_mlp": 1.02166677,
      "epoch": 0.7373294053989058,
      "flos": 33726331854720.0,
      "grad_norm": 1.823929737774238,
      "language_loss": 0.73134887,
      "learning_rate": 6.809911656234569e-07,
      "loss": 0.75385684,
      "num_input_tokens_seen": 131816735,
      "step": 6132,
      "time_per_iteration": 2.7448415756225586
    },
    {
      "auxiliary_loss_clip": 0.01326862,
      "auxiliary_loss_mlp": 0.01023964,
      "balance_loss_clip": 1.04410803,
      "balance_loss_mlp": 1.01721334,
      "epoch": 0.7374496482895448,
      "flos": 21506326427520.0,
      "grad_norm": 2.0918267042129792,
      "language_loss": 0.78255361,
      "learning_rate": 6.804057121616707e-07,
      "loss": 0.80606186,
      "num_input_tokens_seen": 131834940,
      "step": 6133,
      "time_per_iteration": 2.726217269897461
    },
    {
      "auxiliary_loss_clip": 0.0122596,
      "auxiliary_loss_mlp": 0.01026774,
      "balance_loss_clip": 1.04891813,
      "balance_loss_mlp": 1.02001143,
      "epoch": 0.737569891180184,
      "flos": 24936908624640.0,
      "grad_norm": 2.4874079412583274,
      "language_loss": 0.72425199,
      "learning_rate": 6.798204588812888e-07,
      "loss": 0.74677932,
      "num_input_tokens_seen": 131854355,
      "step": 6134,
      "time_per_iteration": 2.6369259357452393
    },
    {
      "auxiliary_loss_clip": 0.01414175,
      "auxiliary_loss_mlp": 0.02565524,
      "balance_loss_clip": 1.03984928,
      "balance_loss_mlp": 0.9999311,
      "epoch": 0.7376901340708231,
      "flos": 20664544222080.0,
      "grad_norm": 1.6955628475015843,
      "language_loss": 0.75494182,
      "learning_rate": 6.792354058710937e-07,
      "loss": 0.79473877,
      "num_input_tokens_seen": 131871825,
      "step": 6135,
      "time_per_iteration": 2.7794926166534424
    },
    {
      "auxiliary_loss_clip": 0.0116923,
      "auxiliary_loss_mlp": 0.01023126,
      "balance_loss_clip": 1.04806852,
      "balance_loss_mlp": 1.01662588,
      "epoch": 0.7378103769614621,
      "flos": 23805794367360.0,
      "grad_norm": 1.794153033073207,
      "language_loss": 0.65202695,
      "learning_rate": 6.786505532198374e-07,
      "loss": 0.67395049,
      "num_input_tokens_seen": 131890770,
      "step": 6136,
      "time_per_iteration": 2.639497756958008
    },
    {
      "auxiliary_loss_clip": 0.01176191,
      "auxiliary_loss_mlp": 0.01028068,
      "balance_loss_clip": 1.05068636,
      "balance_loss_mlp": 1.02039409,
      "epoch": 0.7379306198521013,
      "flos": 22237216369920.0,
      "grad_norm": 2.0862137506075764,
      "language_loss": 0.85393012,
      "learning_rate": 6.780659010162411e-07,
      "loss": 0.87597269,
      "num_input_tokens_seen": 131909720,
      "step": 6137,
      "time_per_iteration": 2.6010000705718994
    },
    {
      "auxiliary_loss_clip": 0.01326651,
      "auxiliary_loss_mlp": 0.01031084,
      "balance_loss_clip": 1.04839587,
      "balance_loss_mlp": 1.0244801,
      "epoch": 0.7380508627427403,
      "flos": 14903108576640.0,
      "grad_norm": 1.8507523683339795,
      "language_loss": 0.8311801,
      "learning_rate": 6.774814493489975e-07,
      "loss": 0.85475749,
      "num_input_tokens_seen": 131927395,
      "step": 6138,
      "time_per_iteration": 2.7583494186401367
    },
    {
      "auxiliary_loss_clip": 0.01221729,
      "auxiliary_loss_mlp": 0.01021796,
      "balance_loss_clip": 1.04988909,
      "balance_loss_mlp": 1.01497722,
      "epoch": 0.7381711056333794,
      "flos": 21685843624320.0,
      "grad_norm": 3.3866301240210044,
      "language_loss": 0.6611588,
      "learning_rate": 6.768971983067655e-07,
      "loss": 0.68359405,
      "num_input_tokens_seen": 131947725,
      "step": 6139,
      "time_per_iteration": 4.480640888214111
    },
    {
      "auxiliary_loss_clip": 0.01066131,
      "auxiliary_loss_mlp": 0.01003366,
      "balance_loss_clip": 1.01061654,
      "balance_loss_mlp": 1.00240076,
      "epoch": 0.7382913485240186,
      "flos": 52404263596800.0,
      "grad_norm": 1.0109881313846756,
      "language_loss": 0.67714351,
      "learning_rate": 6.763131479781772e-07,
      "loss": 0.69783849,
      "num_input_tokens_seen": 131997485,
      "step": 6140,
      "time_per_iteration": 3.008915424346924
    },
    {
      "auxiliary_loss_clip": 0.01265829,
      "auxiliary_loss_mlp": 0.01027469,
      "balance_loss_clip": 1.04726958,
      "balance_loss_mlp": 1.02111566,
      "epoch": 0.7384115914146576,
      "flos": 21798818876160.0,
      "grad_norm": 1.823764479746317,
      "language_loss": 0.76238334,
      "learning_rate": 6.757292984518316e-07,
      "loss": 0.78531623,
      "num_input_tokens_seen": 132016885,
      "step": 6141,
      "time_per_iteration": 3.5446746349334717
    },
    {
      "auxiliary_loss_clip": 0.01119399,
      "auxiliary_loss_mlp": 0.01001863,
      "balance_loss_clip": 1.01069343,
      "balance_loss_mlp": 1.00093961,
      "epoch": 0.7385318343052967,
      "flos": 61494331662720.0,
      "grad_norm": 0.7458236799814033,
      "language_loss": 0.56339931,
      "learning_rate": 6.751456498162981e-07,
      "loss": 0.58461195,
      "num_input_tokens_seen": 132075920,
      "step": 6142,
      "time_per_iteration": 3.0842173099517822
    },
    {
      "auxiliary_loss_clip": 0.01222305,
      "auxiliary_loss_mlp": 0.01024382,
      "balance_loss_clip": 1.04613316,
      "balance_loss_mlp": 1.01771784,
      "epoch": 0.7386520771959358,
      "flos": 17013757697280.0,
      "grad_norm": 4.5428663112970815,
      "language_loss": 0.84983462,
      "learning_rate": 6.745622021601174e-07,
      "loss": 0.87230146,
      "num_input_tokens_seen": 132092945,
      "step": 6143,
      "time_per_iteration": 2.6339495182037354
    },
    {
      "auxiliary_loss_clip": 0.01323608,
      "auxiliary_loss_mlp": 0.01021201,
      "balance_loss_clip": 1.04553175,
      "balance_loss_mlp": 1.01483846,
      "epoch": 0.7387723200865749,
      "flos": 18770759464320.0,
      "grad_norm": 1.783568945529018,
      "language_loss": 0.699049,
      "learning_rate": 6.739789555717954e-07,
      "loss": 0.72249705,
      "num_input_tokens_seen": 132109920,
      "step": 6144,
      "time_per_iteration": 2.667623281478882
    },
    {
      "auxiliary_loss_clip": 0.01171896,
      "auxiliary_loss_mlp": 0.01021236,
      "balance_loss_clip": 1.04770923,
      "balance_loss_mlp": 1.01480806,
      "epoch": 0.738892562977214,
      "flos": 22525542840960.0,
      "grad_norm": 1.980593886583483,
      "language_loss": 0.77783662,
      "learning_rate": 6.733959101398124e-07,
      "loss": 0.79976797,
      "num_input_tokens_seen": 132128050,
      "step": 6145,
      "time_per_iteration": 2.5931572914123535
    },
    {
      "auxiliary_loss_clip": 0.01275832,
      "auxiliary_loss_mlp": 0.0102805,
      "balance_loss_clip": 1.04601026,
      "balance_loss_mlp": 1.02125537,
      "epoch": 0.7390128058678531,
      "flos": 21501478091520.0,
      "grad_norm": 1.7005202565465418,
      "language_loss": 0.81334823,
      "learning_rate": 6.728130659526143e-07,
      "loss": 0.83638704,
      "num_input_tokens_seen": 132145860,
      "step": 6146,
      "time_per_iteration": 2.685086488723755
    },
    {
      "auxiliary_loss_clip": 0.0127521,
      "auxiliary_loss_mlp": 0.01023083,
      "balance_loss_clip": 1.04734564,
      "balance_loss_mlp": 1.0163691,
      "epoch": 0.7391330487584922,
      "flos": 25776176878080.0,
      "grad_norm": 2.444263799134435,
      "language_loss": 0.71669495,
      "learning_rate": 6.7223042309862e-07,
      "loss": 0.73967791,
      "num_input_tokens_seen": 132166060,
      "step": 6147,
      "time_per_iteration": 2.664506196975708
    },
    {
      "auxiliary_loss_clip": 0.01220653,
      "auxiliary_loss_mlp": 0.0102123,
      "balance_loss_clip": 1.04533958,
      "balance_loss_mlp": 1.01474237,
      "epoch": 0.7392532916491312,
      "flos": 28366736636160.0,
      "grad_norm": 3.587018191390677,
      "language_loss": 0.73699939,
      "learning_rate": 6.716479816662144e-07,
      "loss": 0.75941813,
      "num_input_tokens_seen": 132187790,
      "step": 6148,
      "time_per_iteration": 3.632021903991699
    },
    {
      "auxiliary_loss_clip": 0.01279813,
      "auxiliary_loss_mlp": 0.01031738,
      "balance_loss_clip": 1.04808724,
      "balance_loss_mlp": 1.02503824,
      "epoch": 0.7393735345397703,
      "flos": 23585877348480.0,
      "grad_norm": 4.074505306889431,
      "language_loss": 0.73427629,
      "learning_rate": 6.710657417437531e-07,
      "loss": 0.75739181,
      "num_input_tokens_seen": 132207495,
      "step": 6149,
      "time_per_iteration": 2.6814775466918945
    },
    {
      "auxiliary_loss_clip": 0.01274436,
      "auxiliary_loss_mlp": 0.01019802,
      "balance_loss_clip": 1.04700732,
      "balance_loss_mlp": 1.01320386,
      "epoch": 0.7394937774304094,
      "flos": 19974772373760.0,
      "grad_norm": 2.4567580256234525,
      "language_loss": 0.80310583,
      "learning_rate": 6.704837034195628e-07,
      "loss": 0.82604825,
      "num_input_tokens_seen": 132225960,
      "step": 6150,
      "time_per_iteration": 2.658297061920166
    },
    {
      "auxiliary_loss_clip": 0.01221862,
      "auxiliary_loss_mlp": 0.01026905,
      "balance_loss_clip": 1.04824162,
      "balance_loss_mlp": 1.02011299,
      "epoch": 0.7396140203210485,
      "flos": 23478037741440.0,
      "grad_norm": 2.001429538461752,
      "language_loss": 0.85305643,
      "learning_rate": 6.699018667819376e-07,
      "loss": 0.87554407,
      "num_input_tokens_seen": 132245360,
      "step": 6151,
      "time_per_iteration": 2.6916420459747314
    },
    {
      "auxiliary_loss_clip": 0.01223539,
      "auxiliary_loss_mlp": 0.01022,
      "balance_loss_clip": 1.04714322,
      "balance_loss_mlp": 1.0143702,
      "epoch": 0.7397342632116876,
      "flos": 25555433846400.0,
      "grad_norm": 1.7110665086086185,
      "language_loss": 0.72714448,
      "learning_rate": 6.693202319191415e-07,
      "loss": 0.74959981,
      "num_input_tokens_seen": 132267095,
      "step": 6152,
      "time_per_iteration": 2.6567742824554443
    },
    {
      "auxiliary_loss_clip": 0.01179154,
      "auxiliary_loss_mlp": 0.01031034,
      "balance_loss_clip": 1.05524516,
      "balance_loss_mlp": 1.02428341,
      "epoch": 0.7398545061023267,
      "flos": 24755021130240.0,
      "grad_norm": 1.7872100585442539,
      "language_loss": 0.74980086,
      "learning_rate": 6.687387989194084e-07,
      "loss": 0.77190274,
      "num_input_tokens_seen": 132286610,
      "step": 6153,
      "time_per_iteration": 2.6413331031799316
    },
    {
      "auxiliary_loss_clip": 0.01273732,
      "auxiliary_loss_mlp": 0.01028221,
      "balance_loss_clip": 1.04993224,
      "balance_loss_mlp": 1.02125371,
      "epoch": 0.7399747489929658,
      "flos": 16508602776960.0,
      "grad_norm": 2.3448350562895706,
      "language_loss": 0.7966342,
      "learning_rate": 6.681575678709404e-07,
      "loss": 0.81965369,
      "num_input_tokens_seen": 132305300,
      "step": 6154,
      "time_per_iteration": 2.641674518585205
    },
    {
      "auxiliary_loss_clip": 0.0122192,
      "auxiliary_loss_mlp": 0.0102609,
      "balance_loss_clip": 1.04843533,
      "balance_loss_mlp": 1.01905346,
      "epoch": 0.7400949918836048,
      "flos": 24097065753600.0,
      "grad_norm": 3.474892345327748,
      "language_loss": 0.70847154,
      "learning_rate": 6.67576538861911e-07,
      "loss": 0.73095161,
      "num_input_tokens_seen": 132323875,
      "step": 6155,
      "time_per_iteration": 2.671633243560791
    },
    {
      "auxiliary_loss_clip": 0.01271057,
      "auxiliary_loss_mlp": 0.0102409,
      "balance_loss_clip": 1.04686117,
      "balance_loss_mlp": 1.01781034,
      "epoch": 0.740215234774244,
      "flos": 21802517976960.0,
      "grad_norm": 3.2653861480658777,
      "language_loss": 0.82232809,
      "learning_rate": 6.669957119804612e-07,
      "loss": 0.84527951,
      "num_input_tokens_seen": 132345510,
      "step": 6156,
      "time_per_iteration": 2.7344629764556885
    },
    {
      "auxiliary_loss_clip": 0.01281902,
      "auxiliary_loss_mlp": 0.01024813,
      "balance_loss_clip": 1.04795408,
      "balance_loss_mlp": 1.01729083,
      "epoch": 0.7403354776648831,
      "flos": 18733196816640.0,
      "grad_norm": 5.119572515413475,
      "language_loss": 0.72654814,
      "learning_rate": 6.66415087314702e-07,
      "loss": 0.74961525,
      "num_input_tokens_seen": 132360465,
      "step": 6157,
      "time_per_iteration": 2.669663190841675
    },
    {
      "auxiliary_loss_clip": 0.01276483,
      "auxiliary_loss_mlp": 0.01029142,
      "balance_loss_clip": 1.04763985,
      "balance_loss_mlp": 1.02284491,
      "epoch": 0.7404557205555221,
      "flos": 16909581277440.0,
      "grad_norm": 2.56401227761443,
      "language_loss": 0.73168218,
      "learning_rate": 6.65834664952714e-07,
      "loss": 0.75473845,
      "num_input_tokens_seen": 132377915,
      "step": 6158,
      "time_per_iteration": 2.672999620437622
    },
    {
      "auxiliary_loss_clip": 0.01325712,
      "auxiliary_loss_mlp": 0.01022391,
      "balance_loss_clip": 1.04617643,
      "balance_loss_mlp": 1.0152626,
      "epoch": 0.7405759634461613,
      "flos": 21214408596480.0,
      "grad_norm": 1.596677173055138,
      "language_loss": 0.76052791,
      "learning_rate": 6.652544449825457e-07,
      "loss": 0.78400892,
      "num_input_tokens_seen": 132398170,
      "step": 6159,
      "time_per_iteration": 2.7332968711853027
    },
    {
      "auxiliary_loss_clip": 0.01180264,
      "auxiliary_loss_mlp": 0.01031656,
      "balance_loss_clip": 1.0454495,
      "balance_loss_mlp": 1.02429533,
      "epoch": 0.7406962063368003,
      "flos": 20480106862080.0,
      "grad_norm": 1.8921460118562172,
      "language_loss": 0.76471478,
      "learning_rate": 6.646744274922182e-07,
      "loss": 0.78683394,
      "num_input_tokens_seen": 132416615,
      "step": 6160,
      "time_per_iteration": 2.6834092140197754
    },
    {
      "auxiliary_loss_clip": 0.01274084,
      "auxiliary_loss_mlp": 0.01020489,
      "balance_loss_clip": 1.0456152,
      "balance_loss_mlp": 1.01423669,
      "epoch": 0.7408164492274394,
      "flos": 19791915212160.0,
      "grad_norm": 2.830884886145923,
      "language_loss": 0.75762028,
      "learning_rate": 6.640946125697171e-07,
      "loss": 0.78056604,
      "num_input_tokens_seen": 132434145,
      "step": 6161,
      "time_per_iteration": 2.6767778396606445
    },
    {
      "auxiliary_loss_clip": 0.01222864,
      "auxiliary_loss_mlp": 0.0102351,
      "balance_loss_clip": 1.04636574,
      "balance_loss_mlp": 1.01629817,
      "epoch": 0.7409366921180786,
      "flos": 29204855654400.0,
      "grad_norm": 2.048382047129729,
      "language_loss": 0.75823843,
      "learning_rate": 6.635150003030017e-07,
      "loss": 0.78070211,
      "num_input_tokens_seen": 132452670,
      "step": 6162,
      "time_per_iteration": 2.6648457050323486
    },
    {
      "auxiliary_loss_clip": 0.0127116,
      "auxiliary_loss_mlp": 0.01020217,
      "balance_loss_clip": 1.04038954,
      "balance_loss_mlp": 1.01361597,
      "epoch": 0.7410569350087176,
      "flos": 22930004960640.0,
      "grad_norm": 2.3936254852588013,
      "language_loss": 0.8580609,
      "learning_rate": 6.629355907799981e-07,
      "loss": 0.88097465,
      "num_input_tokens_seen": 132472475,
      "step": 6163,
      "time_per_iteration": 2.7692723274230957
    },
    {
      "auxiliary_loss_clip": 0.01225629,
      "auxiliary_loss_mlp": 0.01023832,
      "balance_loss_clip": 1.04740405,
      "balance_loss_mlp": 1.01701021,
      "epoch": 0.7411771778993567,
      "flos": 30440397726720.0,
      "grad_norm": 1.8190722498444787,
      "language_loss": 0.69331247,
      "learning_rate": 6.623563840886015e-07,
      "loss": 0.71580708,
      "num_input_tokens_seen": 132493400,
      "step": 6164,
      "time_per_iteration": 2.676262855529785
    },
    {
      "auxiliary_loss_clip": 0.01219354,
      "auxiliary_loss_mlp": 0.01020503,
      "balance_loss_clip": 1.04502845,
      "balance_loss_mlp": 1.01417649,
      "epoch": 0.7412974207899958,
      "flos": 20522050968960.0,
      "grad_norm": 2.1389580174744296,
      "language_loss": 0.69568503,
      "learning_rate": 6.617773803166795e-07,
      "loss": 0.71808362,
      "num_input_tokens_seen": 132511725,
      "step": 6165,
      "time_per_iteration": 4.522045135498047
    },
    {
      "auxiliary_loss_clip": 0.01278159,
      "auxiliary_loss_mlp": 0.02569993,
      "balance_loss_clip": 1.04896283,
      "balance_loss_mlp": 0.99993038,
      "epoch": 0.7414176636806349,
      "flos": 22090700793600.0,
      "grad_norm": 2.2186821490042443,
      "language_loss": 0.82033646,
      "learning_rate": 6.611985795520634e-07,
      "loss": 0.85881799,
      "num_input_tokens_seen": 132530270,
      "step": 6166,
      "time_per_iteration": 2.7216055393218994
    },
    {
      "auxiliary_loss_clip": 0.01331449,
      "auxiliary_loss_mlp": 0.01021186,
      "balance_loss_clip": 1.04783034,
      "balance_loss_mlp": 1.01433754,
      "epoch": 0.7415379065712739,
      "flos": 25155245445120.0,
      "grad_norm": 2.2711157169518144,
      "language_loss": 0.77351958,
      "learning_rate": 6.606199818825588e-07,
      "loss": 0.79704595,
      "num_input_tokens_seen": 132550725,
      "step": 6167,
      "time_per_iteration": 3.6228413581848145
    },
    {
      "auxiliary_loss_clip": 0.01277644,
      "auxiliary_loss_mlp": 0.01021742,
      "balance_loss_clip": 1.04461598,
      "balance_loss_mlp": 1.01485229,
      "epoch": 0.7416581494619131,
      "flos": 16871731320960.0,
      "grad_norm": 2.038272196802235,
      "language_loss": 0.81994104,
      "learning_rate": 6.600415873959377e-07,
      "loss": 0.84293497,
      "num_input_tokens_seen": 132568600,
      "step": 6168,
      "time_per_iteration": 2.6064140796661377
    },
    {
      "auxiliary_loss_clip": 0.01426614,
      "auxiliary_loss_mlp": 0.0255972,
      "balance_loss_clip": 1.04249763,
      "balance_loss_mlp": 0.99995553,
      "epoch": 0.7417783923525522,
      "flos": 28438881102720.0,
      "grad_norm": 2.2479224658344052,
      "language_loss": 0.64751232,
      "learning_rate": 6.594633961799437e-07,
      "loss": 0.68737561,
      "num_input_tokens_seen": 132587640,
      "step": 6169,
      "time_per_iteration": 2.8245465755462646
    },
    {
      "auxiliary_loss_clip": 0.01331665,
      "auxiliary_loss_mlp": 0.01027962,
      "balance_loss_clip": 1.04672706,
      "balance_loss_mlp": 1.02078271,
      "epoch": 0.7418986352431912,
      "flos": 20084299920000.0,
      "grad_norm": 2.0757582409588324,
      "language_loss": 0.8150543,
      "learning_rate": 6.588854083222857e-07,
      "loss": 0.83865058,
      "num_input_tokens_seen": 132607075,
      "step": 6170,
      "time_per_iteration": 2.6961748600006104
    },
    {
      "auxiliary_loss_clip": 0.01279114,
      "auxiliary_loss_mlp": 0.01027381,
      "balance_loss_clip": 1.04889083,
      "balance_loss_mlp": 1.02037752,
      "epoch": 0.7420188781338304,
      "flos": 18259571059200.0,
      "grad_norm": 2.2072623741189608,
      "language_loss": 0.81112421,
      "learning_rate": 6.583076239106444e-07,
      "loss": 0.83418918,
      "num_input_tokens_seen": 132625580,
      "step": 6171,
      "time_per_iteration": 2.652164936065674
    },
    {
      "auxiliary_loss_clip": 0.0127874,
      "auxiliary_loss_mlp": 0.01025844,
      "balance_loss_clip": 1.04799271,
      "balance_loss_mlp": 1.01853609,
      "epoch": 0.7421391210244694,
      "flos": 13771994319360.0,
      "grad_norm": 2.477547769258423,
      "language_loss": 0.75620103,
      "learning_rate": 6.577300430326707e-07,
      "loss": 0.77924693,
      "num_input_tokens_seen": 132640525,
      "step": 6172,
      "time_per_iteration": 2.6244242191314697
    },
    {
      "auxiliary_loss_clip": 0.01319017,
      "auxiliary_loss_mlp": 0.01025506,
      "balance_loss_clip": 1.04579306,
      "balance_loss_mlp": 1.01855576,
      "epoch": 0.7422593639151085,
      "flos": 15961683317760.0,
      "grad_norm": 2.263507599012854,
      "language_loss": 0.72378159,
      "learning_rate": 6.571526657759821e-07,
      "loss": 0.74722683,
      "num_input_tokens_seen": 132656265,
      "step": 6173,
      "time_per_iteration": 2.6899800300598145
    },
    {
      "auxiliary_loss_clip": 0.01219657,
      "auxiliary_loss_mlp": 0.01028081,
      "balance_loss_clip": 1.04526389,
      "balance_loss_mlp": 1.02087772,
      "epoch": 0.7423796068057477,
      "flos": 30114400867200.0,
      "grad_norm": 1.6654101620058839,
      "language_loss": 0.70658493,
      "learning_rate": 6.565754922281663e-07,
      "loss": 0.72906232,
      "num_input_tokens_seen": 132678510,
      "step": 6174,
      "time_per_iteration": 3.57859206199646
    },
    {
      "auxiliary_loss_clip": 0.01277505,
      "auxiliary_loss_mlp": 0.01024643,
      "balance_loss_clip": 1.04678655,
      "balance_loss_mlp": 1.0176127,
      "epoch": 0.7424998496963867,
      "flos": 20521907314560.0,
      "grad_norm": 3.495495032736239,
      "language_loss": 0.78585285,
      "learning_rate": 6.559985224767801e-07,
      "loss": 0.80887431,
      "num_input_tokens_seen": 132696385,
      "step": 6175,
      "time_per_iteration": 2.6502761840820312
    },
    {
      "auxiliary_loss_clip": 0.01228208,
      "auxiliary_loss_mlp": 0.01031397,
      "balance_loss_clip": 1.04811358,
      "balance_loss_mlp": 1.02429497,
      "epoch": 0.7426200925870258,
      "flos": 21871573873920.0,
      "grad_norm": 2.831937810618245,
      "language_loss": 0.7576409,
      "learning_rate": 6.55421756609349e-07,
      "loss": 0.7802369,
      "num_input_tokens_seen": 132714640,
      "step": 6176,
      "time_per_iteration": 2.7221741676330566
    },
    {
      "auxiliary_loss_clip": 0.01228938,
      "auxiliary_loss_mlp": 0.0102843,
      "balance_loss_clip": 1.05363166,
      "balance_loss_mlp": 1.02160835,
      "epoch": 0.7427403354776649,
      "flos": 26432049265920.0,
      "grad_norm": 2.3554204455130003,
      "language_loss": 0.78962564,
      "learning_rate": 6.54845194713369e-07,
      "loss": 0.81219935,
      "num_input_tokens_seen": 132735590,
      "step": 6177,
      "time_per_iteration": 2.602095603942871
    },
    {
      "auxiliary_loss_clip": 0.01223014,
      "auxiliary_loss_mlp": 0.01026549,
      "balance_loss_clip": 1.048437,
      "balance_loss_mlp": 1.01971197,
      "epoch": 0.742860578368304,
      "flos": 19898390102400.0,
      "grad_norm": 2.5574737756543824,
      "language_loss": 0.79952013,
      "learning_rate": 6.542688368763034e-07,
      "loss": 0.82201576,
      "num_input_tokens_seen": 132753995,
      "step": 6178,
      "time_per_iteration": 2.6224911212921143
    },
    {
      "auxiliary_loss_clip": 0.01221978,
      "auxiliary_loss_mlp": 0.0102498,
      "balance_loss_clip": 1.05015481,
      "balance_loss_mlp": 1.01833093,
      "epoch": 0.742980821258943,
      "flos": 24827201510400.0,
      "grad_norm": 2.5280376847795893,
      "language_loss": 0.76975441,
      "learning_rate": 6.536926831855854e-07,
      "loss": 0.79222393,
      "num_input_tokens_seen": 132773160,
      "step": 6179,
      "time_per_iteration": 2.6224327087402344
    },
    {
      "auxiliary_loss_clip": 0.01266488,
      "auxiliary_loss_mlp": 0.01025161,
      "balance_loss_clip": 1.04732084,
      "balance_loss_mlp": 1.01818109,
      "epoch": 0.7431010641495821,
      "flos": 25228646887680.0,
      "grad_norm": 2.3107245435179453,
      "language_loss": 0.73248649,
      "learning_rate": 6.531167337286165e-07,
      "loss": 0.75540292,
      "num_input_tokens_seen": 132793180,
      "step": 6180,
      "time_per_iteration": 2.6970276832580566
    },
    {
      "auxiliary_loss_clip": 0.01268142,
      "auxiliary_loss_mlp": 0.01025082,
      "balance_loss_clip": 1.04822397,
      "balance_loss_mlp": 1.01883268,
      "epoch": 0.7432213070402213,
      "flos": 21762369550080.0,
      "grad_norm": 1.7152318984043387,
      "language_loss": 0.79823136,
      "learning_rate": 6.52540988592768e-07,
      "loss": 0.82116365,
      "num_input_tokens_seen": 132814200,
      "step": 6181,
      "time_per_iteration": 2.668212890625
    },
    {
      "auxiliary_loss_clip": 0.01274834,
      "auxiliary_loss_mlp": 0.01021977,
      "balance_loss_clip": 1.04740286,
      "balance_loss_mlp": 1.01509535,
      "epoch": 0.7433415499308603,
      "flos": 14793832425600.0,
      "grad_norm": 2.0547363880467113,
      "language_loss": 0.83476758,
      "learning_rate": 6.519654478653814e-07,
      "loss": 0.85773569,
      "num_input_tokens_seen": 132832565,
      "step": 6182,
      "time_per_iteration": 2.739183187484741
    },
    {
      "auxiliary_loss_clip": 0.01176791,
      "auxiliary_loss_mlp": 0.00999544,
      "balance_loss_clip": 1.01084042,
      "balance_loss_mlp": 0.99863797,
      "epoch": 0.7434617928214994,
      "flos": 67155577297920.0,
      "grad_norm": 0.7461348572111697,
      "language_loss": 0.56071663,
      "learning_rate": 6.51390111633763e-07,
      "loss": 0.58247995,
      "num_input_tokens_seen": 132897840,
      "step": 6183,
      "time_per_iteration": 3.2973554134368896
    },
    {
      "auxiliary_loss_clip": 0.01425856,
      "auxiliary_loss_mlp": 0.01027034,
      "balance_loss_clip": 1.0408442,
      "balance_loss_mlp": 1.02012599,
      "epoch": 0.7435820357121385,
      "flos": 27377576928000.0,
      "grad_norm": 1.658433612952696,
      "language_loss": 0.76211667,
      "learning_rate": 6.508149799851932e-07,
      "loss": 0.78664559,
      "num_input_tokens_seen": 132919505,
      "step": 6184,
      "time_per_iteration": 2.840946674346924
    },
    {
      "auxiliary_loss_clip": 0.01268423,
      "auxiliary_loss_mlp": 0.01023271,
      "balance_loss_clip": 1.04721832,
      "balance_loss_mlp": 1.01697361,
      "epoch": 0.7437022786027776,
      "flos": 23987645948160.0,
      "grad_norm": 2.7391642297892673,
      "language_loss": 0.61581111,
      "learning_rate": 6.502400530069183e-07,
      "loss": 0.63872808,
      "num_input_tokens_seen": 132939390,
      "step": 6185,
      "time_per_iteration": 2.7304325103759766
    },
    {
      "auxiliary_loss_clip": 0.01323687,
      "auxiliary_loss_mlp": 0.01027723,
      "balance_loss_clip": 1.04771972,
      "balance_loss_mlp": 1.01987922,
      "epoch": 0.7438225214934167,
      "flos": 21866761451520.0,
      "grad_norm": 1.9796226764208247,
      "language_loss": 0.68382335,
      "learning_rate": 6.496653307861535e-07,
      "loss": 0.70733738,
      "num_input_tokens_seen": 132960060,
      "step": 6186,
      "time_per_iteration": 2.732464075088501
    },
    {
      "auxiliary_loss_clip": 0.01230798,
      "auxiliary_loss_mlp": 0.01029067,
      "balance_loss_clip": 1.04888511,
      "balance_loss_mlp": 1.02232289,
      "epoch": 0.7439427643840558,
      "flos": 20230097224320.0,
      "grad_norm": 1.8643467178595983,
      "language_loss": 0.66104394,
      "learning_rate": 6.490908134100857e-07,
      "loss": 0.68364263,
      "num_input_tokens_seen": 132978525,
      "step": 6187,
      "time_per_iteration": 2.6495859622955322
    },
    {
      "auxiliary_loss_clip": 0.01229755,
      "auxiliary_loss_mlp": 0.01026075,
      "balance_loss_clip": 1.04971886,
      "balance_loss_mlp": 1.0188632,
      "epoch": 0.7440630072746949,
      "flos": 20849915335680.0,
      "grad_norm": 4.223440423536163,
      "language_loss": 0.69228286,
      "learning_rate": 6.48516500965866e-07,
      "loss": 0.71484113,
      "num_input_tokens_seen": 132998460,
      "step": 6188,
      "time_per_iteration": 2.6401450634002686
    },
    {
      "auxiliary_loss_clip": 0.01228842,
      "auxiliary_loss_mlp": 0.01027292,
      "balance_loss_clip": 1.04731894,
      "balance_loss_mlp": 1.01982021,
      "epoch": 0.7441832501653339,
      "flos": 26503762769280.0,
      "grad_norm": 2.154401496009429,
      "language_loss": 0.81692988,
      "learning_rate": 6.479423935406192e-07,
      "loss": 0.83949125,
      "num_input_tokens_seen": 133018445,
      "step": 6189,
      "time_per_iteration": 2.6666197776794434
    },
    {
      "auxiliary_loss_clip": 0.01165353,
      "auxiliary_loss_mlp": 0.0100587,
      "balance_loss_clip": 1.0128324,
      "balance_loss_mlp": 1.00504708,
      "epoch": 0.7443034930559731,
      "flos": 68602848088320.0,
      "grad_norm": 0.8110917338304526,
      "language_loss": 0.61974454,
      "learning_rate": 6.473684912214357e-07,
      "loss": 0.64145678,
      "num_input_tokens_seen": 133082005,
      "step": 6190,
      "time_per_iteration": 3.3562440872192383
    },
    {
      "auxiliary_loss_clip": 0.01223233,
      "auxiliary_loss_mlp": 0.01028389,
      "balance_loss_clip": 1.04996681,
      "balance_loss_mlp": 1.02177286,
      "epoch": 0.7444237359466122,
      "flos": 18654982951680.0,
      "grad_norm": 2.4405602781206595,
      "language_loss": 0.69138944,
      "learning_rate": 6.467947940953778e-07,
      "loss": 0.71390563,
      "num_input_tokens_seen": 133100530,
      "step": 6191,
      "time_per_iteration": 4.377862215042114
    },
    {
      "auxiliary_loss_clip": 0.01269473,
      "auxiliary_loss_mlp": 0.01023145,
      "balance_loss_clip": 1.04581237,
      "balance_loss_mlp": 1.01718783,
      "epoch": 0.7445439788372512,
      "flos": 22817604326400.0,
      "grad_norm": 1.9023915620191663,
      "language_loss": 0.72610456,
      "learning_rate": 6.462213022494732e-07,
      "loss": 0.74903071,
      "num_input_tokens_seen": 133119775,
      "step": 6192,
      "time_per_iteration": 2.7209227085113525
    },
    {
      "auxiliary_loss_clip": 0.01122267,
      "auxiliary_loss_mlp": 0.01001529,
      "balance_loss_clip": 1.01050997,
      "balance_loss_mlp": 1.00065255,
      "epoch": 0.7446642217278904,
      "flos": 67045690615680.0,
      "grad_norm": 0.7687749830276606,
      "language_loss": 0.60979927,
      "learning_rate": 6.456480157707201e-07,
      "loss": 0.63103724,
      "num_input_tokens_seen": 133184550,
      "step": 6193,
      "time_per_iteration": 4.0394651889801025
    },
    {
      "auxiliary_loss_clip": 0.01318939,
      "auxiliary_loss_mlp": 0.01024032,
      "balance_loss_clip": 1.04300511,
      "balance_loss_mlp": 1.01739502,
      "epoch": 0.7447844646185294,
      "flos": 17417465631360.0,
      "grad_norm": 2.1666616921281534,
      "language_loss": 0.85046244,
      "learning_rate": 6.450749347460866e-07,
      "loss": 0.87389207,
      "num_input_tokens_seen": 133201525,
      "step": 6194,
      "time_per_iteration": 2.685598373413086
    },
    {
      "auxiliary_loss_clip": 0.01176473,
      "auxiliary_loss_mlp": 0.01025259,
      "balance_loss_clip": 1.05102038,
      "balance_loss_mlp": 1.01822519,
      "epoch": 0.7449047075091685,
      "flos": 26615876094720.0,
      "grad_norm": 1.934405758446165,
      "language_loss": 0.78885877,
      "learning_rate": 6.445020592625083e-07,
      "loss": 0.81087613,
      "num_input_tokens_seen": 133222175,
      "step": 6195,
      "time_per_iteration": 2.6710145473480225
    },
    {
      "auxiliary_loss_clip": 0.01172448,
      "auxiliary_loss_mlp": 0.01025239,
      "balance_loss_clip": 1.04809999,
      "balance_loss_mlp": 1.01871502,
      "epoch": 0.7450249503998077,
      "flos": 14170458867840.0,
      "grad_norm": 9.12028382521391,
      "language_loss": 0.80821812,
      "learning_rate": 6.4392938940689e-07,
      "loss": 0.83019495,
      "num_input_tokens_seen": 133237590,
      "step": 6196,
      "time_per_iteration": 2.5523126125335693
    },
    {
      "auxiliary_loss_clip": 0.01373751,
      "auxiliary_loss_mlp": 0.02566026,
      "balance_loss_clip": 1.04592419,
      "balance_loss_mlp": 0.99990934,
      "epoch": 0.7451451932904467,
      "flos": 19606687752960.0,
      "grad_norm": 2.4037299847384603,
      "language_loss": 0.71420819,
      "learning_rate": 6.433569252661049e-07,
      "loss": 0.75360596,
      "num_input_tokens_seen": 133255590,
      "step": 6197,
      "time_per_iteration": 2.817814588546753
    },
    {
      "auxiliary_loss_clip": 0.01315797,
      "auxiliary_loss_mlp": 0.01025841,
      "balance_loss_clip": 1.04520679,
      "balance_loss_mlp": 1.01973116,
      "epoch": 0.7452654361810858,
      "flos": 12495405980160.0,
      "grad_norm": 3.4111673806220972,
      "language_loss": 0.71398056,
      "learning_rate": 6.427846669269952e-07,
      "loss": 0.73739696,
      "num_input_tokens_seen": 133273210,
      "step": 6198,
      "time_per_iteration": 2.643432855606079
    },
    {
      "auxiliary_loss_clip": 0.01176859,
      "auxiliary_loss_mlp": 0.01028777,
      "balance_loss_clip": 1.05252433,
      "balance_loss_mlp": 1.02236319,
      "epoch": 0.7453856790717249,
      "flos": 22127329687680.0,
      "grad_norm": 2.040041871240816,
      "language_loss": 0.82699132,
      "learning_rate": 6.422126144763729e-07,
      "loss": 0.84904772,
      "num_input_tokens_seen": 133292600,
      "step": 6199,
      "time_per_iteration": 2.6025140285491943
    },
    {
      "auxiliary_loss_clip": 0.01318755,
      "auxiliary_loss_mlp": 0.02566607,
      "balance_loss_clip": 1.04193974,
      "balance_loss_mlp": 0.99990523,
      "epoch": 0.745505921962364,
      "flos": 20010682995840.0,
      "grad_norm": 2.0326685134013767,
      "language_loss": 0.76746434,
      "learning_rate": 6.416407680010174e-07,
      "loss": 0.80631804,
      "num_input_tokens_seen": 133306960,
      "step": 6200,
      "time_per_iteration": 3.626586437225342
    },
    {
      "auxiliary_loss_clip": 0.01289514,
      "auxiliary_loss_mlp": 0.01022746,
      "balance_loss_clip": 1.04990673,
      "balance_loss_mlp": 1.01586151,
      "epoch": 0.745626164853003,
      "flos": 24677884673280.0,
      "grad_norm": 2.2274175018283318,
      "language_loss": 0.8124699,
      "learning_rate": 6.410691275876774e-07,
      "loss": 0.83559251,
      "num_input_tokens_seen": 133326380,
      "step": 6201,
      "time_per_iteration": 2.7540652751922607
    },
    {
      "auxiliary_loss_clip": 0.01179649,
      "auxiliary_loss_mlp": 0.01033344,
      "balance_loss_clip": 1.04711366,
      "balance_loss_mlp": 1.02627456,
      "epoch": 0.7457464077436422,
      "flos": 14538830797440.0,
      "grad_norm": 2.547374993624888,
      "language_loss": 0.77227032,
      "learning_rate": 6.404976933230704e-07,
      "loss": 0.79440022,
      "num_input_tokens_seen": 133342900,
      "step": 6202,
      "time_per_iteration": 2.7227747440338135
    },
    {
      "auxiliary_loss_clip": 0.01279294,
      "auxiliary_loss_mlp": 0.01027422,
      "balance_loss_clip": 1.04722381,
      "balance_loss_mlp": 1.01978052,
      "epoch": 0.7458666506342813,
      "flos": 34021194600960.0,
      "grad_norm": 2.55403645182253,
      "language_loss": 0.72164202,
      "learning_rate": 6.399264652938813e-07,
      "loss": 0.74470925,
      "num_input_tokens_seen": 133363805,
      "step": 6203,
      "time_per_iteration": 2.7635414600372314
    },
    {
      "auxiliary_loss_clip": 0.01270774,
      "auxiliary_loss_mlp": 0.01026735,
      "balance_loss_clip": 1.04588246,
      "balance_loss_mlp": 1.01977944,
      "epoch": 0.7459868935249203,
      "flos": 24279025075200.0,
      "grad_norm": 1.843109752483123,
      "language_loss": 0.74374574,
      "learning_rate": 6.393554435867679e-07,
      "loss": 0.76672077,
      "num_input_tokens_seen": 133384655,
      "step": 6204,
      "time_per_iteration": 2.721463918685913
    },
    {
      "auxiliary_loss_clip": 0.01322944,
      "auxiliary_loss_mlp": 0.01025205,
      "balance_loss_clip": 1.04497194,
      "balance_loss_mlp": 1.01861262,
      "epoch": 0.7461071364155595,
      "flos": 21908777385600.0,
      "grad_norm": 2.00377586977192,
      "language_loss": 0.83668971,
      "learning_rate": 6.387846282883502e-07,
      "loss": 0.86017114,
      "num_input_tokens_seen": 133401185,
      "step": 6205,
      "time_per_iteration": 2.713993549346924
    },
    {
      "auxiliary_loss_clip": 0.01173231,
      "auxiliary_loss_mlp": 0.01025747,
      "balance_loss_clip": 1.0502615,
      "balance_loss_mlp": 1.01873708,
      "epoch": 0.7462273793061985,
      "flos": 22889712879360.0,
      "grad_norm": 2.9782315327146214,
      "language_loss": 0.76677418,
      "learning_rate": 6.38214019485223e-07,
      "loss": 0.78876394,
      "num_input_tokens_seen": 133420010,
      "step": 6206,
      "time_per_iteration": 2.649962902069092
    },
    {
      "auxiliary_loss_clip": 0.01415195,
      "auxiliary_loss_mlp": 0.01028604,
      "balance_loss_clip": 1.04042697,
      "balance_loss_mlp": 1.02190185,
      "epoch": 0.7463476221968376,
      "flos": 19968451580160.0,
      "grad_norm": 2.0737401019353467,
      "language_loss": 0.71448481,
      "learning_rate": 6.376436172639461e-07,
      "loss": 0.73892283,
      "num_input_tokens_seen": 133437855,
      "step": 6207,
      "time_per_iteration": 2.7121047973632812
    },
    {
      "auxiliary_loss_clip": 0.01383559,
      "auxiliary_loss_mlp": 0.01029899,
      "balance_loss_clip": 1.04437697,
      "balance_loss_mlp": 1.02215087,
      "epoch": 0.7464678650874768,
      "flos": 16836610798080.0,
      "grad_norm": 2.2295708699522834,
      "language_loss": 0.65512085,
      "learning_rate": 6.370734217110487e-07,
      "loss": 0.67925537,
      "num_input_tokens_seen": 133456600,
      "step": 6208,
      "time_per_iteration": 2.8439457416534424
    },
    {
      "auxiliary_loss_clip": 0.01280709,
      "auxiliary_loss_mlp": 0.01027385,
      "balance_loss_clip": 1.05317569,
      "balance_loss_mlp": 1.02000284,
      "epoch": 0.7465881079781158,
      "flos": 48100869843840.0,
      "grad_norm": 2.978742089231542,
      "language_loss": 0.64454889,
      "learning_rate": 6.36503432913031e-07,
      "loss": 0.66762984,
      "num_input_tokens_seen": 133479745,
      "step": 6209,
      "time_per_iteration": 3.1692657470703125
    },
    {
      "auxiliary_loss_clip": 0.01221919,
      "auxiliary_loss_mlp": 0.01026769,
      "balance_loss_clip": 1.04853606,
      "balance_loss_mlp": 1.01969099,
      "epoch": 0.7467083508687549,
      "flos": 19677359761920.0,
      "grad_norm": 2.2618764912246974,
      "language_loss": 0.69214255,
      "learning_rate": 6.359336509563569e-07,
      "loss": 0.71462941,
      "num_input_tokens_seen": 133495765,
      "step": 6210,
      "time_per_iteration": 2.6195778846740723
    },
    {
      "auxiliary_loss_clip": 0.01314187,
      "auxiliary_loss_mlp": 0.01022804,
      "balance_loss_clip": 1.04413903,
      "balance_loss_mlp": 1.01583028,
      "epoch": 0.7468285937593939,
      "flos": 17895436934400.0,
      "grad_norm": 1.8559249344465174,
      "language_loss": 0.80747175,
      "learning_rate": 6.353640759274641e-07,
      "loss": 0.83084166,
      "num_input_tokens_seen": 133514655,
      "step": 6211,
      "time_per_iteration": 2.701646566390991
    },
    {
      "auxiliary_loss_clip": 0.01223505,
      "auxiliary_loss_mlp": 0.01027107,
      "balance_loss_clip": 1.04617643,
      "balance_loss_mlp": 1.02055383,
      "epoch": 0.7469488366500331,
      "flos": 23141446369920.0,
      "grad_norm": 2.686094586760584,
      "language_loss": 0.74481332,
      "learning_rate": 6.347947079127556e-07,
      "loss": 0.7673195,
      "num_input_tokens_seen": 133532555,
      "step": 6212,
      "time_per_iteration": 2.592686176300049
    },
    {
      "auxiliary_loss_clip": 0.01272837,
      "auxiliary_loss_mlp": 0.01022111,
      "balance_loss_clip": 1.04712272,
      "balance_loss_mlp": 1.01527786,
      "epoch": 0.7470690795406721,
      "flos": 16690849407360.0,
      "grad_norm": 2.3397020194551086,
      "language_loss": 0.77213418,
      "learning_rate": 6.342255469986053e-07,
      "loss": 0.7950837,
      "num_input_tokens_seen": 133551300,
      "step": 6213,
      "time_per_iteration": 2.6553595066070557
    },
    {
      "auxiliary_loss_clip": 0.01173144,
      "auxiliary_loss_mlp": 0.0101773,
      "balance_loss_clip": 1.04971766,
      "balance_loss_mlp": 1.01112878,
      "epoch": 0.7471893224313112,
      "flos": 25192700352000.0,
      "grad_norm": 2.1720141077511697,
      "language_loss": 0.76385111,
      "learning_rate": 6.336565932713533e-07,
      "loss": 0.78575993,
      "num_input_tokens_seen": 133570725,
      "step": 6214,
      "time_per_iteration": 2.613816499710083
    },
    {
      "auxiliary_loss_clip": 0.01277915,
      "auxiliary_loss_mlp": 0.01027089,
      "balance_loss_clip": 1.05145383,
      "balance_loss_mlp": 1.01995778,
      "epoch": 0.7473095653219504,
      "flos": 22526225199360.0,
      "grad_norm": 2.290258192399156,
      "language_loss": 0.779181,
      "learning_rate": 6.330878468173088e-07,
      "loss": 0.80223107,
      "num_input_tokens_seen": 133590790,
      "step": 6215,
      "time_per_iteration": 2.736750364303589
    },
    {
      "auxiliary_loss_clip": 0.01219737,
      "auxiliary_loss_mlp": 0.0102382,
      "balance_loss_clip": 1.04544711,
      "balance_loss_mlp": 1.01721263,
      "epoch": 0.7474298082125894,
      "flos": 18113989236480.0,
      "grad_norm": 1.6761854892236625,
      "language_loss": 0.72697723,
      "learning_rate": 6.32519307722752e-07,
      "loss": 0.74941278,
      "num_input_tokens_seen": 133608685,
      "step": 6216,
      "time_per_iteration": 2.5814549922943115
    },
    {
      "auxiliary_loss_clip": 0.01235172,
      "auxiliary_loss_mlp": 0.00998633,
      "balance_loss_clip": 1.02028131,
      "balance_loss_mlp": 0.99757808,
      "epoch": 0.7475500511032285,
      "flos": 62086535193600.0,
      "grad_norm": 0.8358916795645344,
      "language_loss": 0.5498358,
      "learning_rate": 6.31950976073929e-07,
      "loss": 0.57217383,
      "num_input_tokens_seen": 133662775,
      "step": 6217,
      "time_per_iteration": 5.030264377593994
    },
    {
      "auxiliary_loss_clip": 0.01374665,
      "auxiliary_loss_mlp": 0.01025132,
      "balance_loss_clip": 1.04694247,
      "balance_loss_mlp": 1.01833427,
      "epoch": 0.7476702939938676,
      "flos": 17785586165760.0,
      "grad_norm": 1.9598036488623054,
      "language_loss": 0.80569386,
      "learning_rate": 6.31382851957055e-07,
      "loss": 0.82969183,
      "num_input_tokens_seen": 133679595,
      "step": 6218,
      "time_per_iteration": 2.736459970474243
    },
    {
      "auxiliary_loss_clip": 0.01321318,
      "auxiliary_loss_mlp": 0.02563228,
      "balance_loss_clip": 1.0465548,
      "balance_loss_mlp": 0.9999243,
      "epoch": 0.7477905368845067,
      "flos": 27927944092800.0,
      "grad_norm": 3.085368888046895,
      "language_loss": 0.71805328,
      "learning_rate": 6.308149354583143e-07,
      "loss": 0.75689876,
      "num_input_tokens_seen": 133699000,
      "step": 6219,
      "time_per_iteration": 3.8512353897094727
    },
    {
      "auxiliary_loss_clip": 0.01230993,
      "auxiliary_loss_mlp": 0.01030365,
      "balance_loss_clip": 1.05086279,
      "balance_loss_mlp": 1.02306962,
      "epoch": 0.7479107797751458,
      "flos": 26870374932480.0,
      "grad_norm": 1.9029299382187905,
      "language_loss": 0.8191613,
      "learning_rate": 6.302472266638586e-07,
      "loss": 0.84177494,
      "num_input_tokens_seen": 133719540,
      "step": 6220,
      "time_per_iteration": 2.6258554458618164
    },
    {
      "auxiliary_loss_clip": 0.01181791,
      "auxiliary_loss_mlp": 0.01024153,
      "balance_loss_clip": 1.05221343,
      "balance_loss_mlp": 1.01670218,
      "epoch": 0.7480310226657849,
      "flos": 33943375785600.0,
      "grad_norm": 2.144105035960463,
      "language_loss": 0.70089531,
      "learning_rate": 6.296797256598101e-07,
      "loss": 0.72295475,
      "num_input_tokens_seen": 133741020,
      "step": 6221,
      "time_per_iteration": 2.6991336345672607
    },
    {
      "auxiliary_loss_clip": 0.01318181,
      "auxiliary_loss_mlp": 0.01019232,
      "balance_loss_clip": 1.04513574,
      "balance_loss_mlp": 1.01267266,
      "epoch": 0.748151265556424,
      "flos": 24826555065600.0,
      "grad_norm": 1.8202269419668702,
      "language_loss": 0.81000996,
      "learning_rate": 6.291124325322576e-07,
      "loss": 0.8333841,
      "num_input_tokens_seen": 133761145,
      "step": 6222,
      "time_per_iteration": 2.705414056777954
    },
    {
      "auxiliary_loss_clip": 0.0127938,
      "auxiliary_loss_mlp": 0.01022674,
      "balance_loss_clip": 1.04844022,
      "balance_loss_mlp": 1.01553345,
      "epoch": 0.748271508447063,
      "flos": 38399351535360.0,
      "grad_norm": 1.668046434183786,
      "language_loss": 0.62507302,
      "learning_rate": 6.285453473672595e-07,
      "loss": 0.64809346,
      "num_input_tokens_seen": 133783715,
      "step": 6223,
      "time_per_iteration": 2.8758492469787598
    },
    {
      "auxiliary_loss_clip": 0.01174277,
      "auxiliary_loss_mlp": 0.01024879,
      "balance_loss_clip": 1.05016756,
      "balance_loss_mlp": 1.01866841,
      "epoch": 0.7483917513377022,
      "flos": 21541842000000.0,
      "grad_norm": 2.921265657843424,
      "language_loss": 0.75667942,
      "learning_rate": 6.279784702508415e-07,
      "loss": 0.77867103,
      "num_input_tokens_seen": 133804465,
      "step": 6224,
      "time_per_iteration": 2.575937509536743
    },
    {
      "auxiliary_loss_clip": 0.01130837,
      "auxiliary_loss_mlp": 0.00999833,
      "balance_loss_clip": 1.01036882,
      "balance_loss_mlp": 0.99889755,
      "epoch": 0.7485119942283412,
      "flos": 62314532772480.0,
      "grad_norm": 0.7787333355883057,
      "language_loss": 0.58578098,
      "learning_rate": 6.274118012689979e-07,
      "loss": 0.60708761,
      "num_input_tokens_seen": 133866365,
      "step": 6225,
      "time_per_iteration": 3.37434720993042
    },
    {
      "auxiliary_loss_clip": 0.01268728,
      "auxiliary_loss_mlp": 0.01026449,
      "balance_loss_clip": 1.04405403,
      "balance_loss_mlp": 1.01940656,
      "epoch": 0.7486322371189803,
      "flos": 29937613104000.0,
      "grad_norm": 1.5689219261703458,
      "language_loss": 0.68369991,
      "learning_rate": 6.268453405076943e-07,
      "loss": 0.70665169,
      "num_input_tokens_seen": 133888760,
      "step": 6226,
      "time_per_iteration": 3.6518826484680176
    },
    {
      "auxiliary_loss_clip": 0.01273185,
      "auxiliary_loss_mlp": 0.01028611,
      "balance_loss_clip": 1.04598916,
      "balance_loss_mlp": 1.02175951,
      "epoch": 0.7487524800096195,
      "flos": 18949414734720.0,
      "grad_norm": 2.9302514136478424,
      "language_loss": 0.8218137,
      "learning_rate": 6.262790880528592e-07,
      "loss": 0.84483171,
      "num_input_tokens_seen": 133906380,
      "step": 6227,
      "time_per_iteration": 2.647641658782959
    },
    {
      "auxiliary_loss_clip": 0.01334067,
      "auxiliary_loss_mlp": 0.01029813,
      "balance_loss_clip": 1.04370785,
      "balance_loss_mlp": 1.0224489,
      "epoch": 0.7488727229002585,
      "flos": 18697393935360.0,
      "grad_norm": 2.8017058103746266,
      "language_loss": 0.79848087,
      "learning_rate": 6.257130439903951e-07,
      "loss": 0.82211965,
      "num_input_tokens_seen": 133922875,
      "step": 6228,
      "time_per_iteration": 2.692953109741211
    },
    {
      "auxiliary_loss_clip": 0.01178465,
      "auxiliary_loss_mlp": 0.01030738,
      "balance_loss_clip": 1.05205083,
      "balance_loss_mlp": 1.02367747,
      "epoch": 0.7489929657908976,
      "flos": 23623368168960.0,
      "grad_norm": 1.939514489147152,
      "language_loss": 0.80967134,
      "learning_rate": 6.251472084061695e-07,
      "loss": 0.83176339,
      "num_input_tokens_seen": 133941795,
      "step": 6229,
      "time_per_iteration": 2.6251463890075684
    },
    {
      "auxiliary_loss_clip": 0.01221813,
      "auxiliary_loss_mlp": 0.01027573,
      "balance_loss_clip": 1.04842687,
      "balance_loss_mlp": 1.02093065,
      "epoch": 0.7491132086815367,
      "flos": 20551533056640.0,
      "grad_norm": 3.5311608987528147,
      "language_loss": 0.88843358,
      "learning_rate": 6.245815813860191e-07,
      "loss": 0.91092736,
      "num_input_tokens_seen": 133957305,
      "step": 6230,
      "time_per_iteration": 2.6296768188476562
    },
    {
      "auxiliary_loss_clip": 0.01178394,
      "auxiliary_loss_mlp": 0.0102967,
      "balance_loss_clip": 1.05039096,
      "balance_loss_mlp": 1.02219892,
      "epoch": 0.7492334515721758,
      "flos": 23003011353600.0,
      "grad_norm": 2.038011386610414,
      "language_loss": 0.70361817,
      "learning_rate": 6.240161630157495e-07,
      "loss": 0.72569883,
      "num_input_tokens_seen": 133976660,
      "step": 6231,
      "time_per_iteration": 2.573375940322876
    },
    {
      "auxiliary_loss_clip": 0.01179251,
      "auxiliary_loss_mlp": 0.01025243,
      "balance_loss_clip": 1.05209529,
      "balance_loss_mlp": 1.01823938,
      "epoch": 0.7493536944628149,
      "flos": 16398823835520.0,
      "grad_norm": 3.051608268786277,
      "language_loss": 0.70308137,
      "learning_rate": 6.23450953381133e-07,
      "loss": 0.72512627,
      "num_input_tokens_seen": 133994750,
      "step": 6232,
      "time_per_iteration": 2.606398582458496
    },
    {
      "auxiliary_loss_clip": 0.01271659,
      "auxiliary_loss_mlp": 0.01023047,
      "balance_loss_clip": 1.04646218,
      "balance_loss_mlp": 1.01669645,
      "epoch": 0.749473937353454,
      "flos": 15338561155200.0,
      "grad_norm": 1.8318306828413755,
      "language_loss": 0.67886597,
      "learning_rate": 6.228859525679131e-07,
      "loss": 0.70181304,
      "num_input_tokens_seen": 134009165,
      "step": 6233,
      "time_per_iteration": 2.583494186401367
    },
    {
      "auxiliary_loss_clip": 0.01224338,
      "auxiliary_loss_mlp": 0.01023102,
      "balance_loss_clip": 1.04972148,
      "balance_loss_mlp": 1.01627707,
      "epoch": 0.7495941802440931,
      "flos": 18951138587520.0,
      "grad_norm": 2.537618473656424,
      "language_loss": 0.79484689,
      "learning_rate": 6.223211606617986e-07,
      "loss": 0.81732124,
      "num_input_tokens_seen": 134027585,
      "step": 6234,
      "time_per_iteration": 2.588749885559082
    },
    {
      "auxiliary_loss_clip": 0.01222354,
      "auxiliary_loss_mlp": 0.01028197,
      "balance_loss_clip": 1.05190086,
      "balance_loss_mlp": 1.02248955,
      "epoch": 0.7497144231347321,
      "flos": 22492469393280.0,
      "grad_norm": 1.776608233216633,
      "language_loss": 0.8395915,
      "learning_rate": 6.217565777484701e-07,
      "loss": 0.86209702,
      "num_input_tokens_seen": 134046680,
      "step": 6235,
      "time_per_iteration": 2.612347364425659
    },
    {
      "auxiliary_loss_clip": 0.01270485,
      "auxiliary_loss_mlp": 0.02565191,
      "balance_loss_clip": 1.04759645,
      "balance_loss_mlp": 0.99991363,
      "epoch": 0.7498346660253713,
      "flos": 24243509502720.0,
      "grad_norm": 2.027173292677425,
      "language_loss": 0.80173403,
      "learning_rate": 6.211922039135722e-07,
      "loss": 0.84009075,
      "num_input_tokens_seen": 134066825,
      "step": 6236,
      "time_per_iteration": 2.682100534439087
    },
    {
      "auxiliary_loss_clip": 0.0117618,
      "auxiliary_loss_mlp": 0.01022571,
      "balance_loss_clip": 1.05200696,
      "balance_loss_mlp": 1.01615167,
      "epoch": 0.7499549089160104,
      "flos": 24387080163840.0,
      "grad_norm": 2.9193839104557964,
      "language_loss": 0.81114149,
      "learning_rate": 6.206280392427201e-07,
      "loss": 0.83312899,
      "num_input_tokens_seen": 134086410,
      "step": 6237,
      "time_per_iteration": 2.5994954109191895
    },
    {
      "auxiliary_loss_clip": 0.01217989,
      "auxiliary_loss_mlp": 0.01028827,
      "balance_loss_clip": 1.04558873,
      "balance_loss_mlp": 1.02164483,
      "epoch": 0.7500751518066494,
      "flos": 34057320704640.0,
      "grad_norm": 1.8135812013524546,
      "language_loss": 0.73869145,
      "learning_rate": 6.200640838214983e-07,
      "loss": 0.76115966,
      "num_input_tokens_seen": 134109185,
      "step": 6238,
      "time_per_iteration": 2.7624800205230713
    },
    {
      "auxiliary_loss_clip": 0.01174668,
      "auxiliary_loss_mlp": 0.01022057,
      "balance_loss_clip": 1.0504818,
      "balance_loss_mlp": 1.01586163,
      "epoch": 0.7501953946972886,
      "flos": 18843586289280.0,
      "grad_norm": 1.7825597858636124,
      "language_loss": 0.6684745,
      "learning_rate": 6.195003377354578e-07,
      "loss": 0.69044173,
      "num_input_tokens_seen": 134128455,
      "step": 6239,
      "time_per_iteration": 2.5867245197296143
    },
    {
      "auxiliary_loss_clip": 0.01223774,
      "auxiliary_loss_mlp": 0.01028707,
      "balance_loss_clip": 1.04585421,
      "balance_loss_mlp": 1.02184677,
      "epoch": 0.7503156375879276,
      "flos": 20257675891200.0,
      "grad_norm": 2.7195984159482283,
      "language_loss": 0.73467451,
      "learning_rate": 6.189368010701183e-07,
      "loss": 0.75719929,
      "num_input_tokens_seen": 134145515,
      "step": 6240,
      "time_per_iteration": 2.5812289714813232
    },
    {
      "auxiliary_loss_clip": 0.01228966,
      "auxiliary_loss_mlp": 0.01023705,
      "balance_loss_clip": 1.0470438,
      "balance_loss_mlp": 1.01669526,
      "epoch": 0.7504358804785667,
      "flos": 13480040574720.0,
      "grad_norm": 1.8197526702011737,
      "language_loss": 0.7643103,
      "learning_rate": 6.183734739109683e-07,
      "loss": 0.78683704,
      "num_input_tokens_seen": 134163335,
      "step": 6241,
      "time_per_iteration": 2.624143600463867
    },
    {
      "auxiliary_loss_clip": 0.01134333,
      "auxiliary_loss_mlp": 0.01027118,
      "balance_loss_clip": 1.0527761,
      "balance_loss_mlp": 1.01990557,
      "epoch": 0.7505561233692057,
      "flos": 29461042431360.0,
      "grad_norm": 3.2100509338241063,
      "language_loss": 0.69131732,
      "learning_rate": 6.178103563434629e-07,
      "loss": 0.71293181,
      "num_input_tokens_seen": 134182335,
      "step": 6242,
      "time_per_iteration": 2.6519241333007812
    },
    {
      "auxiliary_loss_clip": 0.01176883,
      "auxiliary_loss_mlp": 0.0102866,
      "balance_loss_clip": 1.05202699,
      "balance_loss_mlp": 1.02196944,
      "epoch": 0.7506763662598449,
      "flos": 20302457172480.0,
      "grad_norm": 1.9156463045510121,
      "language_loss": 0.841717,
      "learning_rate": 6.172474484530283e-07,
      "loss": 0.86377245,
      "num_input_tokens_seen": 134201070,
      "step": 6243,
      "time_per_iteration": 3.567047357559204
    },
    {
      "auxiliary_loss_clip": 0.01268137,
      "auxiliary_loss_mlp": 0.01025152,
      "balance_loss_clip": 1.04271722,
      "balance_loss_mlp": 1.01824677,
      "epoch": 0.750796609150484,
      "flos": 37230961939200.0,
      "grad_norm": 2.521099386594235,
      "language_loss": 0.76004755,
      "learning_rate": 6.166847503250563e-07,
      "loss": 0.78298044,
      "num_input_tokens_seen": 134223310,
      "step": 6244,
      "time_per_iteration": 2.7720963954925537
    },
    {
      "auxiliary_loss_clip": 0.01270525,
      "auxiliary_loss_mlp": 0.01024116,
      "balance_loss_clip": 1.04617739,
      "balance_loss_mlp": 1.01702905,
      "epoch": 0.750916852041123,
      "flos": 19609417186560.0,
      "grad_norm": 3.3943868643665773,
      "language_loss": 0.78649306,
      "learning_rate": 6.161222620449078e-07,
      "loss": 0.80943954,
      "num_input_tokens_seen": 134242085,
      "step": 6245,
      "time_per_iteration": 3.6157946586608887
    },
    {
      "auxiliary_loss_clip": 0.01322665,
      "auxiliary_loss_mlp": 0.01026297,
      "balance_loss_clip": 1.04667711,
      "balance_loss_mlp": 1.01914132,
      "epoch": 0.7510370949317622,
      "flos": 25112690807040.0,
      "grad_norm": 2.5406326941022233,
      "language_loss": 0.80043435,
      "learning_rate": 6.155599836979117e-07,
      "loss": 0.82392395,
      "num_input_tokens_seen": 134260770,
      "step": 6246,
      "time_per_iteration": 2.7174274921417236
    },
    {
      "auxiliary_loss_clip": 0.01370364,
      "auxiliary_loss_mlp": 0.0102392,
      "balance_loss_clip": 1.04249167,
      "balance_loss_mlp": 1.0166508,
      "epoch": 0.7511573378224012,
      "flos": 19062282245760.0,
      "grad_norm": 3.084013146784845,
      "language_loss": 0.8181107,
      "learning_rate": 6.149979153693649e-07,
      "loss": 0.84205353,
      "num_input_tokens_seen": 134278025,
      "step": 6247,
      "time_per_iteration": 2.719174385070801
    },
    {
      "auxiliary_loss_clip": 0.01226308,
      "auxiliary_loss_mlp": 0.01031818,
      "balance_loss_clip": 1.04770601,
      "balance_loss_mlp": 1.02451611,
      "epoch": 0.7512775807130403,
      "flos": 19937676602880.0,
      "grad_norm": 1.9938376380588003,
      "language_loss": 0.7703532,
      "learning_rate": 6.144360571445343e-07,
      "loss": 0.79293442,
      "num_input_tokens_seen": 134297170,
      "step": 6248,
      "time_per_iteration": 2.6908886432647705
    },
    {
      "auxiliary_loss_clip": 0.01223247,
      "auxiliary_loss_mlp": 0.01023889,
      "balance_loss_clip": 1.04994309,
      "balance_loss_mlp": 1.01701367,
      "epoch": 0.7513978236036795,
      "flos": 20739920912640.0,
      "grad_norm": 1.9310994199712697,
      "language_loss": 0.80133092,
      "learning_rate": 6.138744091086509e-07,
      "loss": 0.82380223,
      "num_input_tokens_seen": 134316755,
      "step": 6249,
      "time_per_iteration": 2.603250741958618
    },
    {
      "auxiliary_loss_clip": 0.01322344,
      "auxiliary_loss_mlp": 0.01022311,
      "balance_loss_clip": 1.04664874,
      "balance_loss_mlp": 1.01587117,
      "epoch": 0.7515180664943185,
      "flos": 27563163523200.0,
      "grad_norm": 3.8479032605367802,
      "language_loss": 0.72406077,
      "learning_rate": 6.133129713469183e-07,
      "loss": 0.74750733,
      "num_input_tokens_seen": 134335960,
      "step": 6250,
      "time_per_iteration": 2.755819082260132
    },
    {
      "auxiliary_loss_clip": 0.01332158,
      "auxiliary_loss_mlp": 0.01028016,
      "balance_loss_clip": 1.04423451,
      "balance_loss_mlp": 1.02100682,
      "epoch": 0.7516383093849576,
      "flos": 33803181002880.0,
      "grad_norm": 1.8741366260917156,
      "language_loss": 0.63984865,
      "learning_rate": 6.127517439445053e-07,
      "loss": 0.66345042,
      "num_input_tokens_seen": 134356805,
      "step": 6251,
      "time_per_iteration": 2.759401798248291
    },
    {
      "auxiliary_loss_clip": 0.0126955,
      "auxiliary_loss_mlp": 0.01027071,
      "balance_loss_clip": 1.04549956,
      "balance_loss_mlp": 1.0212028,
      "epoch": 0.7517585522755967,
      "flos": 29746172592000.0,
      "grad_norm": 2.802083561051667,
      "language_loss": 0.81585163,
      "learning_rate": 6.121907269865498e-07,
      "loss": 0.83881783,
      "num_input_tokens_seen": 134376295,
      "step": 6252,
      "time_per_iteration": 3.6661221981048584
    },
    {
      "auxiliary_loss_clip": 0.01221089,
      "auxiliary_loss_mlp": 0.01001475,
      "balance_loss_clip": 1.01131225,
      "balance_loss_mlp": 1.00041437,
      "epoch": 0.7518787951662358,
      "flos": 69807974319360.0,
      "grad_norm": 0.9328799528121419,
      "language_loss": 0.67185277,
      "learning_rate": 6.116299205581577e-07,
      "loss": 0.69407839,
      "num_input_tokens_seen": 134431125,
      "step": 6253,
      "time_per_iteration": 3.2214555740356445
    },
    {
      "auxiliary_loss_clip": 0.01179544,
      "auxiliary_loss_mlp": 0.01028893,
      "balance_loss_clip": 1.05040479,
      "balance_loss_mlp": 1.021222,
      "epoch": 0.7519990380568748,
      "flos": 34203225749760.0,
      "grad_norm": 1.8298276358854912,
      "language_loss": 0.68432462,
      "learning_rate": 6.110693247444018e-07,
      "loss": 0.70640898,
      "num_input_tokens_seen": 134452960,
      "step": 6254,
      "time_per_iteration": 2.672105550765991
    },
    {
      "auxiliary_loss_clip": 0.01314855,
      "auxiliary_loss_mlp": 0.01025157,
      "balance_loss_clip": 1.0438205,
      "balance_loss_mlp": 1.01845491,
      "epoch": 0.752119280947514,
      "flos": 21725704742400.0,
      "grad_norm": 2.1678966264859705,
      "language_loss": 0.82797599,
      "learning_rate": 6.105089396303258e-07,
      "loss": 0.85137606,
      "num_input_tokens_seen": 134471350,
      "step": 6255,
      "time_per_iteration": 2.705355167388916
    },
    {
      "auxiliary_loss_clip": 0.01275617,
      "auxiliary_loss_mlp": 0.01027226,
      "balance_loss_clip": 1.04781389,
      "balance_loss_mlp": 1.02011561,
      "epoch": 0.7522395238381531,
      "flos": 32742774668160.0,
      "grad_norm": 2.0891232547884395,
      "language_loss": 0.76118004,
      "learning_rate": 6.099487653009383e-07,
      "loss": 0.78420854,
      "num_input_tokens_seen": 134490695,
      "step": 6256,
      "time_per_iteration": 2.726875066757202
    },
    {
      "auxiliary_loss_clip": 0.01224304,
      "auxiliary_loss_mlp": 0.01022846,
      "balance_loss_clip": 1.04951191,
      "balance_loss_mlp": 1.01701093,
      "epoch": 0.7523597667287921,
      "flos": 23476026579840.0,
      "grad_norm": 1.9855627546290837,
      "language_loss": 0.83022672,
      "learning_rate": 6.093888018412192e-07,
      "loss": 0.85269821,
      "num_input_tokens_seen": 134506885,
      "step": 6257,
      "time_per_iteration": 2.64103364944458
    },
    {
      "auxiliary_loss_clip": 0.01123053,
      "auxiliary_loss_mlp": 0.01001162,
      "balance_loss_clip": 1.01119924,
      "balance_loss_mlp": 1.00024974,
      "epoch": 0.7524800096194313,
      "flos": 67346730501120.0,
      "grad_norm": 0.7008956402672702,
      "language_loss": 0.5461086,
      "learning_rate": 6.088290493361125e-07,
      "loss": 0.5673508,
      "num_input_tokens_seen": 134571770,
      "step": 6258,
      "time_per_iteration": 3.3366308212280273
    },
    {
      "auxiliary_loss_clip": 0.01373236,
      "auxiliary_loss_mlp": 0.01028967,
      "balance_loss_clip": 1.04501438,
      "balance_loss_mlp": 1.02143943,
      "epoch": 0.7526002525100703,
      "flos": 13006055681280.0,
      "grad_norm": 1.9921749948058005,
      "language_loss": 0.71548057,
      "learning_rate": 6.082695078705322e-07,
      "loss": 0.73950255,
      "num_input_tokens_seen": 134589250,
      "step": 6259,
      "time_per_iteration": 2.682206392288208
    },
    {
      "auxiliary_loss_clip": 0.01223934,
      "auxiliary_loss_mlp": 0.01024744,
      "balance_loss_clip": 1.04981256,
      "balance_loss_mlp": 1.01772869,
      "epoch": 0.7527204954007094,
      "flos": 21397229844480.0,
      "grad_norm": 4.574554554113987,
      "language_loss": 0.69066495,
      "learning_rate": 6.077101775293618e-07,
      "loss": 0.71315169,
      "num_input_tokens_seen": 134608075,
      "step": 6260,
      "time_per_iteration": 2.600038528442383
    },
    {
      "auxiliary_loss_clip": 0.01223394,
      "auxiliary_loss_mlp": 0.01025654,
      "balance_loss_clip": 1.04752994,
      "balance_loss_mlp": 1.01834893,
      "epoch": 0.7528407382913486,
      "flos": 18947188091520.0,
      "grad_norm": 3.8497875426713195,
      "language_loss": 0.82372731,
      "learning_rate": 6.071510583974504e-07,
      "loss": 0.84621775,
      "num_input_tokens_seen": 134623260,
      "step": 6261,
      "time_per_iteration": 2.5657777786254883
    },
    {
      "auxiliary_loss_clip": 0.01177624,
      "auxiliary_loss_mlp": 0.01027698,
      "balance_loss_clip": 1.05196762,
      "balance_loss_mlp": 1.02098024,
      "epoch": 0.7529609811819876,
      "flos": 15231798956160.0,
      "grad_norm": 3.0805514366232933,
      "language_loss": 0.7196995,
      "learning_rate": 6.065921505596161e-07,
      "loss": 0.74175268,
      "num_input_tokens_seen": 134641540,
      "step": 6262,
      "time_per_iteration": 2.5643513202667236
    },
    {
      "auxiliary_loss_clip": 0.01319997,
      "auxiliary_loss_mlp": 0.01021567,
      "balance_loss_clip": 1.04676604,
      "balance_loss_mlp": 1.01474476,
      "epoch": 0.7530812240726267,
      "flos": 19354487385600.0,
      "grad_norm": 6.289571373869322,
      "language_loss": 0.77271402,
      "learning_rate": 6.060334541006445e-07,
      "loss": 0.7961297,
      "num_input_tokens_seen": 134660035,
      "step": 6263,
      "time_per_iteration": 2.649203300476074
    },
    {
      "auxiliary_loss_clip": 0.0132539,
      "auxiliary_loss_mlp": 0.01026579,
      "balance_loss_clip": 1.04169118,
      "balance_loss_mlp": 1.01961696,
      "epoch": 0.7532014669632658,
      "flos": 27748247328000.0,
      "grad_norm": 1.8405067803854804,
      "language_loss": 0.69333887,
      "learning_rate": 6.05474969105289e-07,
      "loss": 0.71685857,
      "num_input_tokens_seen": 134683025,
      "step": 6264,
      "time_per_iteration": 2.781316041946411
    },
    {
      "auxiliary_loss_clip": 0.01223417,
      "auxiliary_loss_mlp": 0.01024279,
      "balance_loss_clip": 1.04838991,
      "balance_loss_mlp": 1.01684952,
      "epoch": 0.7533217098539049,
      "flos": 14137421333760.0,
      "grad_norm": 2.949498321056778,
      "language_loss": 0.73749995,
      "learning_rate": 6.049166956582725e-07,
      "loss": 0.75997692,
      "num_input_tokens_seen": 134701290,
      "step": 6265,
      "time_per_iteration": 2.554251194000244
    },
    {
      "auxiliary_loss_clip": 0.01222111,
      "auxiliary_loss_mlp": 0.01021944,
      "balance_loss_clip": 1.04775143,
      "balance_loss_mlp": 1.01556373,
      "epoch": 0.753441952744544,
      "flos": 26429068437120.0,
      "grad_norm": 1.9957709299986253,
      "language_loss": 0.87481868,
      "learning_rate": 6.043586338442841e-07,
      "loss": 0.89725924,
      "num_input_tokens_seen": 134720345,
      "step": 6266,
      "time_per_iteration": 2.6925723552703857
    },
    {
      "auxiliary_loss_clip": 0.01169788,
      "auxiliary_loss_mlp": 0.01021659,
      "balance_loss_clip": 1.04854262,
      "balance_loss_mlp": 1.01507592,
      "epoch": 0.7535621956351831,
      "flos": 23878621192320.0,
      "grad_norm": 1.4306156614819727,
      "language_loss": 0.73073763,
      "learning_rate": 6.038007837479815e-07,
      "loss": 0.75265205,
      "num_input_tokens_seen": 134741450,
      "step": 6267,
      "time_per_iteration": 2.609532117843628
    },
    {
      "auxiliary_loss_clip": 0.0122165,
      "auxiliary_loss_mlp": 0.01023116,
      "balance_loss_clip": 1.0486536,
      "balance_loss_mlp": 1.01639581,
      "epoch": 0.7536824385258222,
      "flos": 21795873960960.0,
      "grad_norm": 2.0300690379358004,
      "language_loss": 0.64324123,
      "learning_rate": 6.032431454539897e-07,
      "loss": 0.66568887,
      "num_input_tokens_seen": 134760295,
      "step": 6268,
      "time_per_iteration": 2.631594657897949
    },
    {
      "auxiliary_loss_clip": 0.013198,
      "auxiliary_loss_mlp": 0.01025593,
      "balance_loss_clip": 1.04515493,
      "balance_loss_mlp": 1.01883435,
      "epoch": 0.7538026814164612,
      "flos": 28911644933760.0,
      "grad_norm": 1.8814573378597863,
      "language_loss": 0.8152746,
      "learning_rate": 6.026857190469014e-07,
      "loss": 0.83872855,
      "num_input_tokens_seen": 134782050,
      "step": 6269,
      "time_per_iteration": 4.791935205459595
    },
    {
      "auxiliary_loss_clip": 0.01273827,
      "auxiliary_loss_mlp": 0.01023651,
      "balance_loss_clip": 1.04696703,
      "balance_loss_mlp": 1.01668024,
      "epoch": 0.7539229243071004,
      "flos": 21104701482240.0,
      "grad_norm": 1.8238713453225737,
      "language_loss": 0.73882818,
      "learning_rate": 6.0212850461128e-07,
      "loss": 0.76180297,
      "num_input_tokens_seen": 134801170,
      "step": 6270,
      "time_per_iteration": 2.66326642036438
    },
    {
      "auxiliary_loss_clip": 0.01271893,
      "auxiliary_loss_mlp": 0.0102441,
      "balance_loss_clip": 1.04417872,
      "balance_loss_mlp": 1.01721573,
      "epoch": 0.7540431671977395,
      "flos": 15158469340800.0,
      "grad_norm": 2.653528689598395,
      "language_loss": 0.74502492,
      "learning_rate": 6.015715022316516e-07,
      "loss": 0.76798797,
      "num_input_tokens_seen": 134819150,
      "step": 6271,
      "time_per_iteration": 3.689634323120117
    },
    {
      "auxiliary_loss_clip": 0.01376685,
      "auxiliary_loss_mlp": 0.01028293,
      "balance_loss_clip": 1.04177952,
      "balance_loss_mlp": 1.02105963,
      "epoch": 0.7541634100883785,
      "flos": 18770579896320.0,
      "grad_norm": 4.369677647015697,
      "language_loss": 0.77833241,
      "learning_rate": 6.010147119925154e-07,
      "loss": 0.80238223,
      "num_input_tokens_seen": 134836905,
      "step": 6272,
      "time_per_iteration": 2.7420337200164795
    },
    {
      "auxiliary_loss_clip": 0.01314751,
      "auxiliary_loss_mlp": 0.01021336,
      "balance_loss_clip": 1.04379892,
      "balance_loss_mlp": 1.01445198,
      "epoch": 0.7542836529790176,
      "flos": 20594770053120.0,
      "grad_norm": 2.6684610039873693,
      "language_loss": 0.6669631,
      "learning_rate": 6.004581339783348e-07,
      "loss": 0.69032395,
      "num_input_tokens_seen": 134855225,
      "step": 6273,
      "time_per_iteration": 2.7293553352355957
    },
    {
      "auxiliary_loss_clip": 0.01229491,
      "auxiliary_loss_mlp": 0.01028046,
      "balance_loss_clip": 1.04997909,
      "balance_loss_mlp": 1.02047026,
      "epoch": 0.7544038958696567,
      "flos": 19095104298240.0,
      "grad_norm": 3.290869433383632,
      "language_loss": 0.68799067,
      "learning_rate": 5.999017682735425e-07,
      "loss": 0.71056604,
      "num_input_tokens_seen": 134871615,
      "step": 6274,
      "time_per_iteration": 2.6805953979492188
    },
    {
      "auxiliary_loss_clip": 0.01428354,
      "auxiliary_loss_mlp": 0.01024257,
      "balance_loss_clip": 1.04302549,
      "balance_loss_mlp": 1.0172534,
      "epoch": 0.7545241387602958,
      "flos": 31723306859520.0,
      "grad_norm": 22.064945291353514,
      "language_loss": 0.66457927,
      "learning_rate": 5.993456149625387e-07,
      "loss": 0.68910539,
      "num_input_tokens_seen": 134892765,
      "step": 6275,
      "time_per_iteration": 2.8822181224823
    },
    {
      "auxiliary_loss_clip": 0.01317606,
      "auxiliary_loss_mlp": 0.01022213,
      "balance_loss_clip": 1.04473674,
      "balance_loss_mlp": 1.01593041,
      "epoch": 0.7546443816509348,
      "flos": 20296495514880.0,
      "grad_norm": 1.8309137063280165,
      "language_loss": 0.8250488,
      "learning_rate": 5.987896741296909e-07,
      "loss": 0.84844697,
      "num_input_tokens_seen": 134910505,
      "step": 6276,
      "time_per_iteration": 2.7634711265563965
    },
    {
      "auxiliary_loss_clip": 0.01273413,
      "auxiliary_loss_mlp": 0.01027492,
      "balance_loss_clip": 1.04876471,
      "balance_loss_mlp": 1.02089381,
      "epoch": 0.754764624541574,
      "flos": 23696159080320.0,
      "grad_norm": 2.0566261733172446,
      "language_loss": 0.77988744,
      "learning_rate": 5.982339458593361e-07,
      "loss": 0.80289644,
      "num_input_tokens_seen": 134930445,
      "step": 6277,
      "time_per_iteration": 2.6642098426818848
    },
    {
      "auxiliary_loss_clip": 0.01220083,
      "auxiliary_loss_mlp": 0.02564001,
      "balance_loss_clip": 1.0493958,
      "balance_loss_mlp": 0.99992168,
      "epoch": 0.7548848674322131,
      "flos": 25337204766720.0,
      "grad_norm": 1.6161488112966702,
      "language_loss": 0.84175545,
      "learning_rate": 5.976784302357767e-07,
      "loss": 0.87959623,
      "num_input_tokens_seen": 134951010,
      "step": 6278,
      "time_per_iteration": 3.5672101974487305
    },
    {
      "auxiliary_loss_clip": 0.01226644,
      "auxiliary_loss_mlp": 0.0102714,
      "balance_loss_clip": 1.04818463,
      "balance_loss_mlp": 1.02029407,
      "epoch": 0.7550051103228521,
      "flos": 19573147428480.0,
      "grad_norm": 3.8817193326729016,
      "language_loss": 0.73654789,
      "learning_rate": 5.971231273432855e-07,
      "loss": 0.75908571,
      "num_input_tokens_seen": 134970495,
      "step": 6279,
      "time_per_iteration": 2.6422367095947266
    },
    {
      "auxiliary_loss_clip": 0.01120451,
      "auxiliary_loss_mlp": 0.01001742,
      "balance_loss_clip": 1.01108479,
      "balance_loss_mlp": 1.00080037,
      "epoch": 0.7551253532134913,
      "flos": 64150068648960.0,
      "grad_norm": 0.813998951886596,
      "language_loss": 0.54529709,
      "learning_rate": 5.965680372661e-07,
      "loss": 0.56651902,
      "num_input_tokens_seen": 135028060,
      "step": 6280,
      "time_per_iteration": 3.1143436431884766
    },
    {
      "auxiliary_loss_clip": 0.01274505,
      "auxiliary_loss_mlp": 0.01017435,
      "balance_loss_clip": 1.04897571,
      "balance_loss_mlp": 1.01130819,
      "epoch": 0.7552455961041303,
      "flos": 26067986968320.0,
      "grad_norm": 1.7482367631342113,
      "language_loss": 0.56402946,
      "learning_rate": 5.960131600884266e-07,
      "loss": 0.58694887,
      "num_input_tokens_seen": 135047330,
      "step": 6281,
      "time_per_iteration": 2.764019012451172
    },
    {
      "auxiliary_loss_clip": 0.01321806,
      "auxiliary_loss_mlp": 0.0102634,
      "balance_loss_clip": 1.04389215,
      "balance_loss_mlp": 1.019521,
      "epoch": 0.7553658389947694,
      "flos": 24498223822080.0,
      "grad_norm": 1.96281666475847,
      "language_loss": 0.76179385,
      "learning_rate": 5.954584958944413e-07,
      "loss": 0.78527528,
      "num_input_tokens_seen": 135065995,
      "step": 6282,
      "time_per_iteration": 2.7051656246185303
    },
    {
      "auxiliary_loss_clip": 0.01323402,
      "auxiliary_loss_mlp": 0.02563886,
      "balance_loss_clip": 1.0432179,
      "balance_loss_mlp": 0.99990463,
      "epoch": 0.7554860818854086,
      "flos": 21799465320960.0,
      "grad_norm": 2.0073899310533156,
      "language_loss": 0.81491971,
      "learning_rate": 5.949040447682854e-07,
      "loss": 0.85379261,
      "num_input_tokens_seen": 135085820,
      "step": 6283,
      "time_per_iteration": 2.692431926727295
    },
    {
      "auxiliary_loss_clip": 0.01278832,
      "auxiliary_loss_mlp": 0.01022138,
      "balance_loss_clip": 1.04695368,
      "balance_loss_mlp": 1.01523912,
      "epoch": 0.7556063247760476,
      "flos": 16362123114240.0,
      "grad_norm": 4.66048957171087,
      "language_loss": 0.68318677,
      "learning_rate": 5.943498067940686e-07,
      "loss": 0.70619643,
      "num_input_tokens_seen": 135102845,
      "step": 6284,
      "time_per_iteration": 2.710841178894043
    },
    {
      "auxiliary_loss_clip": 0.01271801,
      "auxiliary_loss_mlp": 0.01025555,
      "balance_loss_clip": 1.05137372,
      "balance_loss_mlp": 1.01922154,
      "epoch": 0.7557265676666867,
      "flos": 27235155502080.0,
      "grad_norm": 2.0390895924136627,
      "language_loss": 0.81308281,
      "learning_rate": 5.937957820558686e-07,
      "loss": 0.83605635,
      "num_input_tokens_seen": 135122190,
      "step": 6285,
      "time_per_iteration": 2.694814443588257
    },
    {
      "auxiliary_loss_clip": 0.01075194,
      "auxiliary_loss_mlp": 0.0100244,
      "balance_loss_clip": 1.01026452,
      "balance_loss_mlp": 1.00152254,
      "epoch": 0.7558468105573258,
      "flos": 62189131415040.0,
      "grad_norm": 0.8514658805479722,
      "language_loss": 0.65306121,
      "learning_rate": 5.932419706377296e-07,
      "loss": 0.6738376,
      "num_input_tokens_seen": 135180495,
      "step": 6286,
      "time_per_iteration": 3.1912496089935303
    },
    {
      "auxiliary_loss_clip": 0.0131789,
      "auxiliary_loss_mlp": 0.01023935,
      "balance_loss_clip": 1.04715323,
      "balance_loss_mlp": 1.01741719,
      "epoch": 0.7559670534479649,
      "flos": 33249078823680.0,
      "grad_norm": 1.893077587079921,
      "language_loss": 0.74232876,
      "learning_rate": 5.92688372623666e-07,
      "loss": 0.76574695,
      "num_input_tokens_seen": 135199200,
      "step": 6287,
      "time_per_iteration": 2.7779858112335205
    },
    {
      "auxiliary_loss_clip": 0.01221202,
      "auxiliary_loss_mlp": 0.01026001,
      "balance_loss_clip": 1.04601073,
      "balance_loss_mlp": 1.01860392,
      "epoch": 0.7560872963386039,
      "flos": 14064379027200.0,
      "grad_norm": 2.1995240343211635,
      "language_loss": 0.74274671,
      "learning_rate": 5.921349880976574e-07,
      "loss": 0.76521879,
      "num_input_tokens_seen": 135217035,
      "step": 6288,
      "time_per_iteration": 2.580608367919922
    },
    {
      "auxiliary_loss_clip": 0.01277544,
      "auxiliary_loss_mlp": 0.02565571,
      "balance_loss_clip": 1.04594088,
      "balance_loss_mlp": 0.99994129,
      "epoch": 0.7562075392292431,
      "flos": 20412307941120.0,
      "grad_norm": 2.235081666993195,
      "language_loss": 0.82289761,
      "learning_rate": 5.915818171436515e-07,
      "loss": 0.86132872,
      "num_input_tokens_seen": 135236370,
      "step": 6289,
      "time_per_iteration": 2.718810796737671
    },
    {
      "auxiliary_loss_clip": 0.01274077,
      "auxiliary_loss_mlp": 0.01027239,
      "balance_loss_clip": 1.04341531,
      "balance_loss_mlp": 1.01959157,
      "epoch": 0.7563277821198822,
      "flos": 20376792368640.0,
      "grad_norm": 2.917010381467531,
      "language_loss": 0.74279988,
      "learning_rate": 5.910288598455642e-07,
      "loss": 0.76581311,
      "num_input_tokens_seen": 135255720,
      "step": 6290,
      "time_per_iteration": 2.6561546325683594
    },
    {
      "auxiliary_loss_clip": 0.01231,
      "auxiliary_loss_mlp": 0.01024979,
      "balance_loss_clip": 1.04875398,
      "balance_loss_mlp": 1.01757026,
      "epoch": 0.7564480250105212,
      "flos": 18588261438720.0,
      "grad_norm": 2.308221615623373,
      "language_loss": 0.74260187,
      "learning_rate": 5.90476116287278e-07,
      "loss": 0.76516169,
      "num_input_tokens_seen": 135273320,
      "step": 6291,
      "time_per_iteration": 2.55393648147583
    },
    {
      "auxiliary_loss_clip": 0.01271809,
      "auxiliary_loss_mlp": 0.01021194,
      "balance_loss_clip": 1.04820812,
      "balance_loss_mlp": 1.01431274,
      "epoch": 0.7565682679011604,
      "flos": 21215521918080.0,
      "grad_norm": 2.1447554004780645,
      "language_loss": 0.67844522,
      "learning_rate": 5.899235865526456e-07,
      "loss": 0.70137525,
      "num_input_tokens_seen": 135292615,
      "step": 6292,
      "time_per_iteration": 2.6653871536254883
    },
    {
      "auxiliary_loss_clip": 0.01320404,
      "auxiliary_loss_mlp": 0.01029463,
      "balance_loss_clip": 1.04528737,
      "balance_loss_mlp": 1.02301085,
      "epoch": 0.7566885107917994,
      "flos": 20449008662400.0,
      "grad_norm": 1.721989366609964,
      "language_loss": 0.82171917,
      "learning_rate": 5.893712707254825e-07,
      "loss": 0.84521782,
      "num_input_tokens_seen": 135310075,
      "step": 6293,
      "time_per_iteration": 2.684800863265991
    },
    {
      "auxiliary_loss_clip": 0.01366835,
      "auxiliary_loss_mlp": 0.0103097,
      "balance_loss_clip": 1.03994823,
      "balance_loss_mlp": 1.02318883,
      "epoch": 0.7568087536824385,
      "flos": 19025832919680.0,
      "grad_norm": 3.5436703336880444,
      "language_loss": 0.66022241,
      "learning_rate": 5.888191688895769e-07,
      "loss": 0.68420041,
      "num_input_tokens_seen": 135327335,
      "step": 6294,
      "time_per_iteration": 2.782712697982788
    },
    {
      "auxiliary_loss_clip": 0.01175734,
      "auxiliary_loss_mlp": 0.0102284,
      "balance_loss_clip": 1.04834056,
      "balance_loss_mlp": 1.0150615,
      "epoch": 0.7569289965730777,
      "flos": 15225442248960.0,
      "grad_norm": 3.282848727190275,
      "language_loss": 0.6220715,
      "learning_rate": 5.882672811286813e-07,
      "loss": 0.64405727,
      "num_input_tokens_seen": 135343615,
      "step": 6295,
      "time_per_iteration": 3.5912246704101562
    },
    {
      "auxiliary_loss_clip": 0.01175296,
      "auxiliary_loss_mlp": 0.01027953,
      "balance_loss_clip": 1.04903674,
      "balance_loss_mlp": 1.02077389,
      "epoch": 0.7570492394637167,
      "flos": 20769367086720.0,
      "grad_norm": 2.156063962687213,
      "language_loss": 0.68907344,
      "learning_rate": 5.877156075265166e-07,
      "loss": 0.71110588,
      "num_input_tokens_seen": 135359880,
      "step": 6296,
      "time_per_iteration": 2.5461061000823975
    },
    {
      "auxiliary_loss_clip": 0.01274632,
      "auxiliary_loss_mlp": 0.01023233,
      "balance_loss_clip": 1.04341245,
      "balance_loss_mlp": 1.01580048,
      "epoch": 0.7571694823543558,
      "flos": 15664091137920.0,
      "grad_norm": 2.4504963543430796,
      "language_loss": 0.69621205,
      "learning_rate": 5.871641481667715e-07,
      "loss": 0.71919072,
      "num_input_tokens_seen": 135374325,
      "step": 6297,
      "time_per_iteration": 3.5113821029663086
    },
    {
      "auxiliary_loss_clip": 0.01376286,
      "auxiliary_loss_mlp": 0.01026634,
      "balance_loss_clip": 1.04567409,
      "balance_loss_mlp": 1.01938057,
      "epoch": 0.7572897252449949,
      "flos": 25409241492480.0,
      "grad_norm": 2.941750633353927,
      "language_loss": 0.84015191,
      "learning_rate": 5.866129031331011e-07,
      "loss": 0.86418116,
      "num_input_tokens_seen": 135393980,
      "step": 6298,
      "time_per_iteration": 2.736659049987793
    },
    {
      "auxiliary_loss_clip": 0.0127059,
      "auxiliary_loss_mlp": 0.01024419,
      "balance_loss_clip": 1.04409361,
      "balance_loss_mlp": 1.01707578,
      "epoch": 0.757409968135634,
      "flos": 24279348297600.0,
      "grad_norm": 2.350313271549552,
      "language_loss": 0.83365321,
      "learning_rate": 5.8606187250913e-07,
      "loss": 0.85660338,
      "num_input_tokens_seen": 135412030,
      "step": 6299,
      "time_per_iteration": 2.6694650650024414
    },
    {
      "auxiliary_loss_clip": 0.01224555,
      "auxiliary_loss_mlp": 0.02567071,
      "balance_loss_clip": 1.05040801,
      "balance_loss_mlp": 0.99997067,
      "epoch": 0.757530211026273,
      "flos": 24133766474880.0,
      "grad_norm": 2.9636995698056188,
      "language_loss": 0.8420198,
      "learning_rate": 5.855110563784482e-07,
      "loss": 0.8799361,
      "num_input_tokens_seen": 135430565,
      "step": 6300,
      "time_per_iteration": 2.6513469219207764
    },
    {
      "auxiliary_loss_clip": 0.01221381,
      "auxiliary_loss_mlp": 0.02563043,
      "balance_loss_clip": 1.04640102,
      "balance_loss_mlp": 0.99991596,
      "epoch": 0.7576504539169122,
      "flos": 23951807153280.0,
      "grad_norm": 1.9613089379509383,
      "language_loss": 0.64034998,
      "learning_rate": 5.849604548246156e-07,
      "loss": 0.67819422,
      "num_input_tokens_seen": 135451675,
      "step": 6301,
      "time_per_iteration": 2.713392734527588
    },
    {
      "auxiliary_loss_clip": 0.01281171,
      "auxiliary_loss_mlp": 0.02565415,
      "balance_loss_clip": 1.05061126,
      "balance_loss_mlp": 0.99993885,
      "epoch": 0.7577706968075513,
      "flos": 21251360712960.0,
      "grad_norm": 2.1878244644099167,
      "language_loss": 0.80402201,
      "learning_rate": 5.844100679311565e-07,
      "loss": 0.84248787,
      "num_input_tokens_seen": 135470635,
      "step": 6302,
      "time_per_iteration": 2.6313488483428955
    },
    {
      "auxiliary_loss_clip": 0.01270402,
      "auxiliary_loss_mlp": 0.01026308,
      "balance_loss_clip": 1.04786468,
      "balance_loss_mlp": 1.01910794,
      "epoch": 0.7578909396981903,
      "flos": 18296595002880.0,
      "grad_norm": 2.737486751958453,
      "language_loss": 0.76284558,
      "learning_rate": 5.838598957815637e-07,
      "loss": 0.78581274,
      "num_input_tokens_seen": 135487865,
      "step": 6303,
      "time_per_iteration": 2.654484272003174
    },
    {
      "auxiliary_loss_clip": 0.01265717,
      "auxiliary_loss_mlp": 0.01024312,
      "balance_loss_clip": 1.04579544,
      "balance_loss_mlp": 1.01775897,
      "epoch": 0.7580111825888295,
      "flos": 25373869574400.0,
      "grad_norm": 1.5855610806726423,
      "language_loss": 0.85596973,
      "learning_rate": 5.833099384592996e-07,
      "loss": 0.87887001,
      "num_input_tokens_seen": 135508440,
      "step": 6304,
      "time_per_iteration": 3.601959466934204
    },
    {
      "auxiliary_loss_clip": 0.01269132,
      "auxiliary_loss_mlp": 0.0102687,
      "balance_loss_clip": 1.04854047,
      "balance_loss_mlp": 1.01973557,
      "epoch": 0.7581314254794685,
      "flos": 23768662682880.0,
      "grad_norm": 2.3207012740589867,
      "language_loss": 0.71775103,
      "learning_rate": 5.827601960477913e-07,
      "loss": 0.74071097,
      "num_input_tokens_seen": 135526365,
      "step": 6305,
      "time_per_iteration": 2.6713671684265137
    },
    {
      "auxiliary_loss_clip": 0.01223189,
      "auxiliary_loss_mlp": 0.01019657,
      "balance_loss_clip": 1.0470103,
      "balance_loss_mlp": 1.01352072,
      "epoch": 0.7582516683701076,
      "flos": 22054610603520.0,
      "grad_norm": 2.1965720626103886,
      "language_loss": 0.70598632,
      "learning_rate": 5.822106686304344e-07,
      "loss": 0.72841477,
      "num_input_tokens_seen": 135545655,
      "step": 6306,
      "time_per_iteration": 2.6362462043762207
    },
    {
      "auxiliary_loss_clip": 0.01325121,
      "auxiliary_loss_mlp": 0.01025832,
      "balance_loss_clip": 1.04603994,
      "balance_loss_mlp": 1.01888454,
      "epoch": 0.7583719112607467,
      "flos": 31649725848960.0,
      "grad_norm": 1.7572402948020396,
      "language_loss": 0.57980227,
      "learning_rate": 5.816613562905919e-07,
      "loss": 0.60331178,
      "num_input_tokens_seen": 135566840,
      "step": 6307,
      "time_per_iteration": 2.7409708499908447
    },
    {
      "auxiliary_loss_clip": 0.01320691,
      "auxiliary_loss_mlp": 0.01025145,
      "balance_loss_clip": 1.04910409,
      "balance_loss_mlp": 1.0187223,
      "epoch": 0.7584921541513858,
      "flos": 33068376478080.0,
      "grad_norm": 1.8794735660386386,
      "language_loss": 0.69802773,
      "learning_rate": 5.811122591115933e-07,
      "loss": 0.72148609,
      "num_input_tokens_seen": 135587825,
      "step": 6308,
      "time_per_iteration": 2.7690701484680176
    },
    {
      "auxiliary_loss_clip": 0.01328259,
      "auxiliary_loss_mlp": 0.01025397,
      "balance_loss_clip": 1.0527153,
      "balance_loss_mlp": 1.01850712,
      "epoch": 0.7586123970420249,
      "flos": 23326350606720.0,
      "grad_norm": 2.659974698452859,
      "language_loss": 0.71408081,
      "learning_rate": 5.805633771767376e-07,
      "loss": 0.73761737,
      "num_input_tokens_seen": 135605220,
      "step": 6309,
      "time_per_iteration": 2.7120754718780518
    },
    {
      "auxiliary_loss_clip": 0.01176747,
      "auxiliary_loss_mlp": 0.01025633,
      "balance_loss_clip": 1.0493536,
      "balance_loss_mlp": 1.01890969,
      "epoch": 0.7587326399326639,
      "flos": 18334229477760.0,
      "grad_norm": 2.043086645890323,
      "language_loss": 0.77759093,
      "learning_rate": 5.800147105692888e-07,
      "loss": 0.79961473,
      "num_input_tokens_seen": 135624795,
      "step": 6310,
      "time_per_iteration": 2.655167818069458
    },
    {
      "auxiliary_loss_clip": 0.01225822,
      "auxiliary_loss_mlp": 0.01025627,
      "balance_loss_clip": 1.04727018,
      "balance_loss_mlp": 1.01860595,
      "epoch": 0.7588528828233031,
      "flos": 17275080119040.0,
      "grad_norm": 1.7686406187708583,
      "language_loss": 0.79022503,
      "learning_rate": 5.794662593724795e-07,
      "loss": 0.81273949,
      "num_input_tokens_seen": 135643800,
      "step": 6311,
      "time_per_iteration": 2.6210312843322754
    },
    {
      "auxiliary_loss_clip": 0.01175468,
      "auxiliary_loss_mlp": 0.01030169,
      "balance_loss_clip": 1.05062735,
      "balance_loss_mlp": 1.02282858,
      "epoch": 0.7589731257139422,
      "flos": 17713621267200.0,
      "grad_norm": 2.1037384509263823,
      "language_loss": 0.74852085,
      "learning_rate": 5.789180236695091e-07,
      "loss": 0.77057719,
      "num_input_tokens_seen": 135660655,
      "step": 6312,
      "time_per_iteration": 2.5613694190979004
    },
    {
      "auxiliary_loss_clip": 0.01219169,
      "auxiliary_loss_mlp": 0.01022028,
      "balance_loss_clip": 1.04855406,
      "balance_loss_mlp": 1.01559961,
      "epoch": 0.7590933686045812,
      "flos": 15961072786560.0,
      "grad_norm": 1.9089030422057378,
      "language_loss": 0.84846056,
      "learning_rate": 5.78370003543544e-07,
      "loss": 0.8708725,
      "num_input_tokens_seen": 135679410,
      "step": 6313,
      "time_per_iteration": 2.6367571353912354
    },
    {
      "auxiliary_loss_clip": 0.01224438,
      "auxiliary_loss_mlp": 0.02567938,
      "balance_loss_clip": 1.04904437,
      "balance_loss_mlp": 0.99995041,
      "epoch": 0.7592136114952204,
      "flos": 21068072588160.0,
      "grad_norm": 2.3371609562120734,
      "language_loss": 0.83881831,
      "learning_rate": 5.778221990777203e-07,
      "loss": 0.87674212,
      "num_input_tokens_seen": 135697150,
      "step": 6314,
      "time_per_iteration": 2.6249451637268066
    },
    {
      "auxiliary_loss_clip": 0.01268852,
      "auxiliary_loss_mlp": 0.01019934,
      "balance_loss_clip": 1.04680169,
      "balance_loss_mlp": 1.01307023,
      "epoch": 0.7593338543858594,
      "flos": 25297666871040.0,
      "grad_norm": 3.9714871188246548,
      "language_loss": 0.82512867,
      "learning_rate": 5.772746103551372e-07,
      "loss": 0.84801656,
      "num_input_tokens_seen": 135712545,
      "step": 6315,
      "time_per_iteration": 2.7070794105529785
    },
    {
      "auxiliary_loss_clip": 0.01269259,
      "auxiliary_loss_mlp": 0.01024439,
      "balance_loss_clip": 1.04659724,
      "balance_loss_mlp": 1.01782036,
      "epoch": 0.7594540972764985,
      "flos": 31832367528960.0,
      "grad_norm": 3.4868142947686223,
      "language_loss": 0.71883452,
      "learning_rate": 5.767272374588648e-07,
      "loss": 0.74177152,
      "num_input_tokens_seen": 135733950,
      "step": 6316,
      "time_per_iteration": 2.7219178676605225
    },
    {
      "auxiliary_loss_clip": 0.01224059,
      "auxiliary_loss_mlp": 0.0102804,
      "balance_loss_clip": 1.05035067,
      "balance_loss_mlp": 1.02167463,
      "epoch": 0.7595743401671377,
      "flos": 37597250880000.0,
      "grad_norm": 1.7738038317054508,
      "language_loss": 0.78135568,
      "learning_rate": 5.76180080471939e-07,
      "loss": 0.80387664,
      "num_input_tokens_seen": 135757120,
      "step": 6317,
      "time_per_iteration": 2.7574949264526367
    },
    {
      "auxiliary_loss_clip": 0.0117836,
      "auxiliary_loss_mlp": 0.01026387,
      "balance_loss_clip": 1.05115378,
      "balance_loss_mlp": 1.01859963,
      "epoch": 0.7596945830577767,
      "flos": 18287724343680.0,
      "grad_norm": 2.1640104135084526,
      "language_loss": 0.72216511,
      "learning_rate": 5.756331394773631e-07,
      "loss": 0.74421257,
      "num_input_tokens_seen": 135773335,
      "step": 6318,
      "time_per_iteration": 2.5598721504211426
    },
    {
      "auxiliary_loss_clip": 0.01425441,
      "auxiliary_loss_mlp": 0.02567548,
      "balance_loss_clip": 1.04423225,
      "balance_loss_mlp": 0.99990153,
      "epoch": 0.7598148259484158,
      "flos": 22233122219520.0,
      "grad_norm": 1.801745504469567,
      "language_loss": 0.76260245,
      "learning_rate": 5.750864145581071e-07,
      "loss": 0.80253237,
      "num_input_tokens_seen": 135792555,
      "step": 6319,
      "time_per_iteration": 2.7734079360961914
    },
    {
      "auxiliary_loss_clip": 0.0117547,
      "auxiliary_loss_mlp": 0.01024143,
      "balance_loss_clip": 1.05153537,
      "balance_loss_mlp": 1.01761317,
      "epoch": 0.7599350688390549,
      "flos": 27161718145920.0,
      "grad_norm": 5.017659384972666,
      "language_loss": 0.85896051,
      "learning_rate": 5.745399057971085e-07,
      "loss": 0.88095665,
      "num_input_tokens_seen": 135813690,
      "step": 6320,
      "time_per_iteration": 2.5953924655914307
    },
    {
      "auxiliary_loss_clip": 0.01228437,
      "auxiliary_loss_mlp": 0.01030777,
      "balance_loss_clip": 1.0496521,
      "balance_loss_mlp": 1.02360415,
      "epoch": 0.760055311729694,
      "flos": 15560704817280.0,
      "grad_norm": 2.4110012736293074,
      "language_loss": 0.75431013,
      "learning_rate": 5.739936132772738e-07,
      "loss": 0.77690226,
      "num_input_tokens_seen": 135832255,
      "step": 6321,
      "time_per_iteration": 4.4660797119140625
    },
    {
      "auxiliary_loss_clip": 0.01173786,
      "auxiliary_loss_mlp": 0.01027001,
      "balance_loss_clip": 1.04876161,
      "balance_loss_mlp": 1.01979208,
      "epoch": 0.760175554620333,
      "flos": 25155496840320.0,
      "grad_norm": 2.163901674112299,
      "language_loss": 0.7431438,
      "learning_rate": 5.734475370814733e-07,
      "loss": 0.76515168,
      "num_input_tokens_seen": 135851935,
      "step": 6322,
      "time_per_iteration": 2.6504836082458496
    },
    {
      "auxiliary_loss_clip": 0.012271,
      "auxiliary_loss_mlp": 0.0102697,
      "balance_loss_clip": 1.0474087,
      "balance_loss_mlp": 1.01982927,
      "epoch": 0.7602957975109722,
      "flos": 24353791234560.0,
      "grad_norm": 2.2552159738939435,
      "language_loss": 0.78434741,
      "learning_rate": 5.729016772925483e-07,
      "loss": 0.8068881,
      "num_input_tokens_seen": 135873510,
      "step": 6323,
      "time_per_iteration": 3.5863378047943115
    },
    {
      "auxiliary_loss_clip": 0.01373978,
      "auxiliary_loss_mlp": 0.01023478,
      "balance_loss_clip": 1.04815626,
      "balance_loss_mlp": 1.01609874,
      "epoch": 0.7604160404016113,
      "flos": 25192664438400.0,
      "grad_norm": 2.056327465273519,
      "language_loss": 0.70586479,
      "learning_rate": 5.723560339933038e-07,
      "loss": 0.72983938,
      "num_input_tokens_seen": 135893845,
      "step": 6324,
      "time_per_iteration": 2.733764410018921
    },
    {
      "auxiliary_loss_clip": 0.0122517,
      "auxiliary_loss_mlp": 0.02566143,
      "balance_loss_clip": 1.04702926,
      "balance_loss_mlp": 0.99991584,
      "epoch": 0.7605362832922503,
      "flos": 29861841363840.0,
      "grad_norm": 2.4589438924532505,
      "language_loss": 0.65583867,
      "learning_rate": 5.71810607266513e-07,
      "loss": 0.69375181,
      "num_input_tokens_seen": 135912430,
      "step": 6325,
      "time_per_iteration": 2.7091054916381836
    },
    {
      "auxiliary_loss_clip": 0.01227258,
      "auxiliary_loss_mlp": 0.01025903,
      "balance_loss_clip": 1.04927301,
      "balance_loss_mlp": 1.01906085,
      "epoch": 0.7606565261828895,
      "flos": 13917935278080.0,
      "grad_norm": 1.9661942443155997,
      "language_loss": 0.60616797,
      "learning_rate": 5.712653971949184e-07,
      "loss": 0.6286996,
      "num_input_tokens_seen": 135930550,
      "step": 6326,
      "time_per_iteration": 2.6045713424682617
    },
    {
      "auxiliary_loss_clip": 0.01222048,
      "auxiliary_loss_mlp": 0.01031052,
      "balance_loss_clip": 1.04735112,
      "balance_loss_mlp": 1.02384615,
      "epoch": 0.7607767690735285,
      "flos": 18551273408640.0,
      "grad_norm": 2.8599694545261123,
      "language_loss": 0.7518695,
      "learning_rate": 5.707204038612268e-07,
      "loss": 0.77440053,
      "num_input_tokens_seen": 135947980,
      "step": 6327,
      "time_per_iteration": 2.596513271331787
    },
    {
      "auxiliary_loss_clip": 0.01278217,
      "auxiliary_loss_mlp": 0.01027975,
      "balance_loss_clip": 1.05033767,
      "balance_loss_mlp": 1.02024388,
      "epoch": 0.7608970119641676,
      "flos": 20922993555840.0,
      "grad_norm": 2.4130629400980483,
      "language_loss": 0.73829138,
      "learning_rate": 5.701756273481138e-07,
      "loss": 0.76135325,
      "num_input_tokens_seen": 135965400,
      "step": 6328,
      "time_per_iteration": 2.648343801498413
    },
    {
      "auxiliary_loss_clip": 0.01177614,
      "auxiliary_loss_mlp": 0.01027919,
      "balance_loss_clip": 1.04599237,
      "balance_loss_mlp": 1.02138603,
      "epoch": 0.7610172548548068,
      "flos": 23807302738560.0,
      "grad_norm": 1.7607551957522287,
      "language_loss": 0.73910797,
      "learning_rate": 5.696310677382212e-07,
      "loss": 0.76116329,
      "num_input_tokens_seen": 135986795,
      "step": 6329,
      "time_per_iteration": 2.6665496826171875
    },
    {
      "auxiliary_loss_clip": 0.01222953,
      "auxiliary_loss_mlp": 0.01000971,
      "balance_loss_clip": 1.01022935,
      "balance_loss_mlp": 1.00001717,
      "epoch": 0.7611374977454458,
      "flos": 66496580426880.0,
      "grad_norm": 0.8843038822479503,
      "language_loss": 0.61684626,
      "learning_rate": 5.690867251141576e-07,
      "loss": 0.63908547,
      "num_input_tokens_seen": 136053450,
      "step": 6330,
      "time_per_iteration": 4.3029069900512695
    },
    {
      "auxiliary_loss_clip": 0.01130883,
      "auxiliary_loss_mlp": 0.01025795,
      "balance_loss_clip": 1.04916692,
      "balance_loss_mlp": 1.0178256,
      "epoch": 0.7612577406360849,
      "flos": 15633136592640.0,
      "grad_norm": 4.07425225914332,
      "language_loss": 0.91491729,
      "learning_rate": 5.685425995585013e-07,
      "loss": 0.9364841,
      "num_input_tokens_seen": 136071375,
      "step": 6331,
      "time_per_iteration": 2.6235241889953613
    },
    {
      "auxiliary_loss_clip": 0.01179187,
      "auxiliary_loss_mlp": 0.01001127,
      "balance_loss_clip": 1.01063776,
      "balance_loss_mlp": 1.00017965,
      "epoch": 0.761377983526724,
      "flos": 60526253237760.0,
      "grad_norm": 0.7538096601953213,
      "language_loss": 0.58897656,
      "learning_rate": 5.679986911537935e-07,
      "loss": 0.6107797,
      "num_input_tokens_seen": 136138905,
      "step": 6332,
      "time_per_iteration": 3.3508191108703613
    },
    {
      "auxiliary_loss_clip": 0.01370651,
      "auxiliary_loss_mlp": 0.01022495,
      "balance_loss_clip": 1.04670835,
      "balance_loss_mlp": 1.01552379,
      "epoch": 0.7614982264173631,
      "flos": 35772522019200.0,
      "grad_norm": 1.7718832810101364,
      "language_loss": 0.67448497,
      "learning_rate": 5.674549999825462e-07,
      "loss": 0.69841647,
      "num_input_tokens_seen": 136161720,
      "step": 6333,
      "time_per_iteration": 2.825464963912964
    },
    {
      "auxiliary_loss_clip": 0.01121706,
      "auxiliary_loss_mlp": 0.0100214,
      "balance_loss_clip": 1.01036334,
      "balance_loss_mlp": 1.00125742,
      "epoch": 0.7616184693080021,
      "flos": 67925502345600.0,
      "grad_norm": 0.9173419316114866,
      "language_loss": 0.71399301,
      "learning_rate": 5.669115261272363e-07,
      "loss": 0.73523146,
      "num_input_tokens_seen": 136222040,
      "step": 6334,
      "time_per_iteration": 3.2390894889831543
    },
    {
      "auxiliary_loss_clip": 0.01226915,
      "auxiliary_loss_mlp": 0.01022829,
      "balance_loss_clip": 1.0509963,
      "balance_loss_mlp": 1.01552129,
      "epoch": 0.7617387121986413,
      "flos": 20521979141760.0,
      "grad_norm": 2.35391806916075,
      "language_loss": 0.73081839,
      "learning_rate": 5.663682696703081e-07,
      "loss": 0.75331581,
      "num_input_tokens_seen": 136240305,
      "step": 6335,
      "time_per_iteration": 2.696017026901245
    },
    {
      "auxiliary_loss_clip": 0.01173942,
      "auxiliary_loss_mlp": 0.01023404,
      "balance_loss_clip": 1.04988444,
      "balance_loss_mlp": 1.01684475,
      "epoch": 0.7618589550892804,
      "flos": 18624495283200.0,
      "grad_norm": 1.9165617859443422,
      "language_loss": 0.82076812,
      "learning_rate": 5.658252306941746e-07,
      "loss": 0.84274161,
      "num_input_tokens_seen": 136259625,
      "step": 6336,
      "time_per_iteration": 2.659330129623413
    },
    {
      "auxiliary_loss_clip": 0.01371443,
      "auxiliary_loss_mlp": 0.01026341,
      "balance_loss_clip": 1.04677474,
      "balance_loss_mlp": 1.01924527,
      "epoch": 0.7619791979799194,
      "flos": 17453735389440.0,
      "grad_norm": 2.5944498324984284,
      "language_loss": 0.75686073,
      "learning_rate": 5.65282409281212e-07,
      "loss": 0.78083861,
      "num_input_tokens_seen": 136277090,
      "step": 6337,
      "time_per_iteration": 2.8042211532592773
    },
    {
      "auxiliary_loss_clip": 0.01275369,
      "auxiliary_loss_mlp": 0.01024685,
      "balance_loss_clip": 1.04788733,
      "balance_loss_mlp": 1.01755071,
      "epoch": 0.7620994408705585,
      "flos": 14137421333760.0,
      "grad_norm": 2.8596609427690036,
      "language_loss": 0.69944173,
      "learning_rate": 5.64739805513768e-07,
      "loss": 0.72244227,
      "num_input_tokens_seen": 136294635,
      "step": 6338,
      "time_per_iteration": 2.642709732055664
    },
    {
      "auxiliary_loss_clip": 0.01115529,
      "auxiliary_loss_mlp": 0.02504816,
      "balance_loss_clip": 1.00970125,
      "balance_loss_mlp": 0.99989492,
      "epoch": 0.7622196837611976,
      "flos": 70708792527360.0,
      "grad_norm": 0.7927551439912602,
      "language_loss": 0.55635941,
      "learning_rate": 5.641974194741541e-07,
      "loss": 0.59256285,
      "num_input_tokens_seen": 136350320,
      "step": 6339,
      "time_per_iteration": 3.13092303276062
    },
    {
      "auxiliary_loss_clip": 0.01188189,
      "auxiliary_loss_mlp": 0.01017347,
      "balance_loss_clip": 1.03452992,
      "balance_loss_mlp": 1.01614845,
      "epoch": 0.7623399266518367,
      "flos": 60684150447360.0,
      "grad_norm": 0.7963464942061677,
      "language_loss": 0.63733768,
      "learning_rate": 5.636552512446502e-07,
      "loss": 0.65939295,
      "num_input_tokens_seen": 136411375,
      "step": 6340,
      "time_per_iteration": 3.1864728927612305
    },
    {
      "auxiliary_loss_clip": 0.01219218,
      "auxiliary_loss_mlp": 0.01024805,
      "balance_loss_clip": 1.04654527,
      "balance_loss_mlp": 1.01775098,
      "epoch": 0.7624601695424758,
      "flos": 26468893641600.0,
      "grad_norm": 1.6580298578660668,
      "language_loss": 0.77818835,
      "learning_rate": 5.631133009075027e-07,
      "loss": 0.80062854,
      "num_input_tokens_seen": 136430560,
      "step": 6341,
      "time_per_iteration": 2.6504998207092285
    },
    {
      "auxiliary_loss_clip": 0.01224499,
      "auxiliary_loss_mlp": 0.02561144,
      "balance_loss_clip": 1.04954362,
      "balance_loss_mlp": 0.99991286,
      "epoch": 0.7625804124331149,
      "flos": 19135755515520.0,
      "grad_norm": 2.4658359782947614,
      "language_loss": 0.69065249,
      "learning_rate": 5.625715685449242e-07,
      "loss": 0.72850895,
      "num_input_tokens_seen": 136448665,
      "step": 6342,
      "time_per_iteration": 2.60783314704895
    },
    {
      "auxiliary_loss_clip": 0.0122709,
      "auxiliary_loss_mlp": 0.01024359,
      "balance_loss_clip": 1.05282497,
      "balance_loss_mlp": 1.0180707,
      "epoch": 0.762700655323754,
      "flos": 26213101914240.0,
      "grad_norm": 2.6884939738693467,
      "language_loss": 0.71522743,
      "learning_rate": 5.620300542390966e-07,
      "loss": 0.73774189,
      "num_input_tokens_seen": 136469710,
      "step": 6343,
      "time_per_iteration": 2.7197155952453613
    },
    {
      "auxiliary_loss_clip": 0.01171777,
      "auxiliary_loss_mlp": 0.01025715,
      "balance_loss_clip": 1.04606342,
      "balance_loss_mlp": 1.01931369,
      "epoch": 0.762820898214393,
      "flos": 22382582711040.0,
      "grad_norm": 3.309560362732711,
      "language_loss": 0.85277712,
      "learning_rate": 5.614887580721659e-07,
      "loss": 0.87475204,
      "num_input_tokens_seen": 136489855,
      "step": 6344,
      "time_per_iteration": 2.661543846130371
    },
    {
      "auxiliary_loss_clip": 0.01313096,
      "auxiliary_loss_mlp": 0.01021874,
      "balance_loss_clip": 1.04735947,
      "balance_loss_mlp": 1.01470685,
      "epoch": 0.7629411411050322,
      "flos": 15700504550400.0,
      "grad_norm": 2.8495513758690354,
      "language_loss": 0.73645842,
      "learning_rate": 5.609476801262481e-07,
      "loss": 0.75980806,
      "num_input_tokens_seen": 136504715,
      "step": 6345,
      "time_per_iteration": 2.6161885261535645
    },
    {
      "auxiliary_loss_clip": 0.01326244,
      "auxiliary_loss_mlp": 0.01022683,
      "balance_loss_clip": 1.04943061,
      "balance_loss_mlp": 1.01581013,
      "epoch": 0.7630613839956712,
      "flos": 13770342293760.0,
      "grad_norm": 6.430575865007234,
      "language_loss": 0.64449358,
      "learning_rate": 5.604068204834223e-07,
      "loss": 0.66798282,
      "num_input_tokens_seen": 136521610,
      "step": 6346,
      "time_per_iteration": 2.637721061706543
    },
    {
      "auxiliary_loss_clip": 0.01374492,
      "auxiliary_loss_mlp": 0.02567432,
      "balance_loss_clip": 1.04749155,
      "balance_loss_mlp": 0.99991024,
      "epoch": 0.7631816268863103,
      "flos": 14569569861120.0,
      "grad_norm": 2.5277643688928833,
      "language_loss": 0.76891434,
      "learning_rate": 5.598661792257367e-07,
      "loss": 0.80833358,
      "num_input_tokens_seen": 136538655,
      "step": 6347,
      "time_per_iteration": 3.619931697845459
    },
    {
      "auxiliary_loss_clip": 0.01223553,
      "auxiliary_loss_mlp": 0.01024753,
      "balance_loss_clip": 1.04696774,
      "balance_loss_mlp": 1.01808953,
      "epoch": 0.7633018697769495,
      "flos": 19062210418560.0,
      "grad_norm": 2.341477210487527,
      "language_loss": 0.76124513,
      "learning_rate": 5.593257564352071e-07,
      "loss": 0.78372824,
      "num_input_tokens_seen": 136557095,
      "step": 6348,
      "time_per_iteration": 2.6094188690185547
    },
    {
      "auxiliary_loss_clip": 0.01220583,
      "auxiliary_loss_mlp": 0.01023088,
      "balance_loss_clip": 1.04860198,
      "balance_loss_mlp": 1.01607263,
      "epoch": 0.7634221126675885,
      "flos": 22052958577920.0,
      "grad_norm": 2.443947916934433,
      "language_loss": 0.75707972,
      "learning_rate": 5.58785552193815e-07,
      "loss": 0.77951634,
      "num_input_tokens_seen": 136577340,
      "step": 6349,
      "time_per_iteration": 3.5850815773010254
    },
    {
      "auxiliary_loss_clip": 0.01174079,
      "auxiliary_loss_mlp": 0.01027564,
      "balance_loss_clip": 1.04942429,
      "balance_loss_mlp": 1.02108836,
      "epoch": 0.7635423555582276,
      "flos": 29382720825600.0,
      "grad_norm": 1.8056016106256578,
      "language_loss": 0.75427949,
      "learning_rate": 5.582455665835086e-07,
      "loss": 0.77629602,
      "num_input_tokens_seen": 136597635,
      "step": 6350,
      "time_per_iteration": 2.643488645553589
    },
    {
      "auxiliary_loss_clip": 0.01283333,
      "auxiliary_loss_mlp": 0.01031208,
      "balance_loss_clip": 1.0464797,
      "balance_loss_mlp": 1.02378118,
      "epoch": 0.7636625984488667,
      "flos": 17784903807360.0,
      "grad_norm": 3.0254814372482897,
      "language_loss": 0.72408408,
      "learning_rate": 5.577057996862036e-07,
      "loss": 0.74722946,
      "num_input_tokens_seen": 136615260,
      "step": 6351,
      "time_per_iteration": 2.5774903297424316
    },
    {
      "auxiliary_loss_clip": 0.01172487,
      "auxiliary_loss_mlp": 0.01021254,
      "balance_loss_clip": 1.05016351,
      "balance_loss_mlp": 1.01469135,
      "epoch": 0.7637828413395058,
      "flos": 23734583654400.0,
      "grad_norm": 1.6908601970475912,
      "language_loss": 0.7604562,
      "learning_rate": 5.571662515837814e-07,
      "loss": 0.78239357,
      "num_input_tokens_seen": 136637220,
      "step": 6352,
      "time_per_iteration": 2.5868937969207764
    },
    {
      "auxiliary_loss_clip": 0.01271499,
      "auxiliary_loss_mlp": 0.01020889,
      "balance_loss_clip": 1.04817951,
      "balance_loss_mlp": 1.01441264,
      "epoch": 0.7639030842301449,
      "flos": 36283279461120.0,
      "grad_norm": 1.8418662885736523,
      "language_loss": 0.83807063,
      "learning_rate": 5.566269223580926e-07,
      "loss": 0.86099446,
      "num_input_tokens_seen": 136658930,
      "step": 6353,
      "time_per_iteration": 2.758990526199341
    },
    {
      "auxiliary_loss_clip": 0.01226634,
      "auxiliary_loss_mlp": 0.01027837,
      "balance_loss_clip": 1.04864311,
      "balance_loss_mlp": 1.02092028,
      "epoch": 0.764023327120784,
      "flos": 28878104609280.0,
      "grad_norm": 1.7873936643941488,
      "language_loss": 0.75297332,
      "learning_rate": 5.560878120909511e-07,
      "loss": 0.775518,
      "num_input_tokens_seen": 136681530,
      "step": 6354,
      "time_per_iteration": 2.6205759048461914
    },
    {
      "auxiliary_loss_clip": 0.01121981,
      "auxiliary_loss_mlp": 0.00999613,
      "balance_loss_clip": 1.01050353,
      "balance_loss_mlp": 0.99871314,
      "epoch": 0.7641435700114231,
      "flos": 64789711067520.0,
      "grad_norm": 0.8573362773396029,
      "language_loss": 0.58535588,
      "learning_rate": 5.55548920864141e-07,
      "loss": 0.60657179,
      "num_input_tokens_seen": 136742185,
      "step": 6355,
      "time_per_iteration": 3.1821839809417725
    },
    {
      "auxiliary_loss_clip": 0.01224474,
      "auxiliary_loss_mlp": 0.01032981,
      "balance_loss_clip": 1.05105042,
      "balance_loss_mlp": 1.02609658,
      "epoch": 0.7642638129020621,
      "flos": 16835784785280.0,
      "grad_norm": 1.8194907174813417,
      "language_loss": 0.77817619,
      "learning_rate": 5.550102487594113e-07,
      "loss": 0.80075073,
      "num_input_tokens_seen": 136760855,
      "step": 6356,
      "time_per_iteration": 3.4887900352478027
    },
    {
      "auxiliary_loss_clip": 0.01377825,
      "auxiliary_loss_mlp": 0.0256309,
      "balance_loss_clip": 1.04417968,
      "balance_loss_mlp": 0.99988943,
      "epoch": 0.7643840557927013,
      "flos": 30408940391040.0,
      "grad_norm": 1.7314207399295598,
      "language_loss": 0.71496189,
      "learning_rate": 5.54471795858477e-07,
      "loss": 0.75437099,
      "num_input_tokens_seen": 136780925,
      "step": 6357,
      "time_per_iteration": 2.779519557952881
    },
    {
      "auxiliary_loss_clip": 0.01324233,
      "auxiliary_loss_mlp": 0.01021687,
      "balance_loss_clip": 1.04210413,
      "balance_loss_mlp": 1.01513052,
      "epoch": 0.7645042986833404,
      "flos": 16983234115200.0,
      "grad_norm": 2.055811255149393,
      "language_loss": 0.82904482,
      "learning_rate": 5.539335622430235e-07,
      "loss": 0.85250401,
      "num_input_tokens_seen": 136799545,
      "step": 6358,
      "time_per_iteration": 2.6748807430267334
    },
    {
      "auxiliary_loss_clip": 0.01223544,
      "auxiliary_loss_mlp": 0.01026291,
      "balance_loss_clip": 1.04703712,
      "balance_loss_mlp": 1.01953149,
      "epoch": 0.7646245415739794,
      "flos": 17311493531520.0,
      "grad_norm": 2.343368377835373,
      "language_loss": 0.74940574,
      "learning_rate": 5.533955479946975e-07,
      "loss": 0.77190411,
      "num_input_tokens_seen": 136818325,
      "step": 6359,
      "time_per_iteration": 2.6151602268218994
    },
    {
      "auxiliary_loss_clip": 0.01140598,
      "auxiliary_loss_mlp": 0.02505905,
      "balance_loss_clip": 1.03381109,
      "balance_loss_mlp": 0.999856,
      "epoch": 0.7647447844646186,
      "flos": 70402332666240.0,
      "grad_norm": 0.8599922338209752,
      "language_loss": 0.65691185,
      "learning_rate": 5.528577531951173e-07,
      "loss": 0.6933769,
      "num_input_tokens_seen": 136878730,
      "step": 6360,
      "time_per_iteration": 3.262866735458374
    },
    {
      "auxiliary_loss_clip": 0.01275734,
      "auxiliary_loss_mlp": 0.01021492,
      "balance_loss_clip": 1.04701352,
      "balance_loss_mlp": 1.01489711,
      "epoch": 0.7648650273552576,
      "flos": 17675914965120.0,
      "grad_norm": 3.287680098640903,
      "language_loss": 0.74110419,
      "learning_rate": 5.523201779258653e-07,
      "loss": 0.76407647,
      "num_input_tokens_seen": 136897705,
      "step": 6361,
      "time_per_iteration": 2.6236345767974854
    },
    {
      "auxiliary_loss_clip": 0.0117419,
      "auxiliary_loss_mlp": 0.01032654,
      "balance_loss_clip": 1.04879963,
      "balance_loss_mlp": 1.02531099,
      "epoch": 0.7649852702458967,
      "flos": 22162019247360.0,
      "grad_norm": 1.9881197200048353,
      "language_loss": 0.84320873,
      "learning_rate": 5.517828222684912e-07,
      "loss": 0.86527717,
      "num_input_tokens_seen": 136918360,
      "step": 6362,
      "time_per_iteration": 2.61331844329834
    },
    {
      "auxiliary_loss_clip": 0.01170778,
      "auxiliary_loss_mlp": 0.00999132,
      "balance_loss_clip": 1.01129103,
      "balance_loss_mlp": 0.99816054,
      "epoch": 0.7651055131365359,
      "flos": 69848338227840.0,
      "grad_norm": 0.7688963970796907,
      "language_loss": 0.58973449,
      "learning_rate": 5.512456863045117e-07,
      "loss": 0.61143363,
      "num_input_tokens_seen": 136979050,
      "step": 6363,
      "time_per_iteration": 3.2237744331359863
    },
    {
      "auxiliary_loss_clip": 0.01175035,
      "auxiliary_loss_mlp": 0.0102478,
      "balance_loss_clip": 1.04878044,
      "balance_loss_mlp": 1.01746953,
      "epoch": 0.7652257560271749,
      "flos": 19464014931840.0,
      "grad_norm": 2.0772423071249055,
      "language_loss": 0.74018776,
      "learning_rate": 5.507087701154089e-07,
      "loss": 0.76218593,
      "num_input_tokens_seen": 136998970,
      "step": 6364,
      "time_per_iteration": 2.638357400894165
    },
    {
      "auxiliary_loss_clip": 0.01376864,
      "auxiliary_loss_mlp": 0.01024895,
      "balance_loss_clip": 1.04576755,
      "balance_loss_mlp": 1.01817417,
      "epoch": 0.765345998917814,
      "flos": 15961108700160.0,
      "grad_norm": 1.8510273383329774,
      "language_loss": 0.75490171,
      "learning_rate": 5.50172073782634e-07,
      "loss": 0.77891934,
      "num_input_tokens_seen": 137016950,
      "step": 6365,
      "time_per_iteration": 2.7240495681762695
    },
    {
      "auxiliary_loss_clip": 0.01321697,
      "auxiliary_loss_mlp": 0.01025424,
      "balance_loss_clip": 1.04949307,
      "balance_loss_mlp": 1.018888,
      "epoch": 0.7654662418084531,
      "flos": 23659853408640.0,
      "grad_norm": 1.78594469456083,
      "language_loss": 0.87850195,
      "learning_rate": 5.496355973876023e-07,
      "loss": 0.90197319,
      "num_input_tokens_seen": 137036205,
      "step": 6366,
      "time_per_iteration": 2.702868700027466
    },
    {
      "auxiliary_loss_clip": 0.01322706,
      "auxiliary_loss_mlp": 0.02568305,
      "balance_loss_clip": 1.04327083,
      "balance_loss_mlp": 0.99990547,
      "epoch": 0.7655864846990922,
      "flos": 41463608878080.0,
      "grad_norm": 6.089656377979419,
      "language_loss": 0.71362203,
      "learning_rate": 5.490993410116984e-07,
      "loss": 0.75253212,
      "num_input_tokens_seen": 137059195,
      "step": 6367,
      "time_per_iteration": 2.882114887237549
    },
    {
      "auxiliary_loss_clip": 0.01322722,
      "auxiliary_loss_mlp": 0.01024867,
      "balance_loss_clip": 1.04580677,
      "balance_loss_mlp": 1.01786053,
      "epoch": 0.7657067275897312,
      "flos": 43142684088960.0,
      "grad_norm": 1.6118983368832827,
      "language_loss": 0.69863456,
      "learning_rate": 5.485633047362704e-07,
      "loss": 0.72211051,
      "num_input_tokens_seen": 137081200,
      "step": 6368,
      "time_per_iteration": 2.859867572784424
    },
    {
      "auxiliary_loss_clip": 0.01180139,
      "auxiliary_loss_mlp": 0.01026924,
      "balance_loss_clip": 1.05332935,
      "balance_loss_mlp": 1.01993489,
      "epoch": 0.7658269704803703,
      "flos": 17311780840320.0,
      "grad_norm": 2.773695799726802,
      "language_loss": 0.78935415,
      "learning_rate": 5.480274886426341e-07,
      "loss": 0.81142473,
      "num_input_tokens_seen": 137097840,
      "step": 6369,
      "time_per_iteration": 2.6259210109710693
    },
    {
      "auxiliary_loss_clip": 0.0122593,
      "auxiliary_loss_mlp": 0.01021996,
      "balance_loss_clip": 1.0520364,
      "balance_loss_mlp": 1.01543057,
      "epoch": 0.7659472133710095,
      "flos": 12568160977920.0,
      "grad_norm": 2.2402579583670037,
      "language_loss": 0.77890122,
      "learning_rate": 5.474918928120744e-07,
      "loss": 0.80138052,
      "num_input_tokens_seen": 137114335,
      "step": 6370,
      "time_per_iteration": 2.5898306369781494
    },
    {
      "auxiliary_loss_clip": 0.01219138,
      "auxiliary_loss_mlp": 0.01029296,
      "balance_loss_clip": 1.04704833,
      "balance_loss_mlp": 1.02246869,
      "epoch": 0.7660674562616485,
      "flos": 22707430335360.0,
      "grad_norm": 1.941264987992297,
      "language_loss": 0.87429225,
      "learning_rate": 5.469565173258392e-07,
      "loss": 0.89677668,
      "num_input_tokens_seen": 137132850,
      "step": 6371,
      "time_per_iteration": 2.6447532176971436
    },
    {
      "auxiliary_loss_clip": 0.01178463,
      "auxiliary_loss_mlp": 0.01022354,
      "balance_loss_clip": 1.05030262,
      "balance_loss_mlp": 1.01484966,
      "epoch": 0.7661876991522876,
      "flos": 17056455989760.0,
      "grad_norm": 1.8012704733051066,
      "language_loss": 0.64102608,
      "learning_rate": 5.464213622651454e-07,
      "loss": 0.6630342,
      "num_input_tokens_seen": 137150665,
      "step": 6372,
      "time_per_iteration": 3.413572072982788
    },
    {
      "auxiliary_loss_clip": 0.01326648,
      "auxiliary_loss_mlp": 0.01021989,
      "balance_loss_clip": 1.04793692,
      "balance_loss_mlp": 1.01521456,
      "epoch": 0.7663079420429267,
      "flos": 20084228092800.0,
      "grad_norm": 2.073583220253837,
      "language_loss": 0.84306288,
      "learning_rate": 5.458864277111753e-07,
      "loss": 0.86654925,
      "num_input_tokens_seen": 137168500,
      "step": 6373,
      "time_per_iteration": 3.618757486343384
    },
    {
      "auxiliary_loss_clip": 0.01269321,
      "auxiliary_loss_mlp": 0.02561494,
      "balance_loss_clip": 1.04632068,
      "balance_loss_mlp": 0.99990529,
      "epoch": 0.7664281849335658,
      "flos": 12677473042560.0,
      "grad_norm": 2.60596640838214,
      "language_loss": 0.68989909,
      "learning_rate": 5.453517137450769e-07,
      "loss": 0.72820723,
      "num_input_tokens_seen": 137185075,
      "step": 6374,
      "time_per_iteration": 2.651176691055298
    },
    {
      "auxiliary_loss_clip": 0.0122373,
      "auxiliary_loss_mlp": 0.0102204,
      "balance_loss_clip": 1.0505569,
      "balance_loss_mlp": 1.01551378,
      "epoch": 0.7665484278242048,
      "flos": 22345271458560.0,
      "grad_norm": 1.6386470170136365,
      "language_loss": 0.76115036,
      "learning_rate": 5.448172204479684e-07,
      "loss": 0.78360808,
      "num_input_tokens_seen": 137204355,
      "step": 6375,
      "time_per_iteration": 3.4458539485931396
    },
    {
      "auxiliary_loss_clip": 0.01173222,
      "auxiliary_loss_mlp": 0.01028817,
      "balance_loss_clip": 1.04908621,
      "balance_loss_mlp": 1.02150702,
      "epoch": 0.766668670714844,
      "flos": 23617909301760.0,
      "grad_norm": 1.7971847543412012,
      "language_loss": 0.74655199,
      "learning_rate": 5.442829479009294e-07,
      "loss": 0.76857239,
      "num_input_tokens_seen": 137223135,
      "step": 6376,
      "time_per_iteration": 2.624239921569824
    },
    {
      "auxiliary_loss_clip": 0.01130667,
      "auxiliary_loss_mlp": 0.01026729,
      "balance_loss_clip": 1.04805923,
      "balance_loss_mlp": 1.01958799,
      "epoch": 0.7667889136054831,
      "flos": 19427134642560.0,
      "grad_norm": 1.8226818016947173,
      "language_loss": 0.71791887,
      "learning_rate": 5.437488961850103e-07,
      "loss": 0.73949277,
      "num_input_tokens_seen": 137242935,
      "step": 6377,
      "time_per_iteration": 2.619051694869995
    },
    {
      "auxiliary_loss_clip": 0.01364906,
      "auxiliary_loss_mlp": 0.01023564,
      "balance_loss_clip": 1.04463446,
      "balance_loss_mlp": 1.01752353,
      "epoch": 0.7669091564961221,
      "flos": 26866352609280.0,
      "grad_norm": 1.9187707125703009,
      "language_loss": 0.75435752,
      "learning_rate": 5.432150653812258e-07,
      "loss": 0.77824223,
      "num_input_tokens_seen": 137262970,
      "step": 6378,
      "time_per_iteration": 2.756120204925537
    },
    {
      "auxiliary_loss_clip": 0.01222419,
      "auxiliary_loss_mlp": 0.01026785,
      "balance_loss_clip": 1.05013442,
      "balance_loss_mlp": 1.02001381,
      "epoch": 0.7670293993867613,
      "flos": 12385303816320.0,
      "grad_norm": 2.0807618059489443,
      "language_loss": 0.8236661,
      "learning_rate": 5.42681455570557e-07,
      "loss": 0.84615821,
      "num_input_tokens_seen": 137279500,
      "step": 6379,
      "time_per_iteration": 2.58658766746521
    },
    {
      "auxiliary_loss_clip": 0.01173466,
      "auxiliary_loss_mlp": 0.01023039,
      "balance_loss_clip": 1.05024743,
      "balance_loss_mlp": 1.01629198,
      "epoch": 0.7671496422774003,
      "flos": 21762944167680.0,
      "grad_norm": 2.8160949659949943,
      "language_loss": 0.64803928,
      "learning_rate": 5.42148066833954e-07,
      "loss": 0.67000437,
      "num_input_tokens_seen": 137298745,
      "step": 6380,
      "time_per_iteration": 2.614374876022339
    },
    {
      "auxiliary_loss_clip": 0.01173919,
      "auxiliary_loss_mlp": 0.0102105,
      "balance_loss_clip": 1.05061543,
      "balance_loss_mlp": 1.01480937,
      "epoch": 0.7672698851680394,
      "flos": 21069221823360.0,
      "grad_norm": 2.285416712495025,
      "language_loss": 0.75487685,
      "learning_rate": 5.416148992523289e-07,
      "loss": 0.77682656,
      "num_input_tokens_seen": 137317320,
      "step": 6381,
      "time_per_iteration": 2.654759407043457
    },
    {
      "auxiliary_loss_clip": 0.01470306,
      "auxiliary_loss_mlp": 0.01027333,
      "balance_loss_clip": 1.04086769,
      "balance_loss_mlp": 1.02083659,
      "epoch": 0.7673901280586786,
      "flos": 16976697840000.0,
      "grad_norm": 1.9714192319747155,
      "language_loss": 0.78554267,
      "learning_rate": 5.410819529065644e-07,
      "loss": 0.81051898,
      "num_input_tokens_seen": 137335275,
      "step": 6382,
      "time_per_iteration": 3.767184019088745
    },
    {
      "auxiliary_loss_clip": 0.01367621,
      "auxiliary_loss_mlp": 0.01023891,
      "balance_loss_clip": 1.04155076,
      "balance_loss_mlp": 1.01735795,
      "epoch": 0.7675103709493176,
      "flos": 29242669697280.0,
      "grad_norm": 2.0327745553595844,
      "language_loss": 0.6518271,
      "learning_rate": 5.405492278775079e-07,
      "loss": 0.67574227,
      "num_input_tokens_seen": 137355055,
      "step": 6383,
      "time_per_iteration": 2.9145278930664062
    },
    {
      "auxiliary_loss_clip": 0.01277852,
      "auxiliary_loss_mlp": 0.01026105,
      "balance_loss_clip": 1.04433346,
      "balance_loss_mlp": 1.01865721,
      "epoch": 0.7676306138399567,
      "flos": 29023004073600.0,
      "grad_norm": 2.335101918689425,
      "language_loss": 0.79958642,
      "learning_rate": 5.400167242459732e-07,
      "loss": 0.82262605,
      "num_input_tokens_seen": 137374015,
      "step": 6384,
      "time_per_iteration": 2.7245540618896484
    },
    {
      "auxiliary_loss_clip": 0.01221424,
      "auxiliary_loss_mlp": 0.0102341,
      "balance_loss_clip": 1.04753029,
      "balance_loss_mlp": 1.01717246,
      "epoch": 0.7677508567305958,
      "flos": 22565116650240.0,
      "grad_norm": 10.442047718374237,
      "language_loss": 0.80464089,
      "learning_rate": 5.394844420927405e-07,
      "loss": 0.82708919,
      "num_input_tokens_seen": 137393625,
      "step": 6385,
      "time_per_iteration": 2.644745111465454
    },
    {
      "auxiliary_loss_clip": 0.01172842,
      "auxiliary_loss_mlp": 0.0102461,
      "balance_loss_clip": 1.04945946,
      "balance_loss_mlp": 1.01809525,
      "epoch": 0.7678710996212349,
      "flos": 25411432222080.0,
      "grad_norm": 2.4062462420811586,
      "language_loss": 0.73258698,
      "learning_rate": 5.389523814985562e-07,
      "loss": 0.75456154,
      "num_input_tokens_seen": 137413045,
      "step": 6386,
      "time_per_iteration": 2.624577045440674
    },
    {
      "auxiliary_loss_clip": 0.01376322,
      "auxiliary_loss_mlp": 0.01033042,
      "balance_loss_clip": 1.04382646,
      "balance_loss_mlp": 1.02608562,
      "epoch": 0.767991342511874,
      "flos": 26756825063040.0,
      "grad_norm": 1.7679749184341063,
      "language_loss": 0.76124573,
      "learning_rate": 5.384205425441344e-07,
      "loss": 0.78533942,
      "num_input_tokens_seen": 137433955,
      "step": 6387,
      "time_per_iteration": 2.8141911029815674
    },
    {
      "auxiliary_loss_clip": 0.0126843,
      "auxiliary_loss_mlp": 0.01024478,
      "balance_loss_clip": 1.04369438,
      "balance_loss_mlp": 1.01784706,
      "epoch": 0.7681115854025131,
      "flos": 26359509749760.0,
      "grad_norm": 1.6216784590860933,
      "language_loss": 0.84336627,
      "learning_rate": 5.378889253101537e-07,
      "loss": 0.86629534,
      "num_input_tokens_seen": 137454510,
      "step": 6388,
      "time_per_iteration": 2.710787057876587
    },
    {
      "auxiliary_loss_clip": 0.01222594,
      "auxiliary_loss_mlp": 0.01021211,
      "balance_loss_clip": 1.04518962,
      "balance_loss_mlp": 1.01478601,
      "epoch": 0.7682318282931522,
      "flos": 23257043314560.0,
      "grad_norm": 1.6922651094101255,
      "language_loss": 0.81228715,
      "learning_rate": 5.373575298772617e-07,
      "loss": 0.83472514,
      "num_input_tokens_seen": 137473630,
      "step": 6389,
      "time_per_iteration": 2.6289193630218506
    },
    {
      "auxiliary_loss_clip": 0.01125612,
      "auxiliary_loss_mlp": 0.01000475,
      "balance_loss_clip": 1.01079929,
      "balance_loss_mlp": 0.99944353,
      "epoch": 0.7683520711837912,
      "flos": 70072457137920.0,
      "grad_norm": 0.7563001926213312,
      "language_loss": 0.61227423,
      "learning_rate": 5.368263563260689e-07,
      "loss": 0.63353509,
      "num_input_tokens_seen": 137538765,
      "step": 6390,
      "time_per_iteration": 3.297755241394043
    },
    {
      "auxiliary_loss_clip": 0.01223949,
      "auxiliary_loss_mlp": 0.01024931,
      "balance_loss_clip": 1.04659617,
      "balance_loss_mlp": 1.0181365,
      "epoch": 0.7684723140744304,
      "flos": 18624890332800.0,
      "grad_norm": 2.177679176549905,
      "language_loss": 0.64145339,
      "learning_rate": 5.362954047371537e-07,
      "loss": 0.6639421,
      "num_input_tokens_seen": 137557875,
      "step": 6391,
      "time_per_iteration": 2.6172263622283936
    },
    {
      "auxiliary_loss_clip": 0.01324643,
      "auxiliary_loss_mlp": 0.01024143,
      "balance_loss_clip": 1.05180931,
      "balance_loss_mlp": 1.01674032,
      "epoch": 0.7685925569650695,
      "flos": 27452989532160.0,
      "grad_norm": 1.8243167702416283,
      "language_loss": 0.72368777,
      "learning_rate": 5.357646751910627e-07,
      "loss": 0.74717557,
      "num_input_tokens_seen": 137579055,
      "step": 6392,
      "time_per_iteration": 2.713099479675293
    },
    {
      "auxiliary_loss_clip": 0.01271372,
      "auxiliary_loss_mlp": 0.01028193,
      "balance_loss_clip": 1.04685283,
      "balance_loss_mlp": 1.02054918,
      "epoch": 0.7687127998557085,
      "flos": 24535714642560.0,
      "grad_norm": 2.523214208722722,
      "language_loss": 0.80119216,
      "learning_rate": 5.352341677683061e-07,
      "loss": 0.82418787,
      "num_input_tokens_seen": 137600355,
      "step": 6393,
      "time_per_iteration": 2.7305908203125
    },
    {
      "auxiliary_loss_clip": 0.01329323,
      "auxiliary_loss_mlp": 0.0102644,
      "balance_loss_clip": 1.04631996,
      "balance_loss_mlp": 1.01935017,
      "epoch": 0.7688330427463477,
      "flos": 25155963717120.0,
      "grad_norm": 1.790487169270661,
      "language_loss": 0.79074955,
      "learning_rate": 5.347038825493617e-07,
      "loss": 0.81430709,
      "num_input_tokens_seen": 137621885,
      "step": 6394,
      "time_per_iteration": 2.7279727458953857
    },
    {
      "auxiliary_loss_clip": 0.01267545,
      "auxiliary_loss_mlp": 0.01029851,
      "balance_loss_clip": 1.04838228,
      "balance_loss_mlp": 1.02330637,
      "epoch": 0.7689532856369867,
      "flos": 21211284113280.0,
      "grad_norm": 2.726584023331789,
      "language_loss": 0.68433857,
      "learning_rate": 5.341738196146732e-07,
      "loss": 0.70731258,
      "num_input_tokens_seen": 137640230,
      "step": 6395,
      "time_per_iteration": 2.6466407775878906
    },
    {
      "auxiliary_loss_clip": 0.01222439,
      "auxiliary_loss_mlp": 0.01029972,
      "balance_loss_clip": 1.0463109,
      "balance_loss_mlp": 1.02347231,
      "epoch": 0.7690735285276258,
      "flos": 25119083427840.0,
      "grad_norm": 2.693422032812867,
      "language_loss": 0.73623395,
      "learning_rate": 5.336439790446503e-07,
      "loss": 0.75875807,
      "num_input_tokens_seen": 137659330,
      "step": 6396,
      "time_per_iteration": 2.655944347381592
    },
    {
      "auxiliary_loss_clip": 0.01323747,
      "auxiliary_loss_mlp": 0.01023373,
      "balance_loss_clip": 1.04305947,
      "balance_loss_mlp": 1.01595211,
      "epoch": 0.769193771418265,
      "flos": 54744020640000.0,
      "grad_norm": 2.6714074853121517,
      "language_loss": 0.61972147,
      "learning_rate": 5.331143609196711e-07,
      "loss": 0.64319265,
      "num_input_tokens_seen": 137683145,
      "step": 6397,
      "time_per_iteration": 3.0158629417419434
    },
    {
      "auxiliary_loss_clip": 0.01221345,
      "auxiliary_loss_mlp": 0.01025488,
      "balance_loss_clip": 1.04836178,
      "balance_loss_mlp": 1.01855576,
      "epoch": 0.769314014308904,
      "flos": 37341890115840.0,
      "grad_norm": 1.7257421082034516,
      "language_loss": 0.77462626,
      "learning_rate": 5.325849653200758e-07,
      "loss": 0.79709464,
      "num_input_tokens_seen": 137707095,
      "step": 6398,
      "time_per_iteration": 3.8367178440093994
    },
    {
      "auxiliary_loss_clip": 0.01175549,
      "auxiliary_loss_mlp": 0.01021969,
      "balance_loss_clip": 1.05140424,
      "balance_loss_mlp": 1.01524889,
      "epoch": 0.7694342571995431,
      "flos": 20631686256000.0,
      "grad_norm": 1.6760277400774874,
      "language_loss": 0.76595366,
      "learning_rate": 5.32055792326175e-07,
      "loss": 0.78792882,
      "num_input_tokens_seen": 137725520,
      "step": 6399,
      "time_per_iteration": 3.521735191345215
    },
    {
      "auxiliary_loss_clip": 0.01274843,
      "auxiliary_loss_mlp": 0.01025059,
      "balance_loss_clip": 1.04898024,
      "balance_loss_mlp": 1.01784933,
      "epoch": 0.7695545000901821,
      "flos": 24207706621440.0,
      "grad_norm": 2.2236331244873386,
      "language_loss": 0.72574556,
      "learning_rate": 5.315268420182437e-07,
      "loss": 0.74874461,
      "num_input_tokens_seen": 137744195,
      "step": 6400,
      "time_per_iteration": 2.651752233505249
    },
    {
      "auxiliary_loss_clip": 0.01327906,
      "auxiliary_loss_mlp": 0.02564643,
      "balance_loss_clip": 1.04676604,
      "balance_loss_mlp": 0.99991435,
      "epoch": 0.7696747429808213,
      "flos": 28001273708160.0,
      "grad_norm": 2.127844734913713,
      "language_loss": 0.76421332,
      "learning_rate": 5.309981144765221e-07,
      "loss": 0.80313885,
      "num_input_tokens_seen": 137764340,
      "step": 6401,
      "time_per_iteration": 3.5868334770202637
    },
    {
      "auxiliary_loss_clip": 0.01376839,
      "auxiliary_loss_mlp": 0.01023819,
      "balance_loss_clip": 1.04418647,
      "balance_loss_mlp": 1.01694918,
      "epoch": 0.7697949858714603,
      "flos": 11509550323200.0,
      "grad_norm": 2.4744340979915402,
      "language_loss": 0.75721788,
      "learning_rate": 5.304696097812196e-07,
      "loss": 0.78122449,
      "num_input_tokens_seen": 137780940,
      "step": 6402,
      "time_per_iteration": 2.712481737136841
    },
    {
      "auxiliary_loss_clip": 0.01270338,
      "auxiliary_loss_mlp": 0.0102368,
      "balance_loss_clip": 1.04413271,
      "balance_loss_mlp": 1.01615834,
      "epoch": 0.7699152287620994,
      "flos": 26688271956480.0,
      "grad_norm": 3.598583548951113,
      "language_loss": 0.61135828,
      "learning_rate": 5.299413280125078e-07,
      "loss": 0.6342985,
      "num_input_tokens_seen": 137799250,
      "step": 6403,
      "time_per_iteration": 2.6653027534484863
    },
    {
      "auxiliary_loss_clip": 0.01273017,
      "auxiliary_loss_mlp": 0.01033394,
      "balance_loss_clip": 1.04522967,
      "balance_loss_mlp": 1.02622354,
      "epoch": 0.7700354716527386,
      "flos": 16544944362240.0,
      "grad_norm": 2.0702568109659674,
      "language_loss": 0.72861302,
      "learning_rate": 5.294132692505284e-07,
      "loss": 0.75167716,
      "num_input_tokens_seen": 137817660,
      "step": 6404,
      "time_per_iteration": 2.688532590866089
    },
    {
      "auxiliary_loss_clip": 0.01364776,
      "auxiliary_loss_mlp": 0.01026767,
      "balance_loss_clip": 1.04010117,
      "balance_loss_mlp": 1.01990318,
      "epoch": 0.7701557145433776,
      "flos": 19242733196160.0,
      "grad_norm": 2.102711346714565,
      "language_loss": 0.79165745,
      "learning_rate": 5.288854335753861e-07,
      "loss": 0.81557286,
      "num_input_tokens_seen": 137835920,
      "step": 6405,
      "time_per_iteration": 2.74121356010437
    },
    {
      "auxiliary_loss_clip": 0.01226329,
      "auxiliary_loss_mlp": 0.01022316,
      "balance_loss_clip": 1.04867327,
      "balance_loss_mlp": 1.01549423,
      "epoch": 0.7702759574340167,
      "flos": 31685744211840.0,
      "grad_norm": 1.5784211072343581,
      "language_loss": 0.75720203,
      "learning_rate": 5.283578210671551e-07,
      "loss": 0.77968848,
      "num_input_tokens_seen": 137858160,
      "step": 6406,
      "time_per_iteration": 2.734612226486206
    },
    {
      "auxiliary_loss_clip": 0.01276237,
      "auxiliary_loss_mlp": 0.0102306,
      "balance_loss_clip": 1.04723489,
      "balance_loss_mlp": 1.01643467,
      "epoch": 0.7703962003246558,
      "flos": 16800089644800.0,
      "grad_norm": 3.050296277591386,
      "language_loss": 0.76949406,
      "learning_rate": 5.278304318058719e-07,
      "loss": 0.79248703,
      "num_input_tokens_seen": 137876015,
      "step": 6407,
      "time_per_iteration": 2.715555429458618
    },
    {
      "auxiliary_loss_clip": 0.01424308,
      "auxiliary_loss_mlp": 0.01029028,
      "balance_loss_clip": 1.04459691,
      "balance_loss_mlp": 1.02217662,
      "epoch": 0.7705164432152949,
      "flos": 35736072693120.0,
      "grad_norm": 1.7206763388753927,
      "language_loss": 0.79287958,
      "learning_rate": 5.273032658715411e-07,
      "loss": 0.81741285,
      "num_input_tokens_seen": 137898825,
      "step": 6408,
      "time_per_iteration": 3.7544682025909424
    },
    {
      "auxiliary_loss_clip": 0.01367897,
      "auxiliary_loss_mlp": 0.010273,
      "balance_loss_clip": 1.04206109,
      "balance_loss_mlp": 1.0202632,
      "epoch": 0.7706366861059339,
      "flos": 23365960329600.0,
      "grad_norm": 2.1179562836836423,
      "language_loss": 0.76555061,
      "learning_rate": 5.267763233441347e-07,
      "loss": 0.78950262,
      "num_input_tokens_seen": 137919455,
      "step": 6409,
      "time_per_iteration": 2.7582266330718994
    },
    {
      "auxiliary_loss_clip": 0.01226328,
      "auxiliary_loss_mlp": 0.01033633,
      "balance_loss_clip": 1.04942012,
      "balance_loss_mlp": 1.02657604,
      "epoch": 0.7707569289965731,
      "flos": 22929897219840.0,
      "grad_norm": 3.4484466487327077,
      "language_loss": 0.69967902,
      "learning_rate": 5.26249604303588e-07,
      "loss": 0.72227865,
      "num_input_tokens_seen": 137937960,
      "step": 6410,
      "time_per_iteration": 2.641677141189575
    },
    {
      "auxiliary_loss_clip": 0.01174421,
      "auxiliary_loss_mlp": 0.01021831,
      "balance_loss_clip": 1.04992557,
      "balance_loss_mlp": 1.0151794,
      "epoch": 0.7708771718872122,
      "flos": 17420661941760.0,
      "grad_norm": 2.138439361576134,
      "language_loss": 0.78342903,
      "learning_rate": 5.257231088298057e-07,
      "loss": 0.80539155,
      "num_input_tokens_seen": 137956370,
      "step": 6411,
      "time_per_iteration": 2.5665149688720703
    },
    {
      "auxiliary_loss_clip": 0.01220209,
      "auxiliary_loss_mlp": 0.00999196,
      "balance_loss_clip": 1.0115912,
      "balance_loss_mlp": 0.99829596,
      "epoch": 0.7709974147778512,
      "flos": 72241316248320.0,
      "grad_norm": 0.7985990304701299,
      "language_loss": 0.53899252,
      "learning_rate": 5.25196837002655e-07,
      "loss": 0.56118655,
      "num_input_tokens_seen": 138016080,
      "step": 6412,
      "time_per_iteration": 3.30989670753479
    },
    {
      "auxiliary_loss_clip": 0.0126958,
      "auxiliary_loss_mlp": 0.01023811,
      "balance_loss_clip": 1.0456866,
      "balance_loss_mlp": 1.0168556,
      "epoch": 0.7711176576684904,
      "flos": 39859694876160.0,
      "grad_norm": 6.163278091915123,
      "language_loss": 0.68266058,
      "learning_rate": 5.24670788901971e-07,
      "loss": 0.70559454,
      "num_input_tokens_seen": 138039170,
      "step": 6413,
      "time_per_iteration": 2.78220796585083
    },
    {
      "auxiliary_loss_clip": 0.0127679,
      "auxiliary_loss_mlp": 0.01030582,
      "balance_loss_clip": 1.04852724,
      "balance_loss_mlp": 1.02241671,
      "epoch": 0.7712379005591294,
      "flos": 36976391274240.0,
      "grad_norm": 3.013585676152656,
      "language_loss": 0.69064665,
      "learning_rate": 5.241449646075557e-07,
      "loss": 0.71372038,
      "num_input_tokens_seen": 138062395,
      "step": 6414,
      "time_per_iteration": 2.8258821964263916
    },
    {
      "auxiliary_loss_clip": 0.01229859,
      "auxiliary_loss_mlp": 0.01025935,
      "balance_loss_clip": 1.04779184,
      "balance_loss_mlp": 1.01912832,
      "epoch": 0.7713581434497685,
      "flos": 22776773541120.0,
      "grad_norm": 2.0126002298655776,
      "language_loss": 0.72369552,
      "learning_rate": 5.236193641991762e-07,
      "loss": 0.74625349,
      "num_input_tokens_seen": 138080325,
      "step": 6415,
      "time_per_iteration": 2.626279592514038
    },
    {
      "auxiliary_loss_clip": 0.01265638,
      "auxiliary_loss_mlp": 0.01023538,
      "balance_loss_clip": 1.04645908,
      "balance_loss_mlp": 1.01712751,
      "epoch": 0.7714783863404077,
      "flos": 24097460803200.0,
      "grad_norm": 2.0401778793619094,
      "language_loss": 0.69792545,
      "learning_rate": 5.23093987756565e-07,
      "loss": 0.72081721,
      "num_input_tokens_seen": 138099020,
      "step": 6416,
      "time_per_iteration": 2.692166566848755
    },
    {
      "auxiliary_loss_clip": 0.01330639,
      "auxiliary_loss_mlp": 0.01024113,
      "balance_loss_clip": 1.04472113,
      "balance_loss_mlp": 1.01678193,
      "epoch": 0.7715986292310467,
      "flos": 21063655215360.0,
      "grad_norm": 2.0965703144657626,
      "language_loss": 0.75387222,
      "learning_rate": 5.225688353594217e-07,
      "loss": 0.77741981,
      "num_input_tokens_seen": 138118650,
      "step": 6417,
      "time_per_iteration": 2.745495080947876
    },
    {
      "auxiliary_loss_clip": 0.01279066,
      "auxiliary_loss_mlp": 0.02563431,
      "balance_loss_clip": 1.04935801,
      "balance_loss_mlp": 0.99993861,
      "epoch": 0.7717188721216858,
      "flos": 20594877793920.0,
      "grad_norm": 2.197535586187941,
      "language_loss": 0.77938187,
      "learning_rate": 5.220439070874108e-07,
      "loss": 0.81780684,
      "num_input_tokens_seen": 138137890,
      "step": 6418,
      "time_per_iteration": 2.6982784271240234
    },
    {
      "auxiliary_loss_clip": 0.01222065,
      "auxiliary_loss_mlp": 0.01031215,
      "balance_loss_clip": 1.04935265,
      "balance_loss_mlp": 1.02453029,
      "epoch": 0.7718391150123249,
      "flos": 26250951870720.0,
      "grad_norm": 1.6842850282506423,
      "language_loss": 0.70859224,
      "learning_rate": 5.215192030201652e-07,
      "loss": 0.73112506,
      "num_input_tokens_seen": 138158880,
      "step": 6419,
      "time_per_iteration": 2.7129921913146973
    },
    {
      "auxiliary_loss_clip": 0.01310627,
      "auxiliary_loss_mlp": 0.01025155,
      "balance_loss_clip": 1.03911924,
      "balance_loss_mlp": 1.01796639,
      "epoch": 0.771959357902964,
      "flos": 22049762267520.0,
      "grad_norm": 1.85707756961009,
      "language_loss": 0.86287105,
      "learning_rate": 5.209947232372798e-07,
      "loss": 0.8862288,
      "num_input_tokens_seen": 138176370,
      "step": 6420,
      "time_per_iteration": 2.7136452198028564
    },
    {
      "auxiliary_loss_clip": 0.01226259,
      "auxiliary_loss_mlp": 0.02565691,
      "balance_loss_clip": 1.04714346,
      "balance_loss_mlp": 0.99993813,
      "epoch": 0.772079600793603,
      "flos": 30446000248320.0,
      "grad_norm": 2.0496735105689643,
      "language_loss": 0.80952334,
      "learning_rate": 5.204704678183196e-07,
      "loss": 0.84744287,
      "num_input_tokens_seen": 138195105,
      "step": 6421,
      "time_per_iteration": 2.767970085144043
    },
    {
      "auxiliary_loss_clip": 0.01175639,
      "auxiliary_loss_mlp": 0.01029449,
      "balance_loss_clip": 1.05108261,
      "balance_loss_mlp": 1.02243972,
      "epoch": 0.7721998436842422,
      "flos": 12969857750400.0,
      "grad_norm": 2.1006012463167583,
      "language_loss": 0.85202432,
      "learning_rate": 5.19946436842813e-07,
      "loss": 0.87407517,
      "num_input_tokens_seen": 138212235,
      "step": 6422,
      "time_per_iteration": 2.5519728660583496
    },
    {
      "auxiliary_loss_clip": 0.01328239,
      "auxiliary_loss_mlp": 0.01023478,
      "balance_loss_clip": 1.04884684,
      "balance_loss_mlp": 1.01663566,
      "epoch": 0.7723200865748813,
      "flos": 32635509678720.0,
      "grad_norm": 1.7415384055499747,
      "language_loss": 0.68658686,
      "learning_rate": 5.194226303902546e-07,
      "loss": 0.71010411,
      "num_input_tokens_seen": 138231970,
      "step": 6423,
      "time_per_iteration": 2.7906036376953125
    },
    {
      "auxiliary_loss_clip": 0.01270003,
      "auxiliary_loss_mlp": 0.01031465,
      "balance_loss_clip": 1.04536963,
      "balance_loss_mlp": 1.02434492,
      "epoch": 0.7724403294655203,
      "flos": 21105707063040.0,
      "grad_norm": 1.8047932794349952,
      "language_loss": 0.7113446,
      "learning_rate": 5.188990485401072e-07,
      "loss": 0.73435926,
      "num_input_tokens_seen": 138251175,
      "step": 6424,
      "time_per_iteration": 3.556642770767212
    },
    {
      "auxiliary_loss_clip": 0.01224907,
      "auxiliary_loss_mlp": 0.01029786,
      "balance_loss_clip": 1.04912043,
      "balance_loss_mlp": 1.02246666,
      "epoch": 0.7725605723561595,
      "flos": 22090736707200.0,
      "grad_norm": 2.25428871343897,
      "language_loss": 0.86459869,
      "learning_rate": 5.183756913717954e-07,
      "loss": 0.88714564,
      "num_input_tokens_seen": 138270950,
      "step": 6425,
      "time_per_iteration": 3.5687780380249023
    },
    {
      "auxiliary_loss_clip": 0.01269013,
      "auxiliary_loss_mlp": 0.01024313,
      "balance_loss_clip": 1.04732037,
      "balance_loss_mlp": 1.01746416,
      "epoch": 0.7726808152467985,
      "flos": 34495610457600.0,
      "grad_norm": 2.270852678915111,
      "language_loss": 0.73507631,
      "learning_rate": 5.178525589647136e-07,
      "loss": 0.75800955,
      "num_input_tokens_seen": 138292590,
      "step": 6426,
      "time_per_iteration": 2.8162386417388916
    },
    {
      "auxiliary_loss_clip": 0.01279962,
      "auxiliary_loss_mlp": 0.01024724,
      "balance_loss_clip": 1.04778659,
      "balance_loss_mlp": 1.01809573,
      "epoch": 0.7728010581374376,
      "flos": 22306344094080.0,
      "grad_norm": 1.8748520035517338,
      "language_loss": 0.78602397,
      "learning_rate": 5.173296513982197e-07,
      "loss": 0.80907083,
      "num_input_tokens_seen": 138311115,
      "step": 6427,
      "time_per_iteration": 3.540137529373169
    },
    {
      "auxiliary_loss_clip": 0.01333901,
      "auxiliary_loss_mlp": 0.01026818,
      "balance_loss_clip": 1.04814291,
      "balance_loss_mlp": 1.01972163,
      "epoch": 0.7729213010280768,
      "flos": 27126453968640.0,
      "grad_norm": 3.731314244208566,
      "language_loss": 0.65029931,
      "learning_rate": 5.168069687516398e-07,
      "loss": 0.67390645,
      "num_input_tokens_seen": 138330885,
      "step": 6428,
      "time_per_iteration": 2.761063814163208
    },
    {
      "auxiliary_loss_clip": 0.01272299,
      "auxiliary_loss_mlp": 0.01032362,
      "balance_loss_clip": 1.04906189,
      "balance_loss_mlp": 1.02530456,
      "epoch": 0.7730415439187158,
      "flos": 18150223080960.0,
      "grad_norm": 1.9176804301240105,
      "language_loss": 0.72029316,
      "learning_rate": 5.16284511104263e-07,
      "loss": 0.74333978,
      "num_input_tokens_seen": 138350020,
      "step": 6429,
      "time_per_iteration": 2.646824359893799
    },
    {
      "auxiliary_loss_clip": 0.01268937,
      "auxiliary_loss_mlp": 0.01024624,
      "balance_loss_clip": 1.04684663,
      "balance_loss_mlp": 1.01767099,
      "epoch": 0.7731617868093549,
      "flos": 11947480940160.0,
      "grad_norm": 3.0275037319196563,
      "language_loss": 0.80771017,
      "learning_rate": 5.157622785353457e-07,
      "loss": 0.8306458,
      "num_input_tokens_seen": 138368135,
      "step": 6430,
      "time_per_iteration": 2.6584599018096924
    },
    {
      "auxiliary_loss_clip": 0.0112022,
      "auxiliary_loss_mlp": 0.01001513,
      "balance_loss_clip": 1.00989723,
      "balance_loss_mlp": 1.00054729,
      "epoch": 0.7732820296999939,
      "flos": 64201027069440.0,
      "grad_norm": 0.6542652116491539,
      "language_loss": 0.60298318,
      "learning_rate": 5.152402711241113e-07,
      "loss": 0.62420052,
      "num_input_tokens_seen": 138436040,
      "step": 6431,
      "time_per_iteration": 3.2888240814208984
    },
    {
      "auxiliary_loss_clip": 0.01317832,
      "auxiliary_loss_mlp": 0.01025067,
      "balance_loss_clip": 1.04142511,
      "balance_loss_mlp": 1.01832926,
      "epoch": 0.7734022725906331,
      "flos": 25302191984640.0,
      "grad_norm": 1.8409270634393495,
      "language_loss": 0.83216739,
      "learning_rate": 5.147184889497465e-07,
      "loss": 0.85559642,
      "num_input_tokens_seen": 138455510,
      "step": 6432,
      "time_per_iteration": 2.6844303607940674
    },
    {
      "auxiliary_loss_clip": 0.01310625,
      "auxiliary_loss_mlp": 0.01027448,
      "balance_loss_clip": 1.04255605,
      "balance_loss_mlp": 1.0200634,
      "epoch": 0.7735225154812722,
      "flos": 17347440067200.0,
      "grad_norm": 2.666724583133917,
      "language_loss": 0.79889798,
      "learning_rate": 5.141969320914072e-07,
      "loss": 0.82227874,
      "num_input_tokens_seen": 138473015,
      "step": 6433,
      "time_per_iteration": 2.6946356296539307
    },
    {
      "auxiliary_loss_clip": 0.01177975,
      "auxiliary_loss_mlp": 0.01031768,
      "balance_loss_clip": 1.05015993,
      "balance_loss_mlp": 1.02409649,
      "epoch": 0.7736427583719112,
      "flos": 32630086725120.0,
      "grad_norm": 3.472897733931272,
      "language_loss": 0.62583441,
      "learning_rate": 5.136756006282113e-07,
      "loss": 0.64793181,
      "num_input_tokens_seen": 138491680,
      "step": 6434,
      "time_per_iteration": 3.609440803527832
    },
    {
      "auxiliary_loss_clip": 0.01176772,
      "auxiliary_loss_mlp": 0.01026344,
      "balance_loss_clip": 1.05089784,
      "balance_loss_mlp": 1.01919174,
      "epoch": 0.7737630012625504,
      "flos": 19860073269120.0,
      "grad_norm": 3.194374980593014,
      "language_loss": 0.84765595,
      "learning_rate": 5.131544946392446e-07,
      "loss": 0.86968714,
      "num_input_tokens_seen": 138506960,
      "step": 6435,
      "time_per_iteration": 2.6465771198272705
    },
    {
      "auxiliary_loss_clip": 0.01277435,
      "auxiliary_loss_mlp": 0.01035299,
      "balance_loss_clip": 1.05196881,
      "balance_loss_mlp": 1.02768457,
      "epoch": 0.7738832441531894,
      "flos": 36022639397760.0,
      "grad_norm": 2.4628590757949316,
      "language_loss": 0.64373851,
      "learning_rate": 5.126336142035592e-07,
      "loss": 0.66686583,
      "num_input_tokens_seen": 138526995,
      "step": 6436,
      "time_per_iteration": 2.787498950958252
    },
    {
      "auxiliary_loss_clip": 0.01274902,
      "auxiliary_loss_mlp": 0.01028761,
      "balance_loss_clip": 1.04596579,
      "balance_loss_mlp": 1.02168882,
      "epoch": 0.7740034870438285,
      "flos": 13405274415360.0,
      "grad_norm": 4.4384433375599395,
      "language_loss": 0.72224635,
      "learning_rate": 5.121129594001721e-07,
      "loss": 0.74528301,
      "num_input_tokens_seen": 138541260,
      "step": 6437,
      "time_per_iteration": 2.75559401512146
    },
    {
      "auxiliary_loss_clip": 0.01219733,
      "auxiliary_loss_mlp": 0.01029838,
      "balance_loss_clip": 1.04952884,
      "balance_loss_mlp": 1.0228287,
      "epoch": 0.7741237299344677,
      "flos": 22086714384000.0,
      "grad_norm": 1.6150242945474271,
      "language_loss": 0.81413436,
      "learning_rate": 5.115925303080661e-07,
      "loss": 0.83663011,
      "num_input_tokens_seen": 138560970,
      "step": 6438,
      "time_per_iteration": 2.658479928970337
    },
    {
      "auxiliary_loss_clip": 0.01174538,
      "auxiliary_loss_mlp": 0.01021412,
      "balance_loss_clip": 1.04801238,
      "balance_loss_mlp": 1.0150938,
      "epoch": 0.7742439728251067,
      "flos": 19864777950720.0,
      "grad_norm": 1.9785420487531769,
      "language_loss": 0.79639161,
      "learning_rate": 5.110723270061899e-07,
      "loss": 0.81835109,
      "num_input_tokens_seen": 138577460,
      "step": 6439,
      "time_per_iteration": 2.623114824295044
    },
    {
      "auxiliary_loss_clip": 0.01172746,
      "auxiliary_loss_mlp": 0.01023642,
      "balance_loss_clip": 1.04965138,
      "balance_loss_mlp": 1.01706171,
      "epoch": 0.7743642157157458,
      "flos": 16690167048960.0,
      "grad_norm": 3.277821782427305,
      "language_loss": 0.79557353,
      "learning_rate": 5.105523495734572e-07,
      "loss": 0.81753743,
      "num_input_tokens_seen": 138594860,
      "step": 6440,
      "time_per_iteration": 2.59525728225708
    },
    {
      "auxiliary_loss_clip": 0.01176431,
      "auxiliary_loss_mlp": 0.01023795,
      "balance_loss_clip": 1.04912162,
      "balance_loss_mlp": 1.01681495,
      "epoch": 0.7744844586063849,
      "flos": 20304360593280.0,
      "grad_norm": 1.990722522488836,
      "language_loss": 0.7540406,
      "learning_rate": 5.100325980887499e-07,
      "loss": 0.77604282,
      "num_input_tokens_seen": 138614785,
      "step": 6441,
      "time_per_iteration": 2.6053881645202637
    },
    {
      "auxiliary_loss_clip": 0.01181241,
      "auxiliary_loss_mlp": 0.01027218,
      "balance_loss_clip": 1.04847503,
      "balance_loss_mlp": 1.02050984,
      "epoch": 0.774604701497024,
      "flos": 22966705681920.0,
      "grad_norm": 2.056177005806305,
      "language_loss": 0.83324683,
      "learning_rate": 5.095130726309116e-07,
      "loss": 0.85533142,
      "num_input_tokens_seen": 138634960,
      "step": 6442,
      "time_per_iteration": 2.6991353034973145
    },
    {
      "auxiliary_loss_clip": 0.0106446,
      "auxiliary_loss_mlp": 0.01003224,
      "balance_loss_clip": 1.00902939,
      "balance_loss_mlp": 1.0022881,
      "epoch": 0.774724944387663,
      "flos": 60288523073280.0,
      "grad_norm": 0.7975465898478871,
      "language_loss": 0.59002739,
      "learning_rate": 5.089937732787559e-07,
      "loss": 0.61070424,
      "num_input_tokens_seen": 138699520,
      "step": 6443,
      "time_per_iteration": 3.203735113143921
    },
    {
      "auxiliary_loss_clip": 0.01322855,
      "auxiliary_loss_mlp": 0.01024348,
      "balance_loss_clip": 1.04358697,
      "balance_loss_mlp": 1.01702809,
      "epoch": 0.7748451872783022,
      "flos": 26761026954240.0,
      "grad_norm": 2.29623264409304,
      "language_loss": 0.67012691,
      "learning_rate": 5.084747001110592e-07,
      "loss": 0.69359899,
      "num_input_tokens_seen": 138719145,
      "step": 6444,
      "time_per_iteration": 2.753668785095215
    },
    {
      "auxiliary_loss_clip": 0.01227373,
      "auxiliary_loss_mlp": 0.02564501,
      "balance_loss_clip": 1.05399227,
      "balance_loss_mlp": 0.99989426,
      "epoch": 0.7749654301689413,
      "flos": 30338627518080.0,
      "grad_norm": 2.3876572169900405,
      "language_loss": 0.70295572,
      "learning_rate": 5.07955853206564e-07,
      "loss": 0.74087447,
      "num_input_tokens_seen": 138743850,
      "step": 6445,
      "time_per_iteration": 2.717557668685913
    },
    {
      "auxiliary_loss_clip": 0.01227576,
      "auxiliary_loss_mlp": 0.01024068,
      "balance_loss_clip": 1.04928303,
      "balance_loss_mlp": 1.01692724,
      "epoch": 0.7750856730595803,
      "flos": 43179851687040.0,
      "grad_norm": 2.0764614967293795,
      "language_loss": 0.7101016,
      "learning_rate": 5.074372326439807e-07,
      "loss": 0.73261803,
      "num_input_tokens_seen": 138766860,
      "step": 6446,
      "time_per_iteration": 2.799731731414795
    },
    {
      "auxiliary_loss_clip": 0.01319585,
      "auxiliary_loss_mlp": 0.01020654,
      "balance_loss_clip": 1.04331064,
      "balance_loss_mlp": 1.01301837,
      "epoch": 0.7752059159502195,
      "flos": 17640040256640.0,
      "grad_norm": 4.158024971019772,
      "language_loss": 0.73843294,
      "learning_rate": 5.069188385019814e-07,
      "loss": 0.76183534,
      "num_input_tokens_seen": 138784560,
      "step": 6447,
      "time_per_iteration": 2.6515884399414062
    },
    {
      "auxiliary_loss_clip": 0.01377742,
      "auxiliary_loss_mlp": 0.01031319,
      "balance_loss_clip": 1.04215133,
      "balance_loss_mlp": 1.02419019,
      "epoch": 0.7753261588408585,
      "flos": 12677688524160.0,
      "grad_norm": 3.1845236178073075,
      "language_loss": 0.60891461,
      "learning_rate": 5.064006708592077e-07,
      "loss": 0.63300514,
      "num_input_tokens_seen": 138800805,
      "step": 6448,
      "time_per_iteration": 2.760885715484619
    },
    {
      "auxiliary_loss_clip": 0.01263172,
      "auxiliary_loss_mlp": 0.01022989,
      "balance_loss_clip": 1.04614997,
      "balance_loss_mlp": 1.01647758,
      "epoch": 0.7754464017314976,
      "flos": 16690741666560.0,
      "grad_norm": 2.430598373594709,
      "language_loss": 0.75401688,
      "learning_rate": 5.058827297942641e-07,
      "loss": 0.77687848,
      "num_input_tokens_seen": 138815910,
      "step": 6449,
      "time_per_iteration": 2.6097042560577393
    },
    {
      "auxiliary_loss_clip": 0.01179606,
      "auxiliary_loss_mlp": 0.01027068,
      "balance_loss_clip": 1.0480051,
      "balance_loss_mlp": 1.02015662,
      "epoch": 0.7755666446221368,
      "flos": 19718944732800.0,
      "grad_norm": 1.9902076202986638,
      "language_loss": 0.75137115,
      "learning_rate": 5.053650153857237e-07,
      "loss": 0.77343786,
      "num_input_tokens_seen": 138834920,
      "step": 6450,
      "time_per_iteration": 3.5692994594573975
    },
    {
      "auxiliary_loss_clip": 0.0121995,
      "auxiliary_loss_mlp": 0.01024257,
      "balance_loss_clip": 1.04793429,
      "balance_loss_mlp": 1.01755738,
      "epoch": 0.7756868875127758,
      "flos": 18693623007360.0,
      "grad_norm": 3.196147529921042,
      "language_loss": 0.69734943,
      "learning_rate": 5.048475277121214e-07,
      "loss": 0.71979153,
      "num_input_tokens_seen": 138852135,
      "step": 6451,
      "time_per_iteration": 3.5589799880981445
    },
    {
      "auxiliary_loss_clip": 0.01223568,
      "auxiliary_loss_mlp": 0.01024878,
      "balance_loss_clip": 1.04815912,
      "balance_loss_mlp": 1.01786852,
      "epoch": 0.7758071304034149,
      "flos": 28404191543040.0,
      "grad_norm": 2.070193694666023,
      "language_loss": 0.77256083,
      "learning_rate": 5.043302668519598e-07,
      "loss": 0.79504532,
      "num_input_tokens_seen": 138871470,
      "step": 6452,
      "time_per_iteration": 2.707768201828003
    },
    {
      "auxiliary_loss_clip": 0.01226123,
      "auxiliary_loss_mlp": 0.0102831,
      "balance_loss_clip": 1.04702163,
      "balance_loss_mlp": 1.0213629,
      "epoch": 0.775927373294054,
      "flos": 20595344670720.0,
      "grad_norm": 1.8049542897624458,
      "language_loss": 0.72059834,
      "learning_rate": 5.038132328837079e-07,
      "loss": 0.74314266,
      "num_input_tokens_seen": 138889860,
      "step": 6453,
      "time_per_iteration": 3.634740114212036
    },
    {
      "auxiliary_loss_clip": 0.01224745,
      "auxiliary_loss_mlp": 0.0102546,
      "balance_loss_clip": 1.04889369,
      "balance_loss_mlp": 1.01871872,
      "epoch": 0.7760476161846931,
      "flos": 22526368853760.0,
      "grad_norm": 2.75350975775893,
      "language_loss": 0.74589634,
      "learning_rate": 5.032964258857993e-07,
      "loss": 0.7683984,
      "num_input_tokens_seen": 138909955,
      "step": 6454,
      "time_per_iteration": 2.6494932174682617
    },
    {
      "auxiliary_loss_clip": 0.01221299,
      "auxiliary_loss_mlp": 0.01030457,
      "balance_loss_clip": 1.04453373,
      "balance_loss_mlp": 1.02312887,
      "epoch": 0.7761678590753321,
      "flos": 48651488403840.0,
      "grad_norm": 1.6042215221610898,
      "language_loss": 0.68235731,
      "learning_rate": 5.027798459366329e-07,
      "loss": 0.70487487,
      "num_input_tokens_seen": 138935320,
      "step": 6455,
      "time_per_iteration": 2.85683274269104
    },
    {
      "auxiliary_loss_clip": 0.01227883,
      "auxiliary_loss_mlp": 0.0102594,
      "balance_loss_clip": 1.04809773,
      "balance_loss_mlp": 1.01882946,
      "epoch": 0.7762881019659713,
      "flos": 26177047637760.0,
      "grad_norm": 2.7670064408032955,
      "language_loss": 0.63608736,
      "learning_rate": 5.02263493114573e-07,
      "loss": 0.6586256,
      "num_input_tokens_seen": 138957115,
      "step": 6456,
      "time_per_iteration": 2.6154611110687256
    },
    {
      "auxiliary_loss_clip": 0.01174669,
      "auxiliary_loss_mlp": 0.01025748,
      "balance_loss_clip": 1.04982972,
      "balance_loss_mlp": 1.01900673,
      "epoch": 0.7764083448566104,
      "flos": 20588341518720.0,
      "grad_norm": 2.4522007970313338,
      "language_loss": 0.7663309,
      "learning_rate": 5.017473674979502e-07,
      "loss": 0.78833508,
      "num_input_tokens_seen": 138973140,
      "step": 6457,
      "time_per_iteration": 2.5260229110717773
    },
    {
      "auxiliary_loss_clip": 0.01215865,
      "auxiliary_loss_mlp": 0.01003852,
      "balance_loss_clip": 1.01145029,
      "balance_loss_mlp": 1.00294614,
      "epoch": 0.7765285877472494,
      "flos": 67293078560640.0,
      "grad_norm": 0.7428098261801143,
      "language_loss": 0.58259106,
      "learning_rate": 5.01231469165061e-07,
      "loss": 0.60478818,
      "num_input_tokens_seen": 139028965,
      "step": 6458,
      "time_per_iteration": 3.1412134170532227
    },
    {
      "auxiliary_loss_clip": 0.01118528,
      "auxiliary_loss_mlp": 0.01001337,
      "balance_loss_clip": 1.00945067,
      "balance_loss_mlp": 1.0003953,
      "epoch": 0.7766488306378886,
      "flos": 61344476121600.0,
      "grad_norm": 0.8245731656388763,
      "language_loss": 0.56806552,
      "learning_rate": 5.007157981941663e-07,
      "loss": 0.58926415,
      "num_input_tokens_seen": 139094325,
      "step": 6459,
      "time_per_iteration": 4.163167953491211
    },
    {
      "auxiliary_loss_clip": 0.01173262,
      "auxiliary_loss_mlp": 0.00999935,
      "balance_loss_clip": 1.00989461,
      "balance_loss_mlp": 0.99898165,
      "epoch": 0.7767690735285276,
      "flos": 62946199393920.0,
      "grad_norm": 0.8799733442717024,
      "language_loss": 0.67400622,
      "learning_rate": 5.002003546634928e-07,
      "loss": 0.6957382,
      "num_input_tokens_seen": 139150425,
      "step": 6460,
      "time_per_iteration": 3.159785270690918
    },
    {
      "auxiliary_loss_clip": 0.01375949,
      "auxiliary_loss_mlp": 0.01031197,
      "balance_loss_clip": 1.04915714,
      "balance_loss_mlp": 1.02475989,
      "epoch": 0.7768893164191667,
      "flos": 20886400575360.0,
      "grad_norm": 1.607777177927785,
      "language_loss": 0.76212275,
      "learning_rate": 4.996851386512331e-07,
      "loss": 0.78619421,
      "num_input_tokens_seen": 139169130,
      "step": 6461,
      "time_per_iteration": 2.8033576011657715
    },
    {
      "auxiliary_loss_clip": 0.01270171,
      "auxiliary_loss_mlp": 0.01023566,
      "balance_loss_clip": 1.0474782,
      "balance_loss_mlp": 1.01610947,
      "epoch": 0.7770095593098058,
      "flos": 20704584908160.0,
      "grad_norm": 1.7376615256629229,
      "language_loss": 0.83223486,
      "learning_rate": 4.991701502355444e-07,
      "loss": 0.85517216,
      "num_input_tokens_seen": 139189595,
      "step": 6462,
      "time_per_iteration": 2.6751768589019775
    },
    {
      "auxiliary_loss_clip": 0.01224389,
      "auxiliary_loss_mlp": 0.01020236,
      "balance_loss_clip": 1.04592741,
      "balance_loss_mlp": 1.01327443,
      "epoch": 0.7771298022004449,
      "flos": 24717709877760.0,
      "grad_norm": 1.498019520588296,
      "language_loss": 0.75955403,
      "learning_rate": 4.986553894945518e-07,
      "loss": 0.78200024,
      "num_input_tokens_seen": 139210805,
      "step": 6463,
      "time_per_iteration": 2.6463098526000977
    },
    {
      "auxiliary_loss_clip": 0.01272312,
      "auxiliary_loss_mlp": 0.01019195,
      "balance_loss_clip": 1.04311156,
      "balance_loss_mlp": 1.01293707,
      "epoch": 0.777250045091084,
      "flos": 25009232659200.0,
      "grad_norm": 2.0974022312506473,
      "language_loss": 0.86706406,
      "learning_rate": 4.981408565063416e-07,
      "loss": 0.88997912,
      "num_input_tokens_seen": 139230750,
      "step": 6464,
      "time_per_iteration": 2.779273748397827
    },
    {
      "auxiliary_loss_clip": 0.01177203,
      "auxiliary_loss_mlp": 0.01028947,
      "balance_loss_clip": 1.05059814,
      "balance_loss_mlp": 1.0221132,
      "epoch": 0.777370287981723,
      "flos": 20119887319680.0,
      "grad_norm": 2.3149598159024714,
      "language_loss": 0.75997168,
      "learning_rate": 4.976265513489701e-07,
      "loss": 0.78203309,
      "num_input_tokens_seen": 139250720,
      "step": 6465,
      "time_per_iteration": 2.5846400260925293
    },
    {
      "auxiliary_loss_clip": 0.01227199,
      "auxiliary_loss_mlp": 0.01032455,
      "balance_loss_clip": 1.04867876,
      "balance_loss_mlp": 1.02552629,
      "epoch": 0.7774905308723622,
      "flos": 21718809331200.0,
      "grad_norm": 1.9482149599151688,
      "language_loss": 0.80549604,
      "learning_rate": 4.971124741004562e-07,
      "loss": 0.82809258,
      "num_input_tokens_seen": 139269720,
      "step": 6466,
      "time_per_iteration": 2.617812156677246
    },
    {
      "auxiliary_loss_clip": 0.01223409,
      "auxiliary_loss_mlp": 0.01025549,
      "balance_loss_clip": 1.04788828,
      "balance_loss_mlp": 1.01880491,
      "epoch": 0.7776107737630013,
      "flos": 16034115093120.0,
      "grad_norm": 2.2892643036339555,
      "language_loss": 0.76374394,
      "learning_rate": 4.965986248387846e-07,
      "loss": 0.78623354,
      "num_input_tokens_seen": 139288035,
      "step": 6467,
      "time_per_iteration": 2.588573932647705
    },
    {
      "auxiliary_loss_clip": 0.0127712,
      "auxiliary_loss_mlp": 0.01019828,
      "balance_loss_clip": 1.04688787,
      "balance_loss_mlp": 1.01327467,
      "epoch": 0.7777310166536403,
      "flos": 24790895838720.0,
      "grad_norm": 2.2221988408818687,
      "language_loss": 0.77285337,
      "learning_rate": 4.960850036419073e-07,
      "loss": 0.79582286,
      "num_input_tokens_seen": 139307135,
      "step": 6468,
      "time_per_iteration": 2.690772771835327
    },
    {
      "auxiliary_loss_clip": 0.0126847,
      "auxiliary_loss_mlp": 0.01021819,
      "balance_loss_clip": 1.04723501,
      "balance_loss_mlp": 1.01479709,
      "epoch": 0.7778512595442795,
      "flos": 17272530253440.0,
      "grad_norm": 2.3001990075796535,
      "language_loss": 0.78836775,
      "learning_rate": 4.955716105877378e-07,
      "loss": 0.81127059,
      "num_input_tokens_seen": 139325905,
      "step": 6469,
      "time_per_iteration": 2.6428310871124268
    },
    {
      "auxiliary_loss_clip": 0.01226844,
      "auxiliary_loss_mlp": 0.02567025,
      "balance_loss_clip": 1.04805231,
      "balance_loss_mlp": 0.99992454,
      "epoch": 0.7779715024349185,
      "flos": 17748418567680.0,
      "grad_norm": 3.07632692434293,
      "language_loss": 0.83201629,
      "learning_rate": 4.950584457541598e-07,
      "loss": 0.86995494,
      "num_input_tokens_seen": 139344370,
      "step": 6470,
      "time_per_iteration": 2.635831594467163
    },
    {
      "auxiliary_loss_clip": 0.01228064,
      "auxiliary_loss_mlp": 0.01026219,
      "balance_loss_clip": 1.04878712,
      "balance_loss_mlp": 1.01933169,
      "epoch": 0.7780917453255576,
      "flos": 24316875031680.0,
      "grad_norm": 1.4033947656505945,
      "language_loss": 0.81982315,
      "learning_rate": 4.945455092190183e-07,
      "loss": 0.84236598,
      "num_input_tokens_seen": 139365625,
      "step": 6471,
      "time_per_iteration": 2.66037917137146
    },
    {
      "auxiliary_loss_clip": 0.01064568,
      "auxiliary_loss_mlp": 0.01001575,
      "balance_loss_clip": 1.00917745,
      "balance_loss_mlp": 1.00063956,
      "epoch": 0.7782119882161967,
      "flos": 56364601530240.0,
      "grad_norm": 0.6876914192085256,
      "language_loss": 0.5594691,
      "learning_rate": 4.940328010601271e-07,
      "loss": 0.58013046,
      "num_input_tokens_seen": 139430540,
      "step": 6472,
      "time_per_iteration": 3.19032883644104
    },
    {
      "auxiliary_loss_clip": 0.01181722,
      "auxiliary_loss_mlp": 0.01030548,
      "balance_loss_clip": 1.05313277,
      "balance_loss_mlp": 1.02341962,
      "epoch": 0.7783322311068358,
      "flos": 46789986994560.0,
      "grad_norm": 1.8945726642786342,
      "language_loss": 0.76822388,
      "learning_rate": 4.935203213552621e-07,
      "loss": 0.79034662,
      "num_input_tokens_seen": 139454280,
      "step": 6473,
      "time_per_iteration": 2.893869638442993
    },
    {
      "auxiliary_loss_clip": 0.01282066,
      "auxiliary_loss_mlp": 0.01021276,
      "balance_loss_clip": 1.05097568,
      "balance_loss_mlp": 1.01409078,
      "epoch": 0.7784524739974749,
      "flos": 19057864872960.0,
      "grad_norm": 2.414888850228351,
      "language_loss": 0.6688379,
      "learning_rate": 4.930080701821662e-07,
      "loss": 0.69187135,
      "num_input_tokens_seen": 139471745,
      "step": 6474,
      "time_per_iteration": 2.623652696609497
    },
    {
      "auxiliary_loss_clip": 0.01272031,
      "auxiliary_loss_mlp": 0.01025042,
      "balance_loss_clip": 1.04503214,
      "balance_loss_mlp": 1.01811266,
      "epoch": 0.778572716888114,
      "flos": 24791111320320.0,
      "grad_norm": 1.9575521625263195,
      "language_loss": 0.77844357,
      "learning_rate": 4.92496047618548e-07,
      "loss": 0.80141425,
      "num_input_tokens_seen": 139491505,
      "step": 6475,
      "time_per_iteration": 2.69826340675354
    },
    {
      "auxiliary_loss_clip": 0.01221216,
      "auxiliary_loss_mlp": 0.01020587,
      "balance_loss_clip": 1.0495702,
      "balance_loss_mlp": 1.01364017,
      "epoch": 0.7786929597787531,
      "flos": 20078086867200.0,
      "grad_norm": 2.044372343355825,
      "language_loss": 0.77946788,
      "learning_rate": 4.919842537420811e-07,
      "loss": 0.80188584,
      "num_input_tokens_seen": 139508620,
      "step": 6476,
      "time_per_iteration": 3.5619394779205322
    },
    {
      "auxiliary_loss_clip": 0.0127607,
      "auxiliary_loss_mlp": 0.01025712,
      "balance_loss_clip": 1.05034077,
      "balance_loss_mlp": 1.01866388,
      "epoch": 0.7788132026693921,
      "flos": 21872220318720.0,
      "grad_norm": 1.7180282701948317,
      "language_loss": 0.79426229,
      "learning_rate": 4.91472688630404e-07,
      "loss": 0.81728005,
      "num_input_tokens_seen": 139529360,
      "step": 6477,
      "time_per_iteration": 3.631579637527466
    },
    {
      "auxiliary_loss_clip": 0.01173111,
      "auxiliary_loss_mlp": 0.0102628,
      "balance_loss_clip": 1.04987586,
      "balance_loss_mlp": 1.01975965,
      "epoch": 0.7789334455600313,
      "flos": 11181937351680.0,
      "grad_norm": 1.833003560173933,
      "language_loss": 0.73914224,
      "learning_rate": 4.909613523611202e-07,
      "loss": 0.76113617,
      "num_input_tokens_seen": 139546240,
      "step": 6478,
      "time_per_iteration": 2.588418960571289
    },
    {
      "auxiliary_loss_clip": 0.01272005,
      "auxiliary_loss_mlp": 0.02568414,
      "balance_loss_clip": 1.04193997,
      "balance_loss_mlp": 0.99991769,
      "epoch": 0.7790536884506704,
      "flos": 28695427015680.0,
      "grad_norm": 1.8714568499781241,
      "language_loss": 0.7508505,
      "learning_rate": 4.904502450117991e-07,
      "loss": 0.78925467,
      "num_input_tokens_seen": 139567200,
      "step": 6479,
      "time_per_iteration": 3.651624917984009
    },
    {
      "auxiliary_loss_clip": 0.01276015,
      "auxiliary_loss_mlp": 0.01029246,
      "balance_loss_clip": 1.05138147,
      "balance_loss_mlp": 1.02160144,
      "epoch": 0.7791739313413094,
      "flos": 11072302064640.0,
      "grad_norm": 2.5817427306238776,
      "language_loss": 0.71887398,
      "learning_rate": 4.899393666599762e-07,
      "loss": 0.74192655,
      "num_input_tokens_seen": 139583775,
      "step": 6480,
      "time_per_iteration": 2.6607015132904053
    },
    {
      "auxiliary_loss_clip": 0.01172722,
      "auxiliary_loss_mlp": 0.01023133,
      "balance_loss_clip": 1.04736018,
      "balance_loss_mlp": 1.01621032,
      "epoch": 0.7792941742319486,
      "flos": 14679276975360.0,
      "grad_norm": 2.484903616691076,
      "language_loss": 0.72778475,
      "learning_rate": 4.894287173831506e-07,
      "loss": 0.74974328,
      "num_input_tokens_seen": 139599735,
      "step": 6481,
      "time_per_iteration": 2.552361011505127
    },
    {
      "auxiliary_loss_clip": 0.01274017,
      "auxiliary_loss_mlp": 0.01021576,
      "balance_loss_clip": 1.04490197,
      "balance_loss_mlp": 1.01431024,
      "epoch": 0.7794144171225876,
      "flos": 23258874908160.0,
      "grad_norm": 2.6493626537712758,
      "language_loss": 0.84431577,
      "learning_rate": 4.889182972587877e-07,
      "loss": 0.86727166,
      "num_input_tokens_seen": 139619030,
      "step": 6482,
      "time_per_iteration": 2.695523262023926
    },
    {
      "auxiliary_loss_clip": 0.01331952,
      "auxiliary_loss_mlp": 0.01027917,
      "balance_loss_clip": 1.04743624,
      "balance_loss_mlp": 1.02106202,
      "epoch": 0.7795346600132267,
      "flos": 21507080613120.0,
      "grad_norm": 1.959656193006223,
      "language_loss": 0.66469014,
      "learning_rate": 4.884081063643177e-07,
      "loss": 0.68828887,
      "num_input_tokens_seen": 139637690,
      "step": 6483,
      "time_per_iteration": 2.6938936710357666
    },
    {
      "auxiliary_loss_clip": 0.01163048,
      "auxiliary_loss_mlp": 0.01007602,
      "balance_loss_clip": 1.0092603,
      "balance_loss_mlp": 1.00673771,
      "epoch": 0.7796549029038659,
      "flos": 70052273694720.0,
      "grad_norm": 0.8777529746337558,
      "language_loss": 0.52499074,
      "learning_rate": 4.878981447771353e-07,
      "loss": 0.54669726,
      "num_input_tokens_seen": 139692070,
      "step": 6484,
      "time_per_iteration": 3.2099342346191406
    },
    {
      "auxiliary_loss_clip": 0.01319965,
      "auxiliary_loss_mlp": 0.01026008,
      "balance_loss_clip": 1.04691708,
      "balance_loss_mlp": 1.01875663,
      "epoch": 0.7797751457945049,
      "flos": 23989405714560.0,
      "grad_norm": 2.230046136607681,
      "language_loss": 0.73240423,
      "learning_rate": 4.873884125746035e-07,
      "loss": 0.7558639,
      "num_input_tokens_seen": 139713745,
      "step": 6485,
      "time_per_iteration": 2.7443389892578125
    },
    {
      "auxiliary_loss_clip": 0.01272953,
      "auxiliary_loss_mlp": 0.01023572,
      "balance_loss_clip": 1.04679275,
      "balance_loss_mlp": 1.01681602,
      "epoch": 0.779895388685144,
      "flos": 22674751937280.0,
      "grad_norm": 2.7412080053329917,
      "language_loss": 0.72702354,
      "learning_rate": 4.868789098340456e-07,
      "loss": 0.74998879,
      "num_input_tokens_seen": 139731650,
      "step": 6486,
      "time_per_iteration": 3.5360422134399414
    },
    {
      "auxiliary_loss_clip": 0.0132129,
      "auxiliary_loss_mlp": 0.0103125,
      "balance_loss_clip": 1.04552531,
      "balance_loss_mlp": 1.02444351,
      "epoch": 0.7800156315757831,
      "flos": 23768698596480.0,
      "grad_norm": 2.3110555966957205,
      "language_loss": 0.73463035,
      "learning_rate": 4.863696366327543e-07,
      "loss": 0.75815576,
      "num_input_tokens_seen": 139750820,
      "step": 6487,
      "time_per_iteration": 2.755352020263672
    },
    {
      "auxiliary_loss_clip": 0.01225529,
      "auxiliary_loss_mlp": 0.0102301,
      "balance_loss_clip": 1.04662442,
      "balance_loss_mlp": 1.01594687,
      "epoch": 0.7801358744664222,
      "flos": 26429714881920.0,
      "grad_norm": 1.8805190263244949,
      "language_loss": 0.78058797,
      "learning_rate": 4.85860593047986e-07,
      "loss": 0.80307335,
      "num_input_tokens_seen": 139770885,
      "step": 6488,
      "time_per_iteration": 2.669504404067993
    },
    {
      "auxiliary_loss_clip": 0.01315997,
      "auxiliary_loss_mlp": 0.0102644,
      "balance_loss_clip": 1.03969193,
      "balance_loss_mlp": 1.01981199,
      "epoch": 0.7802561173570612,
      "flos": 26322162583680.0,
      "grad_norm": 1.703508273156949,
      "language_loss": 0.74509203,
      "learning_rate": 4.853517791569613e-07,
      "loss": 0.76851642,
      "num_input_tokens_seen": 139793065,
      "step": 6489,
      "time_per_iteration": 2.7530031204223633
    },
    {
      "auxiliary_loss_clip": 0.01276551,
      "auxiliary_loss_mlp": 0.02567899,
      "balance_loss_clip": 1.04543793,
      "balance_loss_mlp": 0.99991739,
      "epoch": 0.7803763602477004,
      "flos": 40333751596800.0,
      "grad_norm": 6.874110933277564,
      "language_loss": 0.66368943,
      "learning_rate": 4.848431950368684e-07,
      "loss": 0.70213401,
      "num_input_tokens_seen": 139815625,
      "step": 6490,
      "time_per_iteration": 2.8528971672058105
    },
    {
      "auxiliary_loss_clip": 0.01063894,
      "auxiliary_loss_mlp": 0.02505062,
      "balance_loss_clip": 1.00878537,
      "balance_loss_mlp": 0.99988699,
      "epoch": 0.7804966031383395,
      "flos": 67001448038400.0,
      "grad_norm": 0.7021159589794077,
      "language_loss": 0.55662942,
      "learning_rate": 4.843348407648569e-07,
      "loss": 0.59231901,
      "num_input_tokens_seen": 139876905,
      "step": 6491,
      "time_per_iteration": 3.1406171321868896
    },
    {
      "auxiliary_loss_clip": 0.01226358,
      "auxiliary_loss_mlp": 0.01024498,
      "balance_loss_clip": 1.04472244,
      "balance_loss_mlp": 1.01737499,
      "epoch": 0.7806168460289785,
      "flos": 17740733057280.0,
      "grad_norm": 2.998797796599976,
      "language_loss": 0.83317745,
      "learning_rate": 4.838267164180457e-07,
      "loss": 0.85568607,
      "num_input_tokens_seen": 139892575,
      "step": 6492,
      "time_per_iteration": 2.5639567375183105
    },
    {
      "auxiliary_loss_clip": 0.0117687,
      "auxiliary_loss_mlp": 0.01024711,
      "balance_loss_clip": 1.05033064,
      "balance_loss_mlp": 1.0173409,
      "epoch": 0.7807370889196176,
      "flos": 23946240545280.0,
      "grad_norm": 2.415225329714158,
      "language_loss": 0.8331672,
      "learning_rate": 4.833188220735156e-07,
      "loss": 0.85518295,
      "num_input_tokens_seen": 139912245,
      "step": 6493,
      "time_per_iteration": 2.5723211765289307
    },
    {
      "auxiliary_loss_clip": 0.01220381,
      "auxiliary_loss_mlp": 0.0102227,
      "balance_loss_clip": 1.04702878,
      "balance_loss_mlp": 1.01501,
      "epoch": 0.7808573318102567,
      "flos": 18989024457600.0,
      "grad_norm": 2.081089953671375,
      "language_loss": 0.75005138,
      "learning_rate": 4.828111578083152e-07,
      "loss": 0.77247787,
      "num_input_tokens_seen": 139929150,
      "step": 6494,
      "time_per_iteration": 2.5768826007843018
    },
    {
      "auxiliary_loss_clip": 0.0127155,
      "auxiliary_loss_mlp": 0.01027003,
      "balance_loss_clip": 1.0489645,
      "balance_loss_mlp": 1.02012122,
      "epoch": 0.7809775747008958,
      "flos": 23980750536960.0,
      "grad_norm": 2.9985034761231324,
      "language_loss": 0.81150115,
      "learning_rate": 4.823037236994556e-07,
      "loss": 0.83448672,
      "num_input_tokens_seen": 139947315,
      "step": 6495,
      "time_per_iteration": 2.6143763065338135
    },
    {
      "auxiliary_loss_clip": 0.01119839,
      "auxiliary_loss_mlp": 0.01001164,
      "balance_loss_clip": 1.00918424,
      "balance_loss_mlp": 1.00021636,
      "epoch": 0.7810978175915348,
      "flos": 68535875180160.0,
      "grad_norm": 0.7195142640987876,
      "language_loss": 0.56294608,
      "learning_rate": 4.817965198239136e-07,
      "loss": 0.58415616,
      "num_input_tokens_seen": 140013775,
      "step": 6496,
      "time_per_iteration": 3.165374994277954
    },
    {
      "auxiliary_loss_clip": 0.01323868,
      "auxiliary_loss_mlp": 0.01028105,
      "balance_loss_clip": 1.04354525,
      "balance_loss_mlp": 1.02057099,
      "epoch": 0.781218060482174,
      "flos": 19642131498240.0,
      "grad_norm": 2.115374394406984,
      "language_loss": 0.74555838,
      "learning_rate": 4.812895462586331e-07,
      "loss": 0.76907814,
      "num_input_tokens_seen": 140031600,
      "step": 6497,
      "time_per_iteration": 2.738694667816162
    },
    {
      "auxiliary_loss_clip": 0.01323168,
      "auxiliary_loss_mlp": 0.01022391,
      "balance_loss_clip": 1.0465889,
      "balance_loss_mlp": 1.01580167,
      "epoch": 0.7813383033728131,
      "flos": 25627865621760.0,
      "grad_norm": 1.8215165768130004,
      "language_loss": 0.8196103,
      "learning_rate": 4.807828030805207e-07,
      "loss": 0.84306592,
      "num_input_tokens_seen": 140050590,
      "step": 6498,
      "time_per_iteration": 2.7651188373565674
    },
    {
      "auxiliary_loss_clip": 0.01225958,
      "auxiliary_loss_mlp": 0.010299,
      "balance_loss_clip": 1.05024862,
      "balance_loss_mlp": 1.02311981,
      "epoch": 0.7814585462634521,
      "flos": 20485924865280.0,
      "grad_norm": 1.9320889928785228,
      "language_loss": 0.67888898,
      "learning_rate": 4.802762903664495e-07,
      "loss": 0.70144749,
      "num_input_tokens_seen": 140069770,
      "step": 6499,
      "time_per_iteration": 2.6465797424316406
    },
    {
      "auxiliary_loss_clip": 0.01279663,
      "auxiliary_loss_mlp": 0.01031456,
      "balance_loss_clip": 1.05115604,
      "balance_loss_mlp": 1.02409482,
      "epoch": 0.7815787891540913,
      "flos": 22304297018880.0,
      "grad_norm": 2.2583347341653206,
      "language_loss": 0.73552895,
      "learning_rate": 4.797700081932565e-07,
      "loss": 0.75864017,
      "num_input_tokens_seen": 140087635,
      "step": 6500,
      "time_per_iteration": 2.6871252059936523
    },
    {
      "auxiliary_loss_clip": 0.01414127,
      "auxiliary_loss_mlp": 0.01024132,
      "balance_loss_clip": 1.03847969,
      "balance_loss_mlp": 1.01752234,
      "epoch": 0.7816990320447303,
      "flos": 22600668136320.0,
      "grad_norm": 2.9132445340393964,
      "language_loss": 0.81931704,
      "learning_rate": 4.792639566377442e-07,
      "loss": 0.84369969,
      "num_input_tokens_seen": 140105045,
      "step": 6501,
      "time_per_iteration": 2.7843241691589355
    },
    {
      "auxiliary_loss_clip": 0.01222279,
      "auxiliary_loss_mlp": 0.01035445,
      "balance_loss_clip": 1.04581738,
      "balance_loss_mlp": 1.02821159,
      "epoch": 0.7818192749353694,
      "flos": 24935974871040.0,
      "grad_norm": 1.9153667362298525,
      "language_loss": 0.7765044,
      "learning_rate": 4.78758135776681e-07,
      "loss": 0.79908168,
      "num_input_tokens_seen": 140124900,
      "step": 6502,
      "time_per_iteration": 3.611147165298462
    },
    {
      "auxiliary_loss_clip": 0.01272259,
      "auxiliary_loss_mlp": 0.01025846,
      "balance_loss_clip": 1.04670751,
      "balance_loss_mlp": 1.01920331,
      "epoch": 0.7819395178260086,
      "flos": 23733039369600.0,
      "grad_norm": 3.788532105645625,
      "language_loss": 0.78860915,
      "learning_rate": 4.782525456867989e-07,
      "loss": 0.81159019,
      "num_input_tokens_seen": 140143755,
      "step": 6503,
      "time_per_iteration": 3.6410610675811768
    },
    {
      "auxiliary_loss_clip": 0.01225239,
      "auxiliary_loss_mlp": 0.01027033,
      "balance_loss_clip": 1.04748225,
      "balance_loss_mlp": 1.01946592,
      "epoch": 0.7820597607166476,
      "flos": 23221671396480.0,
      "grad_norm": 2.9099280109282897,
      "language_loss": 0.83213824,
      "learning_rate": 4.777471864447959e-07,
      "loss": 0.85466099,
      "num_input_tokens_seen": 140164495,
      "step": 6504,
      "time_per_iteration": 2.6986584663391113
    },
    {
      "auxiliary_loss_clip": 0.01275424,
      "auxiliary_loss_mlp": 0.01026083,
      "balance_loss_clip": 1.0448519,
      "balance_loss_mlp": 1.01878381,
      "epoch": 0.7821800036072867,
      "flos": 22309540404480.0,
      "grad_norm": 1.9411003608144712,
      "language_loss": 0.8076731,
      "learning_rate": 4.772420581273344e-07,
      "loss": 0.83068812,
      "num_input_tokens_seen": 140181980,
      "step": 6505,
      "time_per_iteration": 2.6299822330474854
    },
    {
      "auxiliary_loss_clip": 0.01221283,
      "auxiliary_loss_mlp": 0.01024458,
      "balance_loss_clip": 1.04955554,
      "balance_loss_mlp": 1.01750255,
      "epoch": 0.7823002464979258,
      "flos": 21544176384000.0,
      "grad_norm": 2.790792270535227,
      "language_loss": 0.76814222,
      "learning_rate": 4.7673716081104134e-07,
      "loss": 0.79059964,
      "num_input_tokens_seen": 140202155,
      "step": 6506,
      "time_per_iteration": 3.4956796169281006
    },
    {
      "auxiliary_loss_clip": 0.01222659,
      "auxiliary_loss_mlp": 0.01029651,
      "balance_loss_clip": 1.04912543,
      "balance_loss_mlp": 1.02231336,
      "epoch": 0.7824204893885649,
      "flos": 24535642815360.0,
      "grad_norm": 1.8980746584879422,
      "language_loss": 0.84159493,
      "learning_rate": 4.762324945725109e-07,
      "loss": 0.86411798,
      "num_input_tokens_seen": 140221600,
      "step": 6507,
      "time_per_iteration": 2.6636815071105957
    },
    {
      "auxiliary_loss_clip": 0.01269062,
      "auxiliary_loss_mlp": 0.01032264,
      "balance_loss_clip": 1.04886365,
      "balance_loss_mlp": 1.02517724,
      "epoch": 0.782540732279204,
      "flos": 27415211402880.0,
      "grad_norm": 1.686740503529404,
      "language_loss": 0.76004821,
      "learning_rate": 4.7572805948829844e-07,
      "loss": 0.7830615,
      "num_input_tokens_seen": 140241860,
      "step": 6508,
      "time_per_iteration": 2.6994757652282715
    },
    {
      "auxiliary_loss_clip": 0.01375595,
      "auxiliary_loss_mlp": 0.01022186,
      "balance_loss_clip": 1.04403043,
      "balance_loss_mlp": 1.01514637,
      "epoch": 0.7826609751698431,
      "flos": 24353216616960.0,
      "grad_norm": 2.2047525621016373,
      "language_loss": 0.70883805,
      "learning_rate": 4.7522385563492795e-07,
      "loss": 0.73281586,
      "num_input_tokens_seen": 140262160,
      "step": 6509,
      "time_per_iteration": 2.740216016769409
    },
    {
      "auxiliary_loss_clip": 0.01326834,
      "auxiliary_loss_mlp": 0.01022975,
      "balance_loss_clip": 1.0467453,
      "balance_loss_mlp": 1.01590252,
      "epoch": 0.7827812180604822,
      "flos": 23988543788160.0,
      "grad_norm": 1.921535164390875,
      "language_loss": 0.70276666,
      "learning_rate": 4.747198830888863e-07,
      "loss": 0.72626472,
      "num_input_tokens_seen": 140282030,
      "step": 6510,
      "time_per_iteration": 2.726646900177002
    },
    {
      "auxiliary_loss_clip": 0.0126988,
      "auxiliary_loss_mlp": 0.01024365,
      "balance_loss_clip": 1.04672372,
      "balance_loss_mlp": 1.0173645,
      "epoch": 0.7829014609511212,
      "flos": 27454318335360.0,
      "grad_norm": 4.945016399591349,
      "language_loss": 0.68527162,
      "learning_rate": 4.742161419266251e-07,
      "loss": 0.70821404,
      "num_input_tokens_seen": 140301190,
      "step": 6511,
      "time_per_iteration": 3.4950335025787354
    },
    {
      "auxiliary_loss_clip": 0.01227819,
      "auxiliary_loss_mlp": 0.01028263,
      "balance_loss_clip": 1.04884744,
      "balance_loss_mlp": 1.02154279,
      "epoch": 0.7830217038417604,
      "flos": 29204532432000.0,
      "grad_norm": 5.728748363918008,
      "language_loss": 0.65273094,
      "learning_rate": 4.7371263222456304e-07,
      "loss": 0.67529178,
      "num_input_tokens_seen": 140318510,
      "step": 6512,
      "time_per_iteration": 2.5953304767608643
    },
    {
      "auxiliary_loss_clip": 0.01113296,
      "auxiliary_loss_mlp": 0.00999149,
      "balance_loss_clip": 1.00818264,
      "balance_loss_mlp": 0.99813557,
      "epoch": 0.7831419467323995,
      "flos": 60950895822720.0,
      "grad_norm": 0.8092152268437073,
      "language_loss": 0.61319923,
      "learning_rate": 4.7320935405908004e-07,
      "loss": 0.63432366,
      "num_input_tokens_seen": 140379380,
      "step": 6513,
      "time_per_iteration": 3.093642234802246
    },
    {
      "auxiliary_loss_clip": 0.01178159,
      "auxiliary_loss_mlp": 0.01025181,
      "balance_loss_clip": 1.04931998,
      "balance_loss_mlp": 1.01735234,
      "epoch": 0.7832621896230385,
      "flos": 19682531320320.0,
      "grad_norm": 2.1468910131142334,
      "language_loss": 0.84444451,
      "learning_rate": 4.7270630750652475e-07,
      "loss": 0.86647791,
      "num_input_tokens_seen": 140395335,
      "step": 6514,
      "time_per_iteration": 2.4771690368652344
    },
    {
      "auxiliary_loss_clip": 0.01217305,
      "auxiliary_loss_mlp": 0.01027242,
      "balance_loss_clip": 1.0458889,
      "balance_loss_mlp": 1.02069128,
      "epoch": 0.7833824325136777,
      "flos": 25009232659200.0,
      "grad_norm": 1.8197014898160606,
      "language_loss": 0.8055495,
      "learning_rate": 4.7220349264320746e-07,
      "loss": 0.82799494,
      "num_input_tokens_seen": 140414420,
      "step": 6515,
      "time_per_iteration": 2.5470292568206787
    },
    {
      "auxiliary_loss_clip": 0.0111667,
      "auxiliary_loss_mlp": 0.01000683,
      "balance_loss_clip": 1.00914431,
      "balance_loss_mlp": 0.99974674,
      "epoch": 0.7835026754043167,
      "flos": 68800142517120.0,
      "grad_norm": 0.7312329210006615,
      "language_loss": 0.54900229,
      "learning_rate": 4.71700909545407e-07,
      "loss": 0.57017577,
      "num_input_tokens_seen": 140477365,
      "step": 6516,
      "time_per_iteration": 3.133159875869751
    },
    {
      "auxiliary_loss_clip": 0.01225425,
      "auxiliary_loss_mlp": 0.01024281,
      "balance_loss_clip": 1.04728293,
      "balance_loss_mlp": 1.01734638,
      "epoch": 0.7836229182949558,
      "flos": 19864598382720.0,
      "grad_norm": 1.8680053352191384,
      "language_loss": 0.7700007,
      "learning_rate": 4.711985582893627e-07,
      "loss": 0.79249775,
      "num_input_tokens_seen": 140495885,
      "step": 6517,
      "time_per_iteration": 2.6112756729125977
    },
    {
      "auxiliary_loss_clip": 0.01372482,
      "auxiliary_loss_mlp": 0.01030471,
      "balance_loss_clip": 1.04100037,
      "balance_loss_mlp": 1.02367282,
      "epoch": 0.783743161185595,
      "flos": 22965843755520.0,
      "grad_norm": 1.6758813754421278,
      "language_loss": 0.71969861,
      "learning_rate": 4.706964389512811e-07,
      "loss": 0.74372816,
      "num_input_tokens_seen": 140515920,
      "step": 6518,
      "time_per_iteration": 2.691852331161499
    },
    {
      "auxiliary_loss_clip": 0.01173922,
      "auxiliary_loss_mlp": 0.01022963,
      "balance_loss_clip": 1.05054009,
      "balance_loss_mlp": 1.01626647,
      "epoch": 0.783863404076234,
      "flos": 12458489777280.0,
      "grad_norm": 2.0569095019056713,
      "language_loss": 0.87586951,
      "learning_rate": 4.701945516073345e-07,
      "loss": 0.89783841,
      "num_input_tokens_seen": 140533395,
      "step": 6519,
      "time_per_iteration": 2.587524652481079
    },
    {
      "auxiliary_loss_clip": 0.0131759,
      "auxiliary_loss_mlp": 0.01031053,
      "balance_loss_clip": 1.0467416,
      "balance_loss_mlp": 1.02454436,
      "epoch": 0.7839836469668731,
      "flos": 24243940465920.0,
      "grad_norm": 1.8018258922804602,
      "language_loss": 0.75212526,
      "learning_rate": 4.696928963336577e-07,
      "loss": 0.77561164,
      "num_input_tokens_seen": 140552825,
      "step": 6520,
      "time_per_iteration": 2.731614828109741
    },
    {
      "auxiliary_loss_clip": 0.01113335,
      "auxiliary_loss_mlp": 0.00999142,
      "balance_loss_clip": 1.0081768,
      "balance_loss_mlp": 0.99818224,
      "epoch": 0.7841038898575122,
      "flos": 62121978938880.0,
      "grad_norm": 0.8541793200850706,
      "language_loss": 0.60930467,
      "learning_rate": 4.6919147320635224e-07,
      "loss": 0.63042939,
      "num_input_tokens_seen": 140615535,
      "step": 6521,
      "time_per_iteration": 3.1227447986602783
    },
    {
      "auxiliary_loss_clip": 0.01226334,
      "auxiliary_loss_mlp": 0.01025587,
      "balance_loss_clip": 1.04809046,
      "balance_loss_mlp": 1.01862526,
      "epoch": 0.7842241327481513,
      "flos": 20193899293440.0,
      "grad_norm": 2.171653314047841,
      "language_loss": 0.73438227,
      "learning_rate": 4.6869028230148286e-07,
      "loss": 0.75690144,
      "num_input_tokens_seen": 140633330,
      "step": 6522,
      "time_per_iteration": 2.590559959411621
    },
    {
      "auxiliary_loss_clip": 0.01319404,
      "auxiliary_loss_mlp": 0.01028343,
      "balance_loss_clip": 1.04184985,
      "balance_loss_mlp": 1.02154517,
      "epoch": 0.7843443756387903,
      "flos": 28074531496320.0,
      "grad_norm": 2.415744720248586,
      "language_loss": 0.5953331,
      "learning_rate": 4.6818932369507957e-07,
      "loss": 0.61881059,
      "num_input_tokens_seen": 140652830,
      "step": 6523,
      "time_per_iteration": 2.698227882385254
    },
    {
      "auxiliary_loss_clip": 0.0122451,
      "auxiliary_loss_mlp": 0.01028436,
      "balance_loss_clip": 1.05072117,
      "balance_loss_mlp": 1.02159071,
      "epoch": 0.7844646185294295,
      "flos": 21323397438720.0,
      "grad_norm": 2.6063122595264017,
      "language_loss": 0.89035887,
      "learning_rate": 4.676885974631386e-07,
      "loss": 0.91288829,
      "num_input_tokens_seen": 140671190,
      "step": 6524,
      "time_per_iteration": 2.6628284454345703
    },
    {
      "auxiliary_loss_clip": 0.0122321,
      "auxiliary_loss_mlp": 0.01027494,
      "balance_loss_clip": 1.04951406,
      "balance_loss_mlp": 1.02068758,
      "epoch": 0.7845848614200686,
      "flos": 23656585271040.0,
      "grad_norm": 1.9378268111176016,
      "language_loss": 0.81132579,
      "learning_rate": 4.67188103681619e-07,
      "loss": 0.83383286,
      "num_input_tokens_seen": 140690975,
      "step": 6525,
      "time_per_iteration": 2.6460907459259033
    },
    {
      "auxiliary_loss_clip": 0.01227133,
      "auxiliary_loss_mlp": 0.02564861,
      "balance_loss_clip": 1.05278683,
      "balance_loss_mlp": 0.9999544,
      "epoch": 0.7847051043107076,
      "flos": 23402194174080.0,
      "grad_norm": 2.0261819456330876,
      "language_loss": 0.69400871,
      "learning_rate": 4.666878424264453e-07,
      "loss": 0.73192871,
      "num_input_tokens_seen": 140710930,
      "step": 6526,
      "time_per_iteration": 2.6044821739196777
    },
    {
      "auxiliary_loss_clip": 0.0127015,
      "auxiliary_loss_mlp": 0.0102641,
      "balance_loss_clip": 1.04676938,
      "balance_loss_mlp": 1.02044034,
      "epoch": 0.7848253472013467,
      "flos": 19022277473280.0,
      "grad_norm": 1.9073458010830713,
      "language_loss": 0.73604548,
      "learning_rate": 4.661878137735069e-07,
      "loss": 0.75901109,
      "num_input_tokens_seen": 140729120,
      "step": 6527,
      "time_per_iteration": 2.6349282264709473
    },
    {
      "auxiliary_loss_clip": 0.01270448,
      "auxiliary_loss_mlp": 0.01026884,
      "balance_loss_clip": 1.04699183,
      "balance_loss_mlp": 1.02029717,
      "epoch": 0.7849455900919858,
      "flos": 21179180332800.0,
      "grad_norm": 1.949754506054307,
      "language_loss": 0.74986196,
      "learning_rate": 4.656880177986571e-07,
      "loss": 0.77283525,
      "num_input_tokens_seen": 140747665,
      "step": 6528,
      "time_per_iteration": 3.4811906814575195
    },
    {
      "auxiliary_loss_clip": 0.0127723,
      "auxiliary_loss_mlp": 0.01022937,
      "balance_loss_clip": 1.04515779,
      "balance_loss_mlp": 1.01575494,
      "epoch": 0.7850658329826249,
      "flos": 19536482620800.0,
      "grad_norm": 2.5153828254878214,
      "language_loss": 0.81900287,
      "learning_rate": 4.6518845457771607e-07,
      "loss": 0.84200454,
      "num_input_tokens_seen": 140766525,
      "step": 6529,
      "time_per_iteration": 3.5845248699188232
    },
    {
      "auxiliary_loss_clip": 0.01220865,
      "auxiliary_loss_mlp": 0.02564224,
      "balance_loss_clip": 1.04643357,
      "balance_loss_mlp": 0.99992585,
      "epoch": 0.7851860758732639,
      "flos": 12495334152960.0,
      "grad_norm": 1.8929920592932543,
      "language_loss": 0.78963864,
      "learning_rate": 4.646891241864652e-07,
      "loss": 0.82748955,
      "num_input_tokens_seen": 140785090,
      "step": 6530,
      "time_per_iteration": 2.6062588691711426
    },
    {
      "auxiliary_loss_clip": 0.01225835,
      "auxiliary_loss_mlp": 0.01027182,
      "balance_loss_clip": 1.04734898,
      "balance_loss_mlp": 1.01982653,
      "epoch": 0.7853063187639031,
      "flos": 22960959505920.0,
      "grad_norm": 2.103291067480908,
      "language_loss": 0.72877944,
      "learning_rate": 4.6419002670065397e-07,
      "loss": 0.75130963,
      "num_input_tokens_seen": 140804670,
      "step": 6531,
      "time_per_iteration": 2.6322989463806152
    },
    {
      "auxiliary_loss_clip": 0.01325271,
      "auxiliary_loss_mlp": 0.0102596,
      "balance_loss_clip": 1.04848957,
      "balance_loss_mlp": 1.01867962,
      "epoch": 0.7854265616545422,
      "flos": 17347260499200.0,
      "grad_norm": 2.473553960984657,
      "language_loss": 0.86686575,
      "learning_rate": 4.6369116219599445e-07,
      "loss": 0.89037806,
      "num_input_tokens_seen": 140820655,
      "step": 6532,
      "time_per_iteration": 3.56000018119812
    },
    {
      "auxiliary_loss_clip": 0.01319971,
      "auxiliary_loss_mlp": 0.01024515,
      "balance_loss_clip": 1.04363596,
      "balance_loss_mlp": 1.01778269,
      "epoch": 0.7855468045451812,
      "flos": 23838293197440.0,
      "grad_norm": 1.6565436374006373,
      "language_loss": 0.79698825,
      "learning_rate": 4.631925307481637e-07,
      "loss": 0.82043314,
      "num_input_tokens_seen": 140840470,
      "step": 6533,
      "time_per_iteration": 2.682694435119629
    },
    {
      "auxiliary_loss_clip": 0.0127331,
      "auxiliary_loss_mlp": 0.01020688,
      "balance_loss_clip": 1.0482533,
      "balance_loss_mlp": 1.01419151,
      "epoch": 0.7856670474358204,
      "flos": 25666792986240.0,
      "grad_norm": 2.6653858342167527,
      "language_loss": 0.75798607,
      "learning_rate": 4.6269413243280533e-07,
      "loss": 0.78092605,
      "num_input_tokens_seen": 140859890,
      "step": 6534,
      "time_per_iteration": 2.662677764892578
    },
    {
      "auxiliary_loss_clip": 0.01286295,
      "auxiliary_loss_mlp": 0.01025778,
      "balance_loss_clip": 1.0529896,
      "balance_loss_mlp": 1.0179162,
      "epoch": 0.7857872903264594,
      "flos": 18144656472960.0,
      "grad_norm": 3.3616167101560883,
      "language_loss": 0.74311936,
      "learning_rate": 4.621959673255236e-07,
      "loss": 0.76624012,
      "num_input_tokens_seen": 140876190,
      "step": 6535,
      "time_per_iteration": 2.6561100482940674
    },
    {
      "auxiliary_loss_clip": 0.01368702,
      "auxiliary_loss_mlp": 0.01024812,
      "balance_loss_clip": 1.04315031,
      "balance_loss_mlp": 1.01817465,
      "epoch": 0.7859075332170985,
      "flos": 14386138081920.0,
      "grad_norm": 2.4215235256187,
      "language_loss": 0.90665025,
      "learning_rate": 4.6169803550189135e-07,
      "loss": 0.93058538,
      "num_input_tokens_seen": 140891885,
      "step": 6536,
      "time_per_iteration": 2.724778890609741
    },
    {
      "auxiliary_loss_clip": 0.01373038,
      "auxiliary_loss_mlp": 0.01027096,
      "balance_loss_clip": 1.04952478,
      "balance_loss_mlp": 1.0192939,
      "epoch": 0.7860277761077377,
      "flos": 19864059678720.0,
      "grad_norm": 1.888501626962539,
      "language_loss": 0.7735858,
      "learning_rate": 4.6120033703744355e-07,
      "loss": 0.7975871,
      "num_input_tokens_seen": 140910780,
      "step": 6537,
      "time_per_iteration": 3.6518588066101074
    },
    {
      "auxiliary_loss_clip": 0.01265589,
      "auxiliary_loss_mlp": 0.01019448,
      "balance_loss_clip": 1.04520309,
      "balance_loss_mlp": 1.01262593,
      "epoch": 0.7861480189983767,
      "flos": 26396174557440.0,
      "grad_norm": 1.9245845689171825,
      "language_loss": 0.7821722,
      "learning_rate": 4.607028720076822e-07,
      "loss": 0.8050226,
      "num_input_tokens_seen": 140927460,
      "step": 6538,
      "time_per_iteration": 2.692481279373169
    },
    {
      "auxiliary_loss_clip": 0.01220901,
      "auxiliary_loss_mlp": 0.01025817,
      "balance_loss_clip": 1.04786813,
      "balance_loss_mlp": 1.01899791,
      "epoch": 0.7862682618890158,
      "flos": 24236578177920.0,
      "grad_norm": 2.5329052603248443,
      "language_loss": 0.73435932,
      "learning_rate": 4.6020564048807074e-07,
      "loss": 0.75682652,
      "num_input_tokens_seen": 140945135,
      "step": 6539,
      "time_per_iteration": 2.61336350440979
    },
    {
      "auxiliary_loss_clip": 0.01224176,
      "auxiliary_loss_mlp": 0.01024695,
      "balance_loss_clip": 1.04821372,
      "balance_loss_mlp": 1.01805782,
      "epoch": 0.7863885047796549,
      "flos": 47551508259840.0,
      "grad_norm": 1.9306213332539468,
      "language_loss": 0.72018588,
      "learning_rate": 4.5970864255403883e-07,
      "loss": 0.74267459,
      "num_input_tokens_seen": 140966660,
      "step": 6540,
      "time_per_iteration": 2.8369219303131104
    },
    {
      "auxiliary_loss_clip": 0.01218113,
      "auxiliary_loss_mlp": 0.01023684,
      "balance_loss_clip": 1.04696774,
      "balance_loss_mlp": 1.01684391,
      "epoch": 0.786508747670294,
      "flos": 24389234979840.0,
      "grad_norm": 1.9104086502464483,
      "language_loss": 0.82192457,
      "learning_rate": 4.59211878280982e-07,
      "loss": 0.84434247,
      "num_input_tokens_seen": 140986175,
      "step": 6541,
      "time_per_iteration": 2.6960506439208984
    },
    {
      "auxiliary_loss_clip": 0.0127239,
      "auxiliary_loss_mlp": 0.01029075,
      "balance_loss_clip": 1.04670548,
      "balance_loss_mlp": 1.02114749,
      "epoch": 0.786628990560933,
      "flos": 18041234238720.0,
      "grad_norm": 2.169398331492266,
      "language_loss": 0.7014268,
      "learning_rate": 4.587153477442578e-07,
      "loss": 0.72444147,
      "num_input_tokens_seen": 141002490,
      "step": 6542,
      "time_per_iteration": 2.630357027053833
    },
    {
      "auxiliary_loss_clip": 0.01179397,
      "auxiliary_loss_mlp": 0.0102572,
      "balance_loss_clip": 1.05172324,
      "balance_loss_mlp": 1.01805472,
      "epoch": 0.7867492334515722,
      "flos": 25848860048640.0,
      "grad_norm": 5.026102854688579,
      "language_loss": 0.81694633,
      "learning_rate": 4.582190510191899e-07,
      "loss": 0.83899748,
      "num_input_tokens_seen": 141021150,
      "step": 6543,
      "time_per_iteration": 2.6463370323181152
    },
    {
      "auxiliary_loss_clip": 0.0131783,
      "auxiliary_loss_mlp": 0.01022471,
      "balance_loss_clip": 1.0467819,
      "balance_loss_mlp": 1.01596236,
      "epoch": 0.7868694763422113,
      "flos": 16580819070720.0,
      "grad_norm": 2.2695689820513048,
      "language_loss": 0.87532228,
      "learning_rate": 4.5772298818106625e-07,
      "loss": 0.89872533,
      "num_input_tokens_seen": 141036940,
      "step": 6544,
      "time_per_iteration": 2.7277286052703857
    },
    {
      "auxiliary_loss_clip": 0.01235727,
      "auxiliary_loss_mlp": 0.01022588,
      "balance_loss_clip": 1.05089355,
      "balance_loss_mlp": 1.01531577,
      "epoch": 0.7869897192328503,
      "flos": 29386276272000.0,
      "grad_norm": 3.0202176119537802,
      "language_loss": 0.7216152,
      "learning_rate": 4.572271593051384e-07,
      "loss": 0.74419832,
      "num_input_tokens_seen": 141054295,
      "step": 6545,
      "time_per_iteration": 2.7787137031555176
    },
    {
      "auxiliary_loss_clip": 0.01362943,
      "auxiliary_loss_mlp": 0.010223,
      "balance_loss_clip": 1.04328704,
      "balance_loss_mlp": 1.01518941,
      "epoch": 0.7871099621234895,
      "flos": 17128923678720.0,
      "grad_norm": 1.7192610089894722,
      "language_loss": 0.78298163,
      "learning_rate": 4.567315644666245e-07,
      "loss": 0.8068341,
      "num_input_tokens_seen": 141073090,
      "step": 6546,
      "time_per_iteration": 2.723257541656494
    },
    {
      "auxiliary_loss_clip": 0.01319643,
      "auxiliary_loss_mlp": 0.01027869,
      "balance_loss_clip": 1.04568601,
      "balance_loss_mlp": 1.02151167,
      "epoch": 0.7872302050141285,
      "flos": 23440187784960.0,
      "grad_norm": 2.323116678312739,
      "language_loss": 0.85045779,
      "learning_rate": 4.5623620374070507e-07,
      "loss": 0.87393296,
      "num_input_tokens_seen": 141092405,
      "step": 6547,
      "time_per_iteration": 2.7258362770080566
    },
    {
      "auxiliary_loss_clip": 0.01220599,
      "auxiliary_loss_mlp": 0.01002043,
      "balance_loss_clip": 1.00878024,
      "balance_loss_mlp": 1.00107181,
      "epoch": 0.7873504479047676,
      "flos": 65959752689280.0,
      "grad_norm": 0.7559601810709718,
      "language_loss": 0.58327967,
      "learning_rate": 4.557410772025263e-07,
      "loss": 0.60550612,
      "num_input_tokens_seen": 141154355,
      "step": 6548,
      "time_per_iteration": 3.3300929069519043
    },
    {
      "auxiliary_loss_clip": 0.01271804,
      "auxiliary_loss_mlp": 0.01021655,
      "balance_loss_clip": 1.04634678,
      "balance_loss_mlp": 1.01510751,
      "epoch": 0.7874706907954068,
      "flos": 23258336204160.0,
      "grad_norm": 1.899966942458922,
      "language_loss": 0.6680854,
      "learning_rate": 4.5524618492719803e-07,
      "loss": 0.69102001,
      "num_input_tokens_seen": 141173575,
      "step": 6549,
      "time_per_iteration": 2.6873486042022705
    },
    {
      "auxiliary_loss_clip": 0.01224026,
      "auxiliary_loss_mlp": 0.01022273,
      "balance_loss_clip": 1.04808295,
      "balance_loss_mlp": 1.01598775,
      "epoch": 0.7875909336860458,
      "flos": 28767786963840.0,
      "grad_norm": 1.8875013092346509,
      "language_loss": 0.79027241,
      "learning_rate": 4.54751526989795e-07,
      "loss": 0.81273544,
      "num_input_tokens_seen": 141195415,
      "step": 6550,
      "time_per_iteration": 2.6911325454711914
    },
    {
      "auxiliary_loss_clip": 0.01226423,
      "auxiliary_loss_mlp": 0.01026774,
      "balance_loss_clip": 1.04728067,
      "balance_loss_mlp": 1.0202471,
      "epoch": 0.7877111765766849,
      "flos": 18697286194560.0,
      "grad_norm": 2.1652214368593006,
      "language_loss": 0.7923677,
      "learning_rate": 4.5425710346535775e-07,
      "loss": 0.81489968,
      "num_input_tokens_seen": 141213360,
      "step": 6551,
      "time_per_iteration": 2.5745956897735596
    },
    {
      "auxiliary_loss_clip": 0.0122601,
      "auxiliary_loss_mlp": 0.01025624,
      "balance_loss_clip": 1.04893005,
      "balance_loss_mlp": 1.01848602,
      "epoch": 0.787831419467324,
      "flos": 27592968833280.0,
      "grad_norm": 2.4441902949608743,
      "language_loss": 0.81690192,
      "learning_rate": 4.537629144288877e-07,
      "loss": 0.83941823,
      "num_input_tokens_seen": 141230815,
      "step": 6552,
      "time_per_iteration": 2.670820474624634
    },
    {
      "auxiliary_loss_clip": 0.01278724,
      "auxiliary_loss_mlp": 0.01029603,
      "balance_loss_clip": 1.04200053,
      "balance_loss_mlp": 1.02261996,
      "epoch": 0.7879516623579631,
      "flos": 18150187167360.0,
      "grad_norm": 2.197972977301138,
      "language_loss": 0.75359613,
      "learning_rate": 4.5326895995535477e-07,
      "loss": 0.7766794,
      "num_input_tokens_seen": 141249715,
      "step": 6553,
      "time_per_iteration": 2.7191784381866455
    },
    {
      "auxiliary_loss_clip": 0.01219613,
      "auxiliary_loss_mlp": 0.01024537,
      "balance_loss_clip": 1.04722321,
      "balance_loss_mlp": 1.01767039,
      "epoch": 0.7880719052486022,
      "flos": 20339193807360.0,
      "grad_norm": 2.6202734800381675,
      "language_loss": 0.84668213,
      "learning_rate": 4.527752401196907e-07,
      "loss": 0.86912364,
      "num_input_tokens_seen": 141267730,
      "step": 6554,
      "time_per_iteration": 3.551793336868286
    },
    {
      "auxiliary_loss_clip": 0.01271035,
      "auxiliary_loss_mlp": 0.01024502,
      "balance_loss_clip": 1.04619229,
      "balance_loss_mlp": 1.01772809,
      "epoch": 0.7881921481392413,
      "flos": 21653237053440.0,
      "grad_norm": 1.7718158848671413,
      "language_loss": 0.66590428,
      "learning_rate": 4.5228175499679254e-07,
      "loss": 0.68885958,
      "num_input_tokens_seen": 141287315,
      "step": 6555,
      "time_per_iteration": 3.655639410018921
    },
    {
      "auxiliary_loss_clip": 0.01116508,
      "auxiliary_loss_mlp": 0.01002794,
      "balance_loss_clip": 1.0082618,
      "balance_loss_mlp": 1.00185192,
      "epoch": 0.7883123910298804,
      "flos": 68565860058240.0,
      "grad_norm": 1.0163173109035322,
      "language_loss": 0.54510981,
      "learning_rate": 4.5178850466152174e-07,
      "loss": 0.56630284,
      "num_input_tokens_seen": 141346145,
      "step": 6556,
      "time_per_iteration": 3.2305045127868652
    },
    {
      "auxiliary_loss_clip": 0.0127024,
      "auxiliary_loss_mlp": 0.01026661,
      "balance_loss_clip": 1.04411578,
      "balance_loss_mlp": 1.02039337,
      "epoch": 0.7884326339205194,
      "flos": 19318217627520.0,
      "grad_norm": 1.871006743321473,
      "language_loss": 0.81453586,
      "learning_rate": 4.512954891887031e-07,
      "loss": 0.8375048,
      "num_input_tokens_seen": 141364445,
      "step": 6557,
      "time_per_iteration": 2.6806845664978027
    },
    {
      "auxiliary_loss_clip": 0.01277433,
      "auxiliary_loss_mlp": 0.01028058,
      "balance_loss_clip": 1.04870057,
      "balance_loss_mlp": 1.02095318,
      "epoch": 0.7885528768111585,
      "flos": 17784903807360.0,
      "grad_norm": 2.519839607036761,
      "language_loss": 0.83551192,
      "learning_rate": 4.5080270865312806e-07,
      "loss": 0.85856682,
      "num_input_tokens_seen": 141381640,
      "step": 6558,
      "time_per_iteration": 3.5463266372680664
    },
    {
      "auxiliary_loss_clip": 0.01221912,
      "auxiliary_loss_mlp": 0.01027582,
      "balance_loss_clip": 1.04746532,
      "balance_loss_mlp": 1.02094197,
      "epoch": 0.7886731197017977,
      "flos": 18807639753600.0,
      "grad_norm": 1.9870924670401287,
      "language_loss": 0.71021163,
      "learning_rate": 4.5031016312954985e-07,
      "loss": 0.73270655,
      "num_input_tokens_seen": 141399955,
      "step": 6559,
      "time_per_iteration": 2.5864317417144775
    },
    {
      "auxiliary_loss_clip": 0.01232076,
      "auxiliary_loss_mlp": 0.01023,
      "balance_loss_clip": 1.05068541,
      "balance_loss_mlp": 1.01588917,
      "epoch": 0.7887933625924367,
      "flos": 33365358126720.0,
      "grad_norm": 2.3549746822793924,
      "language_loss": 0.74523985,
      "learning_rate": 4.498178526926886e-07,
      "loss": 0.76779062,
      "num_input_tokens_seen": 141420820,
      "step": 6560,
      "time_per_iteration": 2.7205474376678467
    },
    {
      "auxiliary_loss_clip": 0.01175165,
      "auxiliary_loss_mlp": 0.0102337,
      "balance_loss_clip": 1.05079067,
      "balance_loss_mlp": 1.01633668,
      "epoch": 0.7889136054830758,
      "flos": 17019360218880.0,
      "grad_norm": 2.3393432463937476,
      "language_loss": 0.72427732,
      "learning_rate": 4.4932577741722635e-07,
      "loss": 0.74626267,
      "num_input_tokens_seen": 141439350,
      "step": 6561,
      "time_per_iteration": 2.587312698364258
    },
    {
      "auxiliary_loss_clip": 0.01273138,
      "auxiliary_loss_mlp": 0.01022423,
      "balance_loss_clip": 1.04576802,
      "balance_loss_mlp": 1.01527345,
      "epoch": 0.7890338483737149,
      "flos": 29424629018880.0,
      "grad_norm": 2.514820878396196,
      "language_loss": 0.74473941,
      "learning_rate": 4.4883393737780985e-07,
      "loss": 0.76769501,
      "num_input_tokens_seen": 141460300,
      "step": 6562,
      "time_per_iteration": 2.735563278198242
    },
    {
      "auxiliary_loss_clip": 0.01219341,
      "auxiliary_loss_mlp": 0.01027567,
      "balance_loss_clip": 1.04485178,
      "balance_loss_mlp": 1.02026534,
      "epoch": 0.789154091264354,
      "flos": 19971576063360.0,
      "grad_norm": 2.3393367916886563,
      "language_loss": 0.78063256,
      "learning_rate": 4.4834233264905254e-07,
      "loss": 0.8031016,
      "num_input_tokens_seen": 141477315,
      "step": 6563,
      "time_per_iteration": 3.583075761795044
    },
    {
      "auxiliary_loss_clip": 0.01316983,
      "auxiliary_loss_mlp": 0.01027156,
      "balance_loss_clip": 1.04424763,
      "balance_loss_mlp": 1.01992011,
      "epoch": 0.789274334154993,
      "flos": 14537825216640.0,
      "grad_norm": 2.501822297182343,
      "language_loss": 0.70896089,
      "learning_rate": 4.478509633055294e-07,
      "loss": 0.73240232,
      "num_input_tokens_seen": 141495025,
      "step": 6564,
      "time_per_iteration": 2.676697254180908
    },
    {
      "auxiliary_loss_clip": 0.01276271,
      "auxiliary_loss_mlp": 0.01031774,
      "balance_loss_clip": 1.04662037,
      "balance_loss_mlp": 1.02379346,
      "epoch": 0.7893945770456322,
      "flos": 21827403123840.0,
      "grad_norm": 2.901985186213801,
      "language_loss": 0.80219144,
      "learning_rate": 4.473598294217813e-07,
      "loss": 0.8252719,
      "num_input_tokens_seen": 141510450,
      "step": 6565,
      "time_per_iteration": 2.6537797451019287
    },
    {
      "auxiliary_loss_clip": 0.01219215,
      "auxiliary_loss_mlp": 0.01031594,
      "balance_loss_clip": 1.04898143,
      "balance_loss_mlp": 1.02476907,
      "epoch": 0.7895148199362713,
      "flos": 20740639184640.0,
      "grad_norm": 2.3829946053577706,
      "language_loss": 0.71878541,
      "learning_rate": 4.468689310723124e-07,
      "loss": 0.74129355,
      "num_input_tokens_seen": 141528265,
      "step": 6566,
      "time_per_iteration": 2.641378164291382
    },
    {
      "auxiliary_loss_clip": 0.01327062,
      "auxiliary_loss_mlp": 0.01030145,
      "balance_loss_clip": 1.04481721,
      "balance_loss_mlp": 1.02334714,
      "epoch": 0.7896350628269103,
      "flos": 16690669839360.0,
      "grad_norm": 1.8757681927146213,
      "language_loss": 0.78758919,
      "learning_rate": 4.463782683315913e-07,
      "loss": 0.81116128,
      "num_input_tokens_seen": 141547270,
      "step": 6567,
      "time_per_iteration": 2.647749185562134
    },
    {
      "auxiliary_loss_clip": 0.01171895,
      "auxiliary_loss_mlp": 0.01024048,
      "balance_loss_clip": 1.0497092,
      "balance_loss_mlp": 1.01732135,
      "epoch": 0.7897553057175495,
      "flos": 22638374438400.0,
      "grad_norm": 2.0194899763242202,
      "language_loss": 0.7369054,
      "learning_rate": 4.458878412740523e-07,
      "loss": 0.75886482,
      "num_input_tokens_seen": 141566050,
      "step": 6568,
      "time_per_iteration": 2.6744232177734375
    },
    {
      "auxiliary_loss_clip": 0.01221316,
      "auxiliary_loss_mlp": 0.01025144,
      "balance_loss_clip": 1.04856908,
      "balance_loss_mlp": 1.01836681,
      "epoch": 0.7898755486081885,
      "flos": 14537573821440.0,
      "grad_norm": 2.702286605860869,
      "language_loss": 0.78073782,
      "learning_rate": 4.453976499740919e-07,
      "loss": 0.80320239,
      "num_input_tokens_seen": 141583695,
      "step": 6569,
      "time_per_iteration": 2.5739753246307373
    },
    {
      "auxiliary_loss_clip": 0.01222218,
      "auxiliary_loss_mlp": 0.01027554,
      "balance_loss_clip": 1.05005813,
      "balance_loss_mlp": 1.02047634,
      "epoch": 0.7899957914988276,
      "flos": 17238487138560.0,
      "grad_norm": 2.8089052471547147,
      "language_loss": 0.77851856,
      "learning_rate": 4.4490769450607215e-07,
      "loss": 0.80101621,
      "num_input_tokens_seen": 141601320,
      "step": 6570,
      "time_per_iteration": 2.5707685947418213
    },
    {
      "auxiliary_loss_clip": 0.01323714,
      "auxiliary_loss_mlp": 0.01029636,
      "balance_loss_clip": 1.04111528,
      "balance_loss_mlp": 1.02192891,
      "epoch": 0.7901160343894668,
      "flos": 41279351086080.0,
      "grad_norm": 1.9194913857101397,
      "language_loss": 0.72523445,
      "learning_rate": 4.4441797494431845e-07,
      "loss": 0.74876785,
      "num_input_tokens_seen": 141623125,
      "step": 6571,
      "time_per_iteration": 2.8937056064605713
    },
    {
      "auxiliary_loss_clip": 0.012209,
      "auxiliary_loss_mlp": 0.01025152,
      "balance_loss_clip": 1.04988635,
      "balance_loss_mlp": 1.0181483,
      "epoch": 0.7902362772801058,
      "flos": 16837005847680.0,
      "grad_norm": 2.24350718216248,
      "language_loss": 0.78224379,
      "learning_rate": 4.439284913631207e-07,
      "loss": 0.80470431,
      "num_input_tokens_seen": 141640335,
      "step": 6572,
      "time_per_iteration": 2.568086624145508
    },
    {
      "auxiliary_loss_clip": 0.01223577,
      "auxiliary_loss_mlp": 0.01022859,
      "balance_loss_clip": 1.04921508,
      "balance_loss_mlp": 1.01571202,
      "epoch": 0.7903565201707449,
      "flos": 27125987091840.0,
      "grad_norm": 1.875796914280907,
      "language_loss": 0.83897436,
      "learning_rate": 4.434392438367347e-07,
      "loss": 0.86143875,
      "num_input_tokens_seen": 141659760,
      "step": 6573,
      "time_per_iteration": 2.724194288253784
    },
    {
      "auxiliary_loss_clip": 0.01128133,
      "auxiliary_loss_mlp": 0.01027271,
      "balance_loss_clip": 1.04831183,
      "balance_loss_mlp": 1.0199579,
      "epoch": 0.790476763061384,
      "flos": 31025167142400.0,
      "grad_norm": 1.7514729726947278,
      "language_loss": 0.74095935,
      "learning_rate": 4.4295023243937677e-07,
      "loss": 0.7625134,
      "num_input_tokens_seen": 141679965,
      "step": 6574,
      "time_per_iteration": 2.7108240127563477
    },
    {
      "auxiliary_loss_clip": 0.0123077,
      "auxiliary_loss_mlp": 0.01028827,
      "balance_loss_clip": 1.05241346,
      "balance_loss_mlp": 1.02151036,
      "epoch": 0.7905970059520231,
      "flos": 22089084681600.0,
      "grad_norm": 2.021415957806739,
      "language_loss": 0.8039242,
      "learning_rate": 4.4246145724523123e-07,
      "loss": 0.82652014,
      "num_input_tokens_seen": 141697710,
      "step": 6575,
      "time_per_iteration": 2.6468353271484375
    },
    {
      "auxiliary_loss_clip": 0.01319479,
      "auxiliary_loss_mlp": 0.01025821,
      "balance_loss_clip": 1.04677498,
      "balance_loss_mlp": 1.01930034,
      "epoch": 0.7907172488426621,
      "flos": 20558141159040.0,
      "grad_norm": 2.4463129809514124,
      "language_loss": 0.77295578,
      "learning_rate": 4.41972918328444e-07,
      "loss": 0.79640877,
      "num_input_tokens_seen": 141715145,
      "step": 6576,
      "time_per_iteration": 2.7070324420928955
    },
    {
      "auxiliary_loss_clip": 0.01220968,
      "auxiliary_loss_mlp": 0.01028382,
      "balance_loss_clip": 1.04925108,
      "balance_loss_mlp": 1.02110708,
      "epoch": 0.7908374917333013,
      "flos": 30081542901120.0,
      "grad_norm": 7.952828303959024,
      "language_loss": 0.77514976,
      "learning_rate": 4.4148461576312646e-07,
      "loss": 0.7976433,
      "num_input_tokens_seen": 141734810,
      "step": 6577,
      "time_per_iteration": 2.7014849185943604
    },
    {
      "auxiliary_loss_clip": 0.01224463,
      "auxiliary_loss_mlp": 0.01022349,
      "balance_loss_clip": 1.05074596,
      "balance_loss_mlp": 1.01613808,
      "epoch": 0.7909577346239404,
      "flos": 20996359084800.0,
      "grad_norm": 1.4677296907421407,
      "language_loss": 0.74690473,
      "learning_rate": 4.4099654962335343e-07,
      "loss": 0.76937282,
      "num_input_tokens_seen": 141755260,
      "step": 6578,
      "time_per_iteration": 2.668623685836792
    },
    {
      "auxiliary_loss_clip": 0.01279319,
      "auxiliary_loss_mlp": 0.01030637,
      "balance_loss_clip": 1.04808271,
      "balance_loss_mlp": 1.02345145,
      "epoch": 0.7910779775145794,
      "flos": 26247935128320.0,
      "grad_norm": 2.867462476669237,
      "language_loss": 0.75120568,
      "learning_rate": 4.405087199831636e-07,
      "loss": 0.77430522,
      "num_input_tokens_seen": 141775500,
      "step": 6579,
      "time_per_iteration": 3.6217687129974365
    },
    {
      "auxiliary_loss_clip": 0.01272932,
      "auxiliary_loss_mlp": 0.02565017,
      "balance_loss_clip": 1.04395914,
      "balance_loss_mlp": 0.99992275,
      "epoch": 0.7911982204052186,
      "flos": 22564434291840.0,
      "grad_norm": 3.0669859085509925,
      "language_loss": 0.67646235,
      "learning_rate": 4.400211269165619e-07,
      "loss": 0.71484184,
      "num_input_tokens_seen": 141791955,
      "step": 6580,
      "time_per_iteration": 2.697192430496216
    },
    {
      "auxiliary_loss_clip": 0.01176488,
      "auxiliary_loss_mlp": 0.01029356,
      "balance_loss_clip": 1.05271101,
      "balance_loss_mlp": 1.02304721,
      "epoch": 0.7913184632958576,
      "flos": 23112538899840.0,
      "grad_norm": 1.567859321824638,
      "language_loss": 0.76929224,
      "learning_rate": 4.3953377049751416e-07,
      "loss": 0.79135072,
      "num_input_tokens_seen": 141812380,
      "step": 6581,
      "time_per_iteration": 3.5741453170776367
    },
    {
      "auxiliary_loss_clip": 0.01277463,
      "auxiliary_loss_mlp": 0.01023443,
      "balance_loss_clip": 1.04751718,
      "balance_loss_mlp": 1.01646352,
      "epoch": 0.7914387061864967,
      "flos": 12311758719360.0,
      "grad_norm": 2.6132809708175104,
      "language_loss": 0.77840883,
      "learning_rate": 4.390466507999537e-07,
      "loss": 0.80141789,
      "num_input_tokens_seen": 141828130,
      "step": 6582,
      "time_per_iteration": 2.619466781616211
    },
    {
      "auxiliary_loss_clip": 0.01319689,
      "auxiliary_loss_mlp": 0.01028925,
      "balance_loss_clip": 1.04573202,
      "balance_loss_mlp": 1.02148652,
      "epoch": 0.7915589490771359,
      "flos": 17603267708160.0,
      "grad_norm": 7.741934950024072,
      "language_loss": 0.75945878,
      "learning_rate": 4.385597678977748e-07,
      "loss": 0.78294492,
      "num_input_tokens_seen": 141846965,
      "step": 6583,
      "time_per_iteration": 2.6579172611236572
    },
    {
      "auxiliary_loss_clip": 0.01277283,
      "auxiliary_loss_mlp": 0.01022581,
      "balance_loss_clip": 1.04705667,
      "balance_loss_mlp": 1.01533914,
      "epoch": 0.7916791919677749,
      "flos": 25591272641280.0,
      "grad_norm": 8.104146481718471,
      "language_loss": 0.75381112,
      "learning_rate": 4.3807312186483726e-07,
      "loss": 0.77680981,
      "num_input_tokens_seen": 141867685,
      "step": 6584,
      "time_per_iteration": 3.5911638736724854
    },
    {
      "auxiliary_loss_clip": 0.01223928,
      "auxiliary_loss_mlp": 0.01023456,
      "balance_loss_clip": 1.05349445,
      "balance_loss_mlp": 1.01661611,
      "epoch": 0.791799434858414,
      "flos": 18844340474880.0,
      "grad_norm": 2.825929844239629,
      "language_loss": 0.78845692,
      "learning_rate": 4.375867127749655e-07,
      "loss": 0.81093079,
      "num_input_tokens_seen": 141885960,
      "step": 6585,
      "time_per_iteration": 2.6319997310638428
    },
    {
      "auxiliary_loss_clip": 0.01322258,
      "auxiliary_loss_mlp": 0.01027784,
      "balance_loss_clip": 1.04797626,
      "balance_loss_mlp": 1.02055407,
      "epoch": 0.7919196777490531,
      "flos": 25812015672960.0,
      "grad_norm": 2.691243766061732,
      "language_loss": 0.67370105,
      "learning_rate": 4.3710054070194744e-07,
      "loss": 0.69720143,
      "num_input_tokens_seen": 141905655,
      "step": 6586,
      "time_per_iteration": 2.7012381553649902
    },
    {
      "auxiliary_loss_clip": 0.01175756,
      "auxiliary_loss_mlp": 0.02568701,
      "balance_loss_clip": 1.04931259,
      "balance_loss_mlp": 0.99992043,
      "epoch": 0.7920399206396922,
      "flos": 11947624594560.0,
      "grad_norm": 3.4557785986806073,
      "language_loss": 0.6591934,
      "learning_rate": 4.3661460571953455e-07,
      "loss": 0.69663793,
      "num_input_tokens_seen": 141922390,
      "step": 6587,
      "time_per_iteration": 2.5723037719726562
    },
    {
      "auxiliary_loss_clip": 0.01224736,
      "auxiliary_loss_mlp": 0.01026755,
      "balance_loss_clip": 1.04711318,
      "balance_loss_mlp": 1.01971269,
      "epoch": 0.7921601635303313,
      "flos": 21579907438080.0,
      "grad_norm": 1.9143861099342492,
      "language_loss": 0.68555343,
      "learning_rate": 4.36128907901443e-07,
      "loss": 0.70806831,
      "num_input_tokens_seen": 141941985,
      "step": 6588,
      "time_per_iteration": 2.6066060066223145
    },
    {
      "auxiliary_loss_clip": 0.01323966,
      "auxiliary_loss_mlp": 0.01027913,
      "balance_loss_clip": 1.04288173,
      "balance_loss_mlp": 1.02090025,
      "epoch": 0.7922804064209703,
      "flos": 18113989236480.0,
      "grad_norm": 2.3459348269471336,
      "language_loss": 0.72722208,
      "learning_rate": 4.356434473213519e-07,
      "loss": 0.75074089,
      "num_input_tokens_seen": 141959435,
      "step": 6589,
      "time_per_iteration": 3.6153619289398193
    },
    {
      "auxiliary_loss_clip": 0.01270636,
      "auxiliary_loss_mlp": 0.01029599,
      "balance_loss_clip": 1.04970872,
      "balance_loss_mlp": 1.02283096,
      "epoch": 0.7924006493116095,
      "flos": 21652806090240.0,
      "grad_norm": 1.5799464025361234,
      "language_loss": 0.79967403,
      "learning_rate": 4.351582240529068e-07,
      "loss": 0.82267636,
      "num_input_tokens_seen": 141980265,
      "step": 6590,
      "time_per_iteration": 2.717794418334961
    },
    {
      "auxiliary_loss_clip": 0.01175975,
      "auxiliary_loss_mlp": 0.01000447,
      "balance_loss_clip": 1.00864816,
      "balance_loss_mlp": 0.99949932,
      "epoch": 0.7925208922022485,
      "flos": 64242755694720.0,
      "grad_norm": 0.6768582829891229,
      "language_loss": 0.58209193,
      "learning_rate": 4.346732381697149e-07,
      "loss": 0.60385615,
      "num_input_tokens_seen": 142044395,
      "step": 6591,
      "time_per_iteration": 3.3079116344451904
    },
    {
      "auxiliary_loss_clip": 0.01267626,
      "auxiliary_loss_mlp": 0.01020737,
      "balance_loss_clip": 1.04815233,
      "balance_loss_mlp": 1.01442766,
      "epoch": 0.7926411350928876,
      "flos": 16941541403520.0,
      "grad_norm": 2.2567918586585245,
      "language_loss": 0.8104862,
      "learning_rate": 4.3418848974534825e-07,
      "loss": 0.83336979,
      "num_input_tokens_seen": 142061335,
      "step": 6592,
      "time_per_iteration": 2.6682419776916504
    },
    {
      "auxiliary_loss_clip": 0.01327957,
      "auxiliary_loss_mlp": 0.01024679,
      "balance_loss_clip": 1.04674792,
      "balance_loss_mlp": 1.01809013,
      "epoch": 0.7927613779835267,
      "flos": 34459987144320.0,
      "grad_norm": 2.120313042192147,
      "language_loss": 0.68730474,
      "learning_rate": 4.3370397885334276e-07,
      "loss": 0.71083111,
      "num_input_tokens_seen": 142081965,
      "step": 6593,
      "time_per_iteration": 2.7692317962646484
    },
    {
      "auxiliary_loss_clip": 0.01221326,
      "auxiliary_loss_mlp": 0.01025285,
      "balance_loss_clip": 1.04803133,
      "balance_loss_mlp": 1.01850843,
      "epoch": 0.7928816208741658,
      "flos": 18951174501120.0,
      "grad_norm": 2.9750981527269857,
      "language_loss": 0.75619775,
      "learning_rate": 4.3321970556719777e-07,
      "loss": 0.77866387,
      "num_input_tokens_seen": 142100260,
      "step": 6594,
      "time_per_iteration": 2.642742156982422
    },
    {
      "auxiliary_loss_clip": 0.01176204,
      "auxiliary_loss_mlp": 0.01022827,
      "balance_loss_clip": 1.05155504,
      "balance_loss_mlp": 1.0158596,
      "epoch": 0.7930018637648049,
      "flos": 18623022825600.0,
      "grad_norm": 3.0009242392483277,
      "language_loss": 0.72312087,
      "learning_rate": 4.3273566996037856e-07,
      "loss": 0.74511117,
      "num_input_tokens_seen": 142116955,
      "step": 6595,
      "time_per_iteration": 2.5672311782836914
    },
    {
      "auxiliary_loss_clip": 0.01271071,
      "auxiliary_loss_mlp": 0.01028352,
      "balance_loss_clip": 1.04779553,
      "balance_loss_mlp": 1.02147961,
      "epoch": 0.793122106655444,
      "flos": 24530650824960.0,
      "grad_norm": 2.8139297328197723,
      "language_loss": 0.80505371,
      "learning_rate": 4.322518721063113e-07,
      "loss": 0.82804799,
      "num_input_tokens_seen": 142135505,
      "step": 6596,
      "time_per_iteration": 2.6597180366516113
    },
    {
      "auxiliary_loss_clip": 0.01223308,
      "auxiliary_loss_mlp": 0.01028551,
      "balance_loss_clip": 1.05057168,
      "balance_loss_mlp": 1.02126169,
      "epoch": 0.7932423495460831,
      "flos": 34421203434240.0,
      "grad_norm": 1.8903001131241843,
      "language_loss": 0.7005105,
      "learning_rate": 4.3176831207838906e-07,
      "loss": 0.72302902,
      "num_input_tokens_seen": 142158915,
      "step": 6597,
      "time_per_iteration": 2.721423625946045
    },
    {
      "auxiliary_loss_clip": 0.01219674,
      "auxiliary_loss_mlp": 0.01026241,
      "balance_loss_clip": 1.05002785,
      "balance_loss_mlp": 1.01915669,
      "epoch": 0.7933625924367221,
      "flos": 26980333441920.0,
      "grad_norm": 1.8130764961350527,
      "language_loss": 0.74653459,
      "learning_rate": 4.3128498994996685e-07,
      "loss": 0.76899374,
      "num_input_tokens_seen": 142178390,
      "step": 6598,
      "time_per_iteration": 2.6749420166015625
    },
    {
      "auxiliary_loss_clip": 0.0122911,
      "auxiliary_loss_mlp": 0.01030689,
      "balance_loss_clip": 1.05007553,
      "balance_loss_mlp": 1.02376342,
      "epoch": 0.7934828353273613,
      "flos": 29568630643200.0,
      "grad_norm": 4.0903082306213605,
      "language_loss": 0.71573693,
      "learning_rate": 4.308019057943646e-07,
      "loss": 0.73833489,
      "num_input_tokens_seen": 142200115,
      "step": 6599,
      "time_per_iteration": 2.6582255363464355
    },
    {
      "auxiliary_loss_clip": 0.01280857,
      "auxiliary_loss_mlp": 0.01028322,
      "balance_loss_clip": 1.04769373,
      "balance_loss_mlp": 1.02111244,
      "epoch": 0.7936030782180004,
      "flos": 28615381557120.0,
      "grad_norm": 2.0239458813922964,
      "language_loss": 0.7465328,
      "learning_rate": 4.3031905968486535e-07,
      "loss": 0.76962459,
      "num_input_tokens_seen": 142220945,
      "step": 6600,
      "time_per_iteration": 2.8021457195281982
    },
    {
      "auxiliary_loss_clip": 0.01368641,
      "auxiliary_loss_mlp": 0.01025909,
      "balance_loss_clip": 1.04901361,
      "balance_loss_mlp": 1.0197016,
      "epoch": 0.7937233211086394,
      "flos": 16392574869120.0,
      "grad_norm": 2.473529499068388,
      "language_loss": 0.68804109,
      "learning_rate": 4.298364516947162e-07,
      "loss": 0.71198654,
      "num_input_tokens_seen": 142238175,
      "step": 6601,
      "time_per_iteration": 2.6996238231658936
    },
    {
      "auxiliary_loss_clip": 0.01370838,
      "auxiliary_loss_mlp": 0.01025556,
      "balance_loss_clip": 1.04344296,
      "balance_loss_mlp": 1.01881433,
      "epoch": 0.7938435639992786,
      "flos": 22013420682240.0,
      "grad_norm": 2.262413517321873,
      "language_loss": 0.65668094,
      "learning_rate": 4.293540818971295e-07,
      "loss": 0.68064493,
      "num_input_tokens_seen": 142255980,
      "step": 6602,
      "time_per_iteration": 2.761214256286621
    },
    {
      "auxiliary_loss_clip": 0.01228737,
      "auxiliary_loss_mlp": 0.01025146,
      "balance_loss_clip": 1.04831731,
      "balance_loss_mlp": 1.01786506,
      "epoch": 0.7939638068899176,
      "flos": 22197032029440.0,
      "grad_norm": 2.0840774183567565,
      "language_loss": 0.76662505,
      "learning_rate": 4.2887195036527934e-07,
      "loss": 0.78916383,
      "num_input_tokens_seen": 142274785,
      "step": 6603,
      "time_per_iteration": 2.6247200965881348
    },
    {
      "auxiliary_loss_clip": 0.01218016,
      "auxiliary_loss_mlp": 0.01027163,
      "balance_loss_clip": 1.04480445,
      "balance_loss_mlp": 1.02007866,
      "epoch": 0.7940840497805567,
      "flos": 17745186343680.0,
      "grad_norm": 2.9665866810480988,
      "language_loss": 0.73440588,
      "learning_rate": 4.28390057172306e-07,
      "loss": 0.75685769,
      "num_input_tokens_seen": 142291290,
      "step": 6604,
      "time_per_iteration": 2.572831869125366
    },
    {
      "auxiliary_loss_clip": 0.01323138,
      "auxiliary_loss_mlp": 0.01028349,
      "balance_loss_clip": 1.04296446,
      "balance_loss_mlp": 1.02104473,
      "epoch": 0.7942042926711959,
      "flos": 23805435231360.0,
      "grad_norm": 2.3654628528850865,
      "language_loss": 0.7217024,
      "learning_rate": 4.279084023913111e-07,
      "loss": 0.74521726,
      "num_input_tokens_seen": 142309165,
      "step": 6605,
      "time_per_iteration": 2.7234859466552734
    },
    {
      "auxiliary_loss_clip": 0.01219588,
      "auxiliary_loss_mlp": 0.0102503,
      "balance_loss_clip": 1.04739165,
      "balance_loss_mlp": 1.01856291,
      "epoch": 0.7943245355618349,
      "flos": 19244959839360.0,
      "grad_norm": 2.102307108573259,
      "language_loss": 0.69564432,
      "learning_rate": 4.2742698609536096e-07,
      "loss": 0.71809053,
      "num_input_tokens_seen": 142327475,
      "step": 6606,
      "time_per_iteration": 3.502429962158203
    },
    {
      "auxiliary_loss_clip": 0.01276587,
      "auxiliary_loss_mlp": 0.01026034,
      "balance_loss_clip": 1.04922497,
      "balance_loss_mlp": 1.01915288,
      "epoch": 0.794444778452474,
      "flos": 25007616547200.0,
      "grad_norm": 1.8721397953749253,
      "language_loss": 0.78235471,
      "learning_rate": 4.2694580835748706e-07,
      "loss": 0.80538094,
      "num_input_tokens_seen": 142347335,
      "step": 6607,
      "time_per_iteration": 2.6978325843811035
    },
    {
      "auxiliary_loss_clip": 0.01173003,
      "auxiliary_loss_mlp": 0.0102559,
      "balance_loss_clip": 1.04549289,
      "balance_loss_mlp": 1.01867008,
      "epoch": 0.7945650213431131,
      "flos": 23221491828480.0,
      "grad_norm": 2.093957466592328,
      "language_loss": 0.7381168,
      "learning_rate": 4.264648692506836e-07,
      "loss": 0.76010275,
      "num_input_tokens_seen": 142366125,
      "step": 6608,
      "time_per_iteration": 3.6061301231384277
    },
    {
      "auxiliary_loss_clip": 0.01270341,
      "auxiliary_loss_mlp": 0.01029612,
      "balance_loss_clip": 1.04593027,
      "balance_loss_mlp": 1.02267432,
      "epoch": 0.7946852642337522,
      "flos": 26062887237120.0,
      "grad_norm": 2.1045714395347965,
      "language_loss": 0.72103018,
      "learning_rate": 4.2598416884790824e-07,
      "loss": 0.74402976,
      "num_input_tokens_seen": 142385175,
      "step": 6609,
      "time_per_iteration": 2.717087745666504
    },
    {
      "auxiliary_loss_clip": 0.01282507,
      "auxiliary_loss_mlp": 0.01028284,
      "balance_loss_clip": 1.046417,
      "balance_loss_mlp": 1.02091956,
      "epoch": 0.7948055071243912,
      "flos": 23769704177280.0,
      "grad_norm": 2.3802609052777948,
      "language_loss": 0.80832213,
      "learning_rate": 4.255037072220828e-07,
      "loss": 0.83143008,
      "num_input_tokens_seen": 142406545,
      "step": 6610,
      "time_per_iteration": 3.5911757946014404
    },
    {
      "auxiliary_loss_clip": 0.01171575,
      "auxiliary_loss_mlp": 0.0102383,
      "balance_loss_clip": 1.04793286,
      "balance_loss_mlp": 1.01736283,
      "epoch": 0.7949257500150304,
      "flos": 21980814111360.0,
      "grad_norm": 1.743181465224468,
      "language_loss": 0.7206012,
      "learning_rate": 4.2502348444609293e-07,
      "loss": 0.74255526,
      "num_input_tokens_seen": 142426165,
      "step": 6611,
      "time_per_iteration": 2.579723358154297
    },
    {
      "auxiliary_loss_clip": 0.01369767,
      "auxiliary_loss_mlp": 0.01024973,
      "balance_loss_clip": 1.04085243,
      "balance_loss_mlp": 1.01821399,
      "epoch": 0.7950459929056695,
      "flos": 25774129802880.0,
      "grad_norm": 1.6644626944516545,
      "language_loss": 0.69379443,
      "learning_rate": 4.2454350059278844e-07,
      "loss": 0.71774185,
      "num_input_tokens_seen": 142447225,
      "step": 6612,
      "time_per_iteration": 2.7681305408477783
    },
    {
      "auxiliary_loss_clip": 0.01271366,
      "auxiliary_loss_mlp": 0.01027396,
      "balance_loss_clip": 1.04311693,
      "balance_loss_mlp": 1.02088702,
      "epoch": 0.7951662357963085,
      "flos": 22158068751360.0,
      "grad_norm": 1.9127727093014686,
      "language_loss": 0.84556329,
      "learning_rate": 4.240637557349824e-07,
      "loss": 0.86855096,
      "num_input_tokens_seen": 142464440,
      "step": 6613,
      "time_per_iteration": 2.674767255783081
    },
    {
      "auxiliary_loss_clip": 0.01264726,
      "auxiliary_loss_mlp": 0.01024387,
      "balance_loss_clip": 1.0458076,
      "balance_loss_mlp": 1.01707923,
      "epoch": 0.7952864786869477,
      "flos": 24641938137600.0,
      "grad_norm": 1.7503328366054725,
      "language_loss": 0.66438103,
      "learning_rate": 4.235842499454516e-07,
      "loss": 0.68727219,
      "num_input_tokens_seen": 142484355,
      "step": 6614,
      "time_per_iteration": 2.66776967048645
    },
    {
      "auxiliary_loss_clip": 0.01275371,
      "auxiliary_loss_mlp": 0.01030002,
      "balance_loss_clip": 1.04821181,
      "balance_loss_mlp": 1.02300119,
      "epoch": 0.7954067215775867,
      "flos": 21830922656640.0,
      "grad_norm": 2.045718487402676,
      "language_loss": 0.82663572,
      "learning_rate": 4.2310498329693687e-07,
      "loss": 0.84968948,
      "num_input_tokens_seen": 142505255,
      "step": 6615,
      "time_per_iteration": 3.6162827014923096
    },
    {
      "auxiliary_loss_clip": 0.01225462,
      "auxiliary_loss_mlp": 0.01019866,
      "balance_loss_clip": 1.04974484,
      "balance_loss_mlp": 1.01238585,
      "epoch": 0.7955269644682258,
      "flos": 24060652341120.0,
      "grad_norm": 1.8477882507103678,
      "language_loss": 0.80806947,
      "learning_rate": 4.2262595586214164e-07,
      "loss": 0.83052278,
      "num_input_tokens_seen": 142526350,
      "step": 6616,
      "time_per_iteration": 2.622640371322632
    },
    {
      "auxiliary_loss_clip": 0.01228029,
      "auxiliary_loss_mlp": 0.01028283,
      "balance_loss_clip": 1.05001426,
      "balance_loss_mlp": 1.02125823,
      "epoch": 0.795647207358865,
      "flos": 25010741030400.0,
      "grad_norm": 1.7384081731205656,
      "language_loss": 0.77016824,
      "learning_rate": 4.221471677137358e-07,
      "loss": 0.79273134,
      "num_input_tokens_seen": 142547165,
      "step": 6617,
      "time_per_iteration": 2.6558423042297363
    },
    {
      "auxiliary_loss_clip": 0.01263528,
      "auxiliary_loss_mlp": 0.01025889,
      "balance_loss_clip": 1.04594338,
      "balance_loss_mlp": 1.01941633,
      "epoch": 0.795767450249504,
      "flos": 14648358343680.0,
      "grad_norm": 2.3603069590725108,
      "language_loss": 0.70073247,
      "learning_rate": 4.216686189243492e-07,
      "loss": 0.72362661,
      "num_input_tokens_seen": 142565955,
      "step": 6618,
      "time_per_iteration": 2.6570870876312256
    },
    {
      "auxiliary_loss_clip": 0.01316861,
      "auxiliary_loss_mlp": 0.01025414,
      "balance_loss_clip": 1.04487669,
      "balance_loss_mlp": 1.01807094,
      "epoch": 0.7958876931401431,
      "flos": 18547897530240.0,
      "grad_norm": 1.5881220212206542,
      "language_loss": 0.72821677,
      "learning_rate": 4.211903095665785e-07,
      "loss": 0.75163954,
      "num_input_tokens_seen": 142585340,
      "step": 6619,
      "time_per_iteration": 2.6535756587982178
    },
    {
      "auxiliary_loss_clip": 0.01218565,
      "auxiliary_loss_mlp": 0.01027302,
      "balance_loss_clip": 1.04776609,
      "balance_loss_mlp": 1.0208137,
      "epoch": 0.7960079360307821,
      "flos": 21543960902400.0,
      "grad_norm": 1.966605937756381,
      "language_loss": 0.75288701,
      "learning_rate": 4.2071223971298277e-07,
      "loss": 0.77534568,
      "num_input_tokens_seen": 142602525,
      "step": 6620,
      "time_per_iteration": 2.5839157104492188
    },
    {
      "auxiliary_loss_clip": 0.01226376,
      "auxiliary_loss_mlp": 0.01023221,
      "balance_loss_clip": 1.04744124,
      "balance_loss_mlp": 1.01574063,
      "epoch": 0.7961281789214213,
      "flos": 25481745095040.0,
      "grad_norm": 2.6976608247297125,
      "language_loss": 0.61285633,
      "learning_rate": 4.2023440943608433e-07,
      "loss": 0.63535225,
      "num_input_tokens_seen": 142622490,
      "step": 6621,
      "time_per_iteration": 2.6712915897369385
    },
    {
      "auxiliary_loss_clip": 0.01224174,
      "auxiliary_loss_mlp": 0.01025307,
      "balance_loss_clip": 1.0478586,
      "balance_loss_mlp": 1.01907492,
      "epoch": 0.7962484218120603,
      "flos": 21944436612480.0,
      "grad_norm": 1.5805080498653639,
      "language_loss": 0.78519738,
      "learning_rate": 4.1975681880837023e-07,
      "loss": 0.80769217,
      "num_input_tokens_seen": 142642495,
      "step": 6622,
      "time_per_iteration": 2.6104848384857178
    },
    {
      "auxiliary_loss_clip": 0.0132031,
      "auxiliary_loss_mlp": 0.01026032,
      "balance_loss_clip": 1.04225218,
      "balance_loss_mlp": 1.01907587,
      "epoch": 0.7963686647026994,
      "flos": 18876264687360.0,
      "grad_norm": 1.90683854846277,
      "language_loss": 0.8254177,
      "learning_rate": 4.192794679022895e-07,
      "loss": 0.84888113,
      "num_input_tokens_seen": 142660820,
      "step": 6623,
      "time_per_iteration": 2.6752336025238037
    },
    {
      "auxiliary_loss_clip": 0.01224466,
      "auxiliary_loss_mlp": 0.01030563,
      "balance_loss_clip": 1.04695058,
      "balance_loss_mlp": 1.02401555,
      "epoch": 0.7964889075933386,
      "flos": 29716582763520.0,
      "grad_norm": 2.0868180937107432,
      "language_loss": 0.72090715,
      "learning_rate": 4.1880235679025743e-07,
      "loss": 0.74345744,
      "num_input_tokens_seen": 142680915,
      "step": 6624,
      "time_per_iteration": 2.763941764831543
    },
    {
      "auxiliary_loss_clip": 0.01425457,
      "auxiliary_loss_mlp": 0.01029506,
      "balance_loss_clip": 1.04131508,
      "balance_loss_mlp": 1.02241611,
      "epoch": 0.7966091504839776,
      "flos": 29491458272640.0,
      "grad_norm": 1.8945800486188473,
      "language_loss": 0.63855994,
      "learning_rate": 4.1832548554464986e-07,
      "loss": 0.6631096,
      "num_input_tokens_seen": 142699210,
      "step": 6625,
      "time_per_iteration": 2.7469444274902344
    },
    {
      "auxiliary_loss_clip": 0.01110941,
      "auxiliary_loss_mlp": 0.01002413,
      "balance_loss_clip": 1.00888407,
      "balance_loss_mlp": 1.00149477,
      "epoch": 0.7967293933746167,
      "flos": 67288697101440.0,
      "grad_norm": 0.7400008932190706,
      "language_loss": 0.58670127,
      "learning_rate": 4.178488542378098e-07,
      "loss": 0.60783482,
      "num_input_tokens_seen": 142756790,
      "step": 6626,
      "time_per_iteration": 3.1250414848327637
    },
    {
      "auxiliary_loss_clip": 0.01177394,
      "auxiliary_loss_mlp": 0.01027547,
      "balance_loss_clip": 1.0508213,
      "balance_loss_mlp": 1.02012062,
      "epoch": 0.7968496362652558,
      "flos": 25554679660800.0,
      "grad_norm": 1.7748111154840907,
      "language_loss": 0.8926754,
      "learning_rate": 4.173724629420401e-07,
      "loss": 0.91472483,
      "num_input_tokens_seen": 142778150,
      "step": 6627,
      "time_per_iteration": 2.643479108810425
    },
    {
      "auxiliary_loss_clip": 0.01278687,
      "auxiliary_loss_mlp": 0.01025898,
      "balance_loss_clip": 1.04851544,
      "balance_loss_mlp": 1.01864374,
      "epoch": 0.7969698791558949,
      "flos": 14501088581760.0,
      "grad_norm": 3.7967914450180262,
      "language_loss": 0.68203473,
      "learning_rate": 4.168963117296087e-07,
      "loss": 0.70508051,
      "num_input_tokens_seen": 142795485,
      "step": 6628,
      "time_per_iteration": 2.627826690673828
    },
    {
      "auxiliary_loss_clip": 0.01174434,
      "auxiliary_loss_mlp": 0.01023694,
      "balance_loss_clip": 1.04960692,
      "balance_loss_mlp": 1.0167172,
      "epoch": 0.797090122046534,
      "flos": 22127545169280.0,
      "grad_norm": 2.388201585128572,
      "language_loss": 0.7612949,
      "learning_rate": 4.1642040067274876e-07,
      "loss": 0.78327614,
      "num_input_tokens_seen": 142815155,
      "step": 6629,
      "time_per_iteration": 2.6335997581481934
    },
    {
      "auxiliary_loss_clip": 0.0127716,
      "auxiliary_loss_mlp": 0.01026593,
      "balance_loss_clip": 1.04763412,
      "balance_loss_mlp": 1.01994419,
      "epoch": 0.7972103649371731,
      "flos": 19897671830400.0,
      "grad_norm": 1.8539583044423484,
      "language_loss": 0.72597766,
      "learning_rate": 4.1594472984365493e-07,
      "loss": 0.74901521,
      "num_input_tokens_seen": 142833840,
      "step": 6630,
      "time_per_iteration": 2.645946979522705
    },
    {
      "auxiliary_loss_clip": 0.01223043,
      "auxiliary_loss_mlp": 0.01026322,
      "balance_loss_clip": 1.04910481,
      "balance_loss_mlp": 1.01951849,
      "epoch": 0.7973306078278122,
      "flos": 36058621847040.0,
      "grad_norm": 1.9409853836155337,
      "language_loss": 0.77800435,
      "learning_rate": 4.154692993144862e-07,
      "loss": 0.80049801,
      "num_input_tokens_seen": 142853610,
      "step": 6631,
      "time_per_iteration": 2.6939005851745605
    },
    {
      "auxiliary_loss_clip": 0.0117495,
      "auxiliary_loss_mlp": 0.02564464,
      "balance_loss_clip": 1.04936337,
      "balance_loss_mlp": 0.99992561,
      "epoch": 0.7974508507184512,
      "flos": 21360600950400.0,
      "grad_norm": 2.415306032995808,
      "language_loss": 0.71782446,
      "learning_rate": 4.1499410915736476e-07,
      "loss": 0.75521851,
      "num_input_tokens_seen": 142872540,
      "step": 6632,
      "time_per_iteration": 3.5175857543945312
    },
    {
      "auxiliary_loss_clip": 0.01118608,
      "auxiliary_loss_mlp": 0.01000358,
      "balance_loss_clip": 1.0100981,
      "balance_loss_mlp": 0.99945164,
      "epoch": 0.7975710936090904,
      "flos": 68253115317120.0,
      "grad_norm": 0.7657919169831581,
      "language_loss": 0.64176095,
      "learning_rate": 4.145191594443762e-07,
      "loss": 0.66295058,
      "num_input_tokens_seen": 142936895,
      "step": 6633,
      "time_per_iteration": 4.259446859359741
    },
    {
      "auxiliary_loss_clip": 0.01320621,
      "auxiliary_loss_mlp": 0.01029296,
      "balance_loss_clip": 1.04523683,
      "balance_loss_mlp": 1.02195585,
      "epoch": 0.7976913364997295,
      "flos": 22492433479680.0,
      "grad_norm": 1.7989134932137414,
      "language_loss": 0.70771545,
      "learning_rate": 4.140444502475713e-07,
      "loss": 0.73121464,
      "num_input_tokens_seen": 142956445,
      "step": 6634,
      "time_per_iteration": 2.689101457595825
    },
    {
      "auxiliary_loss_clip": 0.01225242,
      "auxiliary_loss_mlp": 0.01026118,
      "balance_loss_clip": 1.04793358,
      "balance_loss_mlp": 1.0192219,
      "epoch": 0.7978115793903685,
      "flos": 15263220378240.0,
      "grad_norm": 6.428747355059764,
      "language_loss": 0.70235312,
      "learning_rate": 4.1356998163896216e-07,
      "loss": 0.72486669,
      "num_input_tokens_seen": 142973495,
      "step": 6635,
      "time_per_iteration": 3.494041919708252
    },
    {
      "auxiliary_loss_clip": 0.01325978,
      "auxiliary_loss_mlp": 0.01025083,
      "balance_loss_clip": 1.04678547,
      "balance_loss_mlp": 1.0180912,
      "epoch": 0.7979318222810077,
      "flos": 19719232041600.0,
      "grad_norm": 1.9914035101663885,
      "language_loss": 0.74774706,
      "learning_rate": 4.130957536905255e-07,
      "loss": 0.77125764,
      "num_input_tokens_seen": 142991510,
      "step": 6636,
      "time_per_iteration": 2.776224136352539
    },
    {
      "auxiliary_loss_clip": 0.01180638,
      "auxiliary_loss_mlp": 0.01029399,
      "balance_loss_clip": 1.04838133,
      "balance_loss_mlp": 1.02193308,
      "epoch": 0.7980520651716467,
      "flos": 15560273854080.0,
      "grad_norm": 2.739028165300597,
      "language_loss": 0.71560186,
      "learning_rate": 4.1262176647420134e-07,
      "loss": 0.73770225,
      "num_input_tokens_seen": 143009675,
      "step": 6637,
      "time_per_iteration": 2.611011266708374
    },
    {
      "auxiliary_loss_clip": 0.01280272,
      "auxiliary_loss_mlp": 0.01029503,
      "balance_loss_clip": 1.04872203,
      "balance_loss_mlp": 1.0229044,
      "epoch": 0.7981723080622858,
      "flos": 22309432663680.0,
      "grad_norm": 1.8033313589043347,
      "language_loss": 0.80202639,
      "learning_rate": 4.121480200618923e-07,
      "loss": 0.82512408,
      "num_input_tokens_seen": 143029330,
      "step": 6638,
      "time_per_iteration": 2.6852447986602783
    },
    {
      "auxiliary_loss_clip": 0.01272433,
      "auxiliary_loss_mlp": 0.01022561,
      "balance_loss_clip": 1.04737389,
      "balance_loss_mlp": 1.01519704,
      "epoch": 0.798292550952925,
      "flos": 22929573997440.0,
      "grad_norm": 2.055115061046622,
      "language_loss": 0.79873085,
      "learning_rate": 4.116745145254674e-07,
      "loss": 0.82168078,
      "num_input_tokens_seen": 143048865,
      "step": 6639,
      "time_per_iteration": 2.6521058082580566
    },
    {
      "auxiliary_loss_clip": 0.01162049,
      "auxiliary_loss_mlp": 0.01004236,
      "balance_loss_clip": 1.00692391,
      "balance_loss_mlp": 1.00331235,
      "epoch": 0.798412793843564,
      "flos": 64497936890880.0,
      "grad_norm": 0.7621568453102635,
      "language_loss": 0.57931942,
      "learning_rate": 4.1120124993675476e-07,
      "loss": 0.60098225,
      "num_input_tokens_seen": 143113295,
      "step": 6640,
      "time_per_iteration": 3.2148122787475586
    },
    {
      "auxiliary_loss_clip": 0.01279502,
      "auxiliary_loss_mlp": 0.01026292,
      "balance_loss_clip": 1.04547751,
      "balance_loss_mlp": 1.01870465,
      "epoch": 0.7985330367342031,
      "flos": 13586910514560.0,
      "grad_norm": 3.903859083734488,
      "language_loss": 0.61948192,
      "learning_rate": 4.107282263675498e-07,
      "loss": 0.64253986,
      "num_input_tokens_seen": 143130965,
      "step": 6641,
      "time_per_iteration": 3.600423812866211
    },
    {
      "auxiliary_loss_clip": 0.0116416,
      "auxiliary_loss_mlp": 0.0250768,
      "balance_loss_clip": 1.00711107,
      "balance_loss_mlp": 0.99990219,
      "epoch": 0.7986532796248422,
      "flos": 67698797656320.0,
      "grad_norm": 0.7605436544049067,
      "language_loss": 0.52438164,
      "learning_rate": 4.1025544388960907e-07,
      "loss": 0.56110007,
      "num_input_tokens_seen": 143192005,
      "step": 6642,
      "time_per_iteration": 3.210226058959961
    },
    {
      "auxiliary_loss_clip": 0.01219556,
      "auxiliary_loss_mlp": 0.01024942,
      "balance_loss_clip": 1.04898679,
      "balance_loss_mlp": 1.01764369,
      "epoch": 0.7987735225154813,
      "flos": 22455373622400.0,
      "grad_norm": 1.8230659555391795,
      "language_loss": 0.71447772,
      "learning_rate": 4.097829025746538e-07,
      "loss": 0.73692268,
      "num_input_tokens_seen": 143213550,
      "step": 6643,
      "time_per_iteration": 2.7129478454589844
    },
    {
      "auxiliary_loss_clip": 0.01113094,
      "auxiliary_loss_mlp": 0.01001707,
      "balance_loss_clip": 1.00831914,
      "balance_loss_mlp": 1.00082493,
      "epoch": 0.7988937654061203,
      "flos": 68864098682880.0,
      "grad_norm": 0.6576465258881536,
      "language_loss": 0.60959661,
      "learning_rate": 4.0931060249436757e-07,
      "loss": 0.63074458,
      "num_input_tokens_seen": 143277390,
      "step": 6644,
      "time_per_iteration": 3.213172674179077
    },
    {
      "auxiliary_loss_clip": 0.01224053,
      "auxiliary_loss_mlp": 0.0102444,
      "balance_loss_clip": 1.05177569,
      "balance_loss_mlp": 1.0173943,
      "epoch": 0.7990140082967595,
      "flos": 20806893820800.0,
      "grad_norm": 2.190067378099445,
      "language_loss": 0.69624674,
      "learning_rate": 4.088385437203978e-07,
      "loss": 0.7187317,
      "num_input_tokens_seen": 143294400,
      "step": 6645,
      "time_per_iteration": 2.612171173095703
    },
    {
      "auxiliary_loss_clip": 0.01174538,
      "auxiliary_loss_mlp": 0.01025156,
      "balance_loss_clip": 1.04871297,
      "balance_loss_mlp": 1.01786304,
      "epoch": 0.7991342511873986,
      "flos": 18985289443200.0,
      "grad_norm": 2.5030923088194967,
      "language_loss": 0.77768946,
      "learning_rate": 4.083667263243564e-07,
      "loss": 0.79968643,
      "num_input_tokens_seen": 143312745,
      "step": 6646,
      "time_per_iteration": 2.577841281890869
    },
    {
      "auxiliary_loss_clip": 0.01229794,
      "auxiliary_loss_mlp": 0.01024069,
      "balance_loss_clip": 1.05461478,
      "balance_loss_mlp": 1.01723552,
      "epoch": 0.7992544940780376,
      "flos": 20816805974400.0,
      "grad_norm": 6.1073038973103895,
      "language_loss": 0.71773988,
      "learning_rate": 4.0789515037781653e-07,
      "loss": 0.74027848,
      "num_input_tokens_seen": 143333470,
      "step": 6647,
      "time_per_iteration": 2.6117942333221436
    },
    {
      "auxiliary_loss_clip": 0.01228075,
      "auxiliary_loss_mlp": 0.01027397,
      "balance_loss_clip": 1.04964781,
      "balance_loss_mlp": 1.02036381,
      "epoch": 0.7993747369686768,
      "flos": 12640772321280.0,
      "grad_norm": 2.0826045737615586,
      "language_loss": 0.82670832,
      "learning_rate": 4.0742381595231755e-07,
      "loss": 0.84926307,
      "num_input_tokens_seen": 143350195,
      "step": 6648,
      "time_per_iteration": 2.608567476272583
    },
    {
      "auxiliary_loss_clip": 0.01225608,
      "auxiliary_loss_mlp": 0.01024164,
      "balance_loss_clip": 1.04751027,
      "balance_loss_mlp": 1.01773214,
      "epoch": 0.7994949798593158,
      "flos": 20078769225600.0,
      "grad_norm": 1.979064265243511,
      "language_loss": 0.78285515,
      "learning_rate": 4.06952723119359e-07,
      "loss": 0.80535287,
      "num_input_tokens_seen": 143370070,
      "step": 6649,
      "time_per_iteration": 2.6663105487823486
    },
    {
      "auxiliary_loss_clip": 0.01267842,
      "auxiliary_loss_mlp": 0.0103091,
      "balance_loss_clip": 1.04626429,
      "balance_loss_mlp": 1.02386165,
      "epoch": 0.7996152227499549,
      "flos": 38654209509120.0,
      "grad_norm": 2.0914517767057395,
      "language_loss": 0.67399764,
      "learning_rate": 4.0648187195040504e-07,
      "loss": 0.69698524,
      "num_input_tokens_seen": 143392275,
      "step": 6650,
      "time_per_iteration": 2.8233542442321777
    },
    {
      "auxiliary_loss_clip": 0.0111076,
      "auxiliary_loss_mlp": 0.01000927,
      "balance_loss_clip": 1.008865,
      "balance_loss_mlp": 0.99995595,
      "epoch": 0.799735465640594,
      "flos": 70243821947520.0,
      "grad_norm": 0.8163662007306148,
      "language_loss": 0.6755954,
      "learning_rate": 4.060112625168848e-07,
      "loss": 0.69671237,
      "num_input_tokens_seen": 143457385,
      "step": 6651,
      "time_per_iteration": 3.26642107963562
    },
    {
      "auxiliary_loss_clip": 0.01175563,
      "auxiliary_loss_mlp": 0.01026879,
      "balance_loss_clip": 1.04992712,
      "balance_loss_mlp": 1.01937473,
      "epoch": 0.7998557085312331,
      "flos": 24240995550720.0,
      "grad_norm": 2.50810137207009,
      "language_loss": 0.74187768,
      "learning_rate": 4.055408948901886e-07,
      "loss": 0.76390207,
      "num_input_tokens_seen": 143478785,
      "step": 6652,
      "time_per_iteration": 2.6128408908843994
    },
    {
      "auxiliary_loss_clip": 0.01229764,
      "auxiliary_loss_mlp": 0.01023126,
      "balance_loss_clip": 1.05042982,
      "balance_loss_mlp": 1.01582408,
      "epoch": 0.7999759514218722,
      "flos": 27564025449600.0,
      "grad_norm": 1.6871353641182028,
      "language_loss": 0.7134884,
      "learning_rate": 4.050707691416708e-07,
      "loss": 0.73601735,
      "num_input_tokens_seen": 143500095,
      "step": 6653,
      "time_per_iteration": 2.640617609024048
    },
    {
      "auxiliary_loss_clip": 0.01110367,
      "auxiliary_loss_mlp": 0.01000077,
      "balance_loss_clip": 1.00878263,
      "balance_loss_mlp": 0.99911112,
      "epoch": 0.8000961943125112,
      "flos": 67337428878720.0,
      "grad_norm": 0.6718449193686803,
      "language_loss": 0.59713471,
      "learning_rate": 4.046008853426495e-07,
      "loss": 0.61823916,
      "num_input_tokens_seen": 143563410,
      "step": 6654,
      "time_per_iteration": 3.255549192428589
    },
    {
      "auxiliary_loss_clip": 0.01320697,
      "auxiliary_loss_mlp": 0.01027164,
      "balance_loss_clip": 1.04562402,
      "balance_loss_mlp": 1.02001405,
      "epoch": 0.8002164372031504,
      "flos": 28733815676160.0,
      "grad_norm": 2.6039228488041477,
      "language_loss": 0.62786257,
      "learning_rate": 4.0413124356440464e-07,
      "loss": 0.6513412,
      "num_input_tokens_seen": 143587455,
      "step": 6655,
      "time_per_iteration": 2.764796733856201
    },
    {
      "auxiliary_loss_clip": 0.0137379,
      "auxiliary_loss_mlp": 0.01025469,
      "balance_loss_clip": 1.04234493,
      "balance_loss_mlp": 1.01818848,
      "epoch": 0.8003366800937894,
      "flos": 17639429725440.0,
      "grad_norm": 2.0684844634992903,
      "language_loss": 0.82283747,
      "learning_rate": 4.0366184387818223e-07,
      "loss": 0.84683007,
      "num_input_tokens_seen": 143605915,
      "step": 6656,
      "time_per_iteration": 2.712739944458008
    },
    {
      "auxiliary_loss_clip": 0.01181676,
      "auxiliary_loss_mlp": 0.01029171,
      "balance_loss_clip": 1.05184722,
      "balance_loss_mlp": 1.0215416,
      "epoch": 0.8004569229844285,
      "flos": 25995303797760.0,
      "grad_norm": 2.1752781907040517,
      "language_loss": 0.85272014,
      "learning_rate": 4.0319268635518797e-07,
      "loss": 0.87482858,
      "num_input_tokens_seen": 143626490,
      "step": 6657,
      "time_per_iteration": 2.679807424545288
    },
    {
      "auxiliary_loss_clip": 0.01224531,
      "auxiliary_loss_mlp": 0.01023936,
      "balance_loss_clip": 1.04716492,
      "balance_loss_mlp": 1.01765966,
      "epoch": 0.8005771658750677,
      "flos": 20812352688000.0,
      "grad_norm": 2.077679397144356,
      "language_loss": 0.75190496,
      "learning_rate": 4.027237710665943e-07,
      "loss": 0.77438962,
      "num_input_tokens_seen": 143644955,
      "step": 6658,
      "time_per_iteration": 3.537364959716797
    },
    {
      "auxiliary_loss_clip": 0.01324718,
      "auxiliary_loss_mlp": 0.01024142,
      "balance_loss_clip": 1.04380906,
      "balance_loss_mlp": 1.01702785,
      "epoch": 0.8006974087657067,
      "flos": 25812626204160.0,
      "grad_norm": 2.2185016798233037,
      "language_loss": 0.69555247,
      "learning_rate": 4.022550980835344e-07,
      "loss": 0.71904105,
      "num_input_tokens_seen": 143667200,
      "step": 6659,
      "time_per_iteration": 3.721189022064209
    },
    {
      "auxiliary_loss_clip": 0.0131925,
      "auxiliary_loss_mlp": 0.01029876,
      "balance_loss_clip": 1.04264593,
      "balance_loss_mlp": 1.02297986,
      "epoch": 0.8008176516563458,
      "flos": 17164690646400.0,
      "grad_norm": 2.719632408062194,
      "language_loss": 0.79483283,
      "learning_rate": 4.017866674771051e-07,
      "loss": 0.81832415,
      "num_input_tokens_seen": 143684685,
      "step": 6660,
      "time_per_iteration": 3.6000237464904785
    },
    {
      "auxiliary_loss_clip": 0.01366591,
      "auxiliary_loss_mlp": 0.01023351,
      "balance_loss_clip": 1.04038239,
      "balance_loss_mlp": 1.01648474,
      "epoch": 0.8009378945469849,
      "flos": 24207311571840.0,
      "grad_norm": 1.7156909884374745,
      "language_loss": 0.74671519,
      "learning_rate": 4.013184793183688e-07,
      "loss": 0.77061456,
      "num_input_tokens_seen": 143706780,
      "step": 6661,
      "time_per_iteration": 2.7734885215759277
    },
    {
      "auxiliary_loss_clip": 0.01223334,
      "auxiliary_loss_mlp": 0.01026537,
      "balance_loss_clip": 1.04674673,
      "balance_loss_mlp": 1.01933646,
      "epoch": 0.801058137437624,
      "flos": 19787318271360.0,
      "grad_norm": 3.2982898725232537,
      "language_loss": 0.72712266,
      "learning_rate": 4.008505336783472e-07,
      "loss": 0.74962133,
      "num_input_tokens_seen": 143724505,
      "step": 6662,
      "time_per_iteration": 2.6017673015594482
    },
    {
      "auxiliary_loss_clip": 0.01215862,
      "auxiliary_loss_mlp": 0.01026613,
      "balance_loss_clip": 1.04646027,
      "balance_loss_mlp": 1.02012515,
      "epoch": 0.801178380328263,
      "flos": 18659400324480.0,
      "grad_norm": 2.252747977353427,
      "language_loss": 0.81066442,
      "learning_rate": 4.003828306280284e-07,
      "loss": 0.83308917,
      "num_input_tokens_seen": 143742180,
      "step": 6663,
      "time_per_iteration": 2.6275711059570312
    },
    {
      "auxiliary_loss_clip": 0.01123803,
      "auxiliary_loss_mlp": 0.01025978,
      "balance_loss_clip": 1.04833269,
      "balance_loss_mlp": 1.0191977,
      "epoch": 0.8012986232189022,
      "flos": 15706573948800.0,
      "grad_norm": 2.8779595155325115,
      "language_loss": 0.78045571,
      "learning_rate": 3.999153702383626e-07,
      "loss": 0.80195349,
      "num_input_tokens_seen": 143760070,
      "step": 6664,
      "time_per_iteration": 2.6399126052856445
    },
    {
      "auxiliary_loss_clip": 0.01227654,
      "auxiliary_loss_mlp": 0.01029253,
      "balance_loss_clip": 1.04829574,
      "balance_loss_mlp": 1.02161145,
      "epoch": 0.8014188661095413,
      "flos": 28584139703040.0,
      "grad_norm": 7.573085870682205,
      "language_loss": 0.73866951,
      "learning_rate": 3.9944815258026263e-07,
      "loss": 0.76123852,
      "num_input_tokens_seen": 143781890,
      "step": 6665,
      "time_per_iteration": 2.7393569946289062
    },
    {
      "auxiliary_loss_clip": 0.01228402,
      "auxiliary_loss_mlp": 0.01023318,
      "balance_loss_clip": 1.04876339,
      "balance_loss_mlp": 1.01620138,
      "epoch": 0.8015391090001803,
      "flos": 29310360877440.0,
      "grad_norm": 1.7228681256034981,
      "language_loss": 0.8309164,
      "learning_rate": 3.989811777246057e-07,
      "loss": 0.85343367,
      "num_input_tokens_seen": 143802060,
      "step": 6666,
      "time_per_iteration": 2.7058985233306885
    },
    {
      "auxiliary_loss_clip": 0.01061862,
      "auxiliary_loss_mlp": 0.01002387,
      "balance_loss_clip": 1.00751519,
      "balance_loss_mlp": 1.00149298,
      "epoch": 0.8016593518908195,
      "flos": 70397340675840.0,
      "grad_norm": 0.8473748373653676,
      "language_loss": 0.66219085,
      "learning_rate": 3.985144457422305e-07,
      "loss": 0.68283337,
      "num_input_tokens_seen": 143856345,
      "step": 6667,
      "time_per_iteration": 4.068999767303467
    },
    {
      "auxiliary_loss_clip": 0.01174158,
      "auxiliary_loss_mlp": 0.01031047,
      "balance_loss_clip": 1.0494585,
      "balance_loss_mlp": 1.0246129,
      "epoch": 0.8017795947814585,
      "flos": 26026114688640.0,
      "grad_norm": 1.90450931134373,
      "language_loss": 0.76820219,
      "learning_rate": 3.9804795670394096e-07,
      "loss": 0.79025429,
      "num_input_tokens_seen": 143876470,
      "step": 6668,
      "time_per_iteration": 2.62266206741333
    },
    {
      "auxiliary_loss_clip": 0.01266984,
      "auxiliary_loss_mlp": 0.01026822,
      "balance_loss_clip": 1.04576886,
      "balance_loss_mlp": 1.0202055,
      "epoch": 0.8018998376720976,
      "flos": 22087181260800.0,
      "grad_norm": 2.0081702695733683,
      "language_loss": 0.70462835,
      "learning_rate": 3.975817106805022e-07,
      "loss": 0.72756642,
      "num_input_tokens_seen": 143895170,
      "step": 6669,
      "time_per_iteration": 2.675126791000366
    },
    {
      "auxiliary_loss_clip": 0.01323369,
      "auxiliary_loss_mlp": 0.01029517,
      "balance_loss_clip": 1.04639828,
      "balance_loss_mlp": 1.0223794,
      "epoch": 0.8020200805627368,
      "flos": 34568545023360.0,
      "grad_norm": 3.9007349378643292,
      "language_loss": 0.65277016,
      "learning_rate": 3.97115707742645e-07,
      "loss": 0.67629904,
      "num_input_tokens_seen": 143915845,
      "step": 6670,
      "time_per_iteration": 2.83388614654541
    },
    {
      "auxiliary_loss_clip": 0.01277045,
      "auxiliary_loss_mlp": 0.01026017,
      "balance_loss_clip": 1.04906726,
      "balance_loss_mlp": 1.01907301,
      "epoch": 0.8021403234533758,
      "flos": 20120354196480.0,
      "grad_norm": 2.290491276473115,
      "language_loss": 0.64973199,
      "learning_rate": 3.966499479610599e-07,
      "loss": 0.67276257,
      "num_input_tokens_seen": 143933940,
      "step": 6671,
      "time_per_iteration": 2.6538848876953125
    },
    {
      "auxiliary_loss_clip": 0.01320072,
      "auxiliary_loss_mlp": 0.01032208,
      "balance_loss_clip": 1.04748702,
      "balance_loss_mlp": 1.02514172,
      "epoch": 0.8022605663440149,
      "flos": 27746200252800.0,
      "grad_norm": 2.1138433176033056,
      "language_loss": 0.64854586,
      "learning_rate": 3.9618443140640225e-07,
      "loss": 0.67206872,
      "num_input_tokens_seen": 143952850,
      "step": 6672,
      "time_per_iteration": 2.7405736446380615
    },
    {
      "auxiliary_loss_clip": 0.01271239,
      "auxiliary_loss_mlp": 0.01002382,
      "balance_loss_clip": 1.00765753,
      "balance_loss_mlp": 1.00143993,
      "epoch": 0.802380809234654,
      "flos": 60244998768000.0,
      "grad_norm": 0.6864079553418202,
      "language_loss": 0.51274478,
      "learning_rate": 3.957191581492918e-07,
      "loss": 0.53548098,
      "num_input_tokens_seen": 144013610,
      "step": 6673,
      "time_per_iteration": 3.3522908687591553
    },
    {
      "auxiliary_loss_clip": 0.01268539,
      "auxiliary_loss_mlp": 0.01026439,
      "balance_loss_clip": 1.04578304,
      "balance_loss_mlp": 1.0199033,
      "epoch": 0.8025010521252931,
      "flos": 15080722352640.0,
      "grad_norm": 3.064995768141159,
      "language_loss": 0.71763325,
      "learning_rate": 3.952541282603097e-07,
      "loss": 0.740583,
      "num_input_tokens_seen": 144028715,
      "step": 6674,
      "time_per_iteration": 2.6060752868652344
    },
    {
      "auxiliary_loss_clip": 0.0122236,
      "auxiliary_loss_mlp": 0.01030035,
      "balance_loss_clip": 1.04912329,
      "balance_loss_mlp": 1.0231235,
      "epoch": 0.8026212950159322,
      "flos": 22163527618560.0,
      "grad_norm": 1.7378585788681622,
      "language_loss": 0.8368746,
      "learning_rate": 3.9478934181000013e-07,
      "loss": 0.85939854,
      "num_input_tokens_seen": 144048740,
      "step": 6675,
      "time_per_iteration": 2.6135292053222656
    },
    {
      "auxiliary_loss_clip": 0.01177958,
      "auxiliary_loss_mlp": 0.01030036,
      "balance_loss_clip": 1.04970694,
      "balance_loss_mlp": 1.02281189,
      "epoch": 0.8027415379065713,
      "flos": 17675986792320.0,
      "grad_norm": 2.4535129586636324,
      "language_loss": 0.849572,
      "learning_rate": 3.943247988688714e-07,
      "loss": 0.87165195,
      "num_input_tokens_seen": 144067435,
      "step": 6676,
      "time_per_iteration": 2.581153392791748
    },
    {
      "auxiliary_loss_clip": 0.01224932,
      "auxiliary_loss_mlp": 0.01028324,
      "balance_loss_clip": 1.04706287,
      "balance_loss_mlp": 1.02186036,
      "epoch": 0.8028617807972104,
      "flos": 21979593048960.0,
      "grad_norm": 1.8386457015330784,
      "language_loss": 0.71883476,
      "learning_rate": 3.938604995073933e-07,
      "loss": 0.74136734,
      "num_input_tokens_seen": 144085905,
      "step": 6677,
      "time_per_iteration": 2.620859384536743
    },
    {
      "auxiliary_loss_clip": 0.01276639,
      "auxiliary_loss_mlp": 0.01031638,
      "balance_loss_clip": 1.04733217,
      "balance_loss_mlp": 1.02453291,
      "epoch": 0.8029820236878494,
      "flos": 26428457905920.0,
      "grad_norm": 1.6823240748244546,
      "language_loss": 0.65520179,
      "learning_rate": 3.9339644379600157e-07,
      "loss": 0.67828453,
      "num_input_tokens_seen": 144105735,
      "step": 6678,
      "time_per_iteration": 2.705271005630493
    },
    {
      "auxiliary_loss_clip": 0.01126545,
      "auxiliary_loss_mlp": 0.0102702,
      "balance_loss_clip": 1.04982471,
      "balance_loss_mlp": 1.02082753,
      "epoch": 0.8031022665784886,
      "flos": 17676489582720.0,
      "grad_norm": 2.0691056825130656,
      "language_loss": 0.71219981,
      "learning_rate": 3.929326318050907e-07,
      "loss": 0.7337355,
      "num_input_tokens_seen": 144123405,
      "step": 6679,
      "time_per_iteration": 2.555757761001587
    },
    {
      "auxiliary_loss_clip": 0.01168699,
      "auxiliary_loss_mlp": 0.01022502,
      "balance_loss_clip": 1.04565287,
      "balance_loss_mlp": 1.01537895,
      "epoch": 0.8032225094691277,
      "flos": 15450279431040.0,
      "grad_norm": 3.9848858527152085,
      "language_loss": 0.79224223,
      "learning_rate": 3.924690636050225e-07,
      "loss": 0.81415427,
      "num_input_tokens_seen": 144140815,
      "step": 6680,
      "time_per_iteration": 2.582589864730835
    },
    {
      "auxiliary_loss_clip": 0.01226491,
      "auxiliary_loss_mlp": 0.01028038,
      "balance_loss_clip": 1.05047953,
      "balance_loss_mlp": 1.02095723,
      "epoch": 0.8033427523597667,
      "flos": 26179202453760.0,
      "grad_norm": 1.9071726687069162,
      "language_loss": 0.72933114,
      "learning_rate": 3.9200573926611915e-07,
      "loss": 0.75187641,
      "num_input_tokens_seen": 144162230,
      "step": 6681,
      "time_per_iteration": 2.643369436264038
    },
    {
      "auxiliary_loss_clip": 0.01228035,
      "auxiliary_loss_mlp": 0.01026734,
      "balance_loss_clip": 1.05186129,
      "balance_loss_mlp": 1.01984119,
      "epoch": 0.8034629952504058,
      "flos": 21324905809920.0,
      "grad_norm": 1.9788995699012273,
      "language_loss": 0.73229277,
      "learning_rate": 3.9154265885866613e-07,
      "loss": 0.75484037,
      "num_input_tokens_seen": 144181540,
      "step": 6682,
      "time_per_iteration": 2.6120049953460693
    },
    {
      "auxiliary_loss_clip": 0.01224329,
      "auxiliary_loss_mlp": 0.0101992,
      "balance_loss_clip": 1.04990959,
      "balance_loss_mlp": 1.01290774,
      "epoch": 0.8035832381410449,
      "flos": 21651585027840.0,
      "grad_norm": 3.3140809390015344,
      "language_loss": 0.74741828,
      "learning_rate": 3.9107982245291394e-07,
      "loss": 0.76986074,
      "num_input_tokens_seen": 144199665,
      "step": 6683,
      "time_per_iteration": 3.5193560123443604
    },
    {
      "auxiliary_loss_clip": 0.01323582,
      "auxiliary_loss_mlp": 0.01023884,
      "balance_loss_clip": 1.04888368,
      "balance_loss_mlp": 1.01702619,
      "epoch": 0.803703481031684,
      "flos": 20518818744960.0,
      "grad_norm": 1.9760793177861198,
      "language_loss": 0.7751323,
      "learning_rate": 3.9061723011907245e-07,
      "loss": 0.79860699,
      "num_input_tokens_seen": 144219020,
      "step": 6684,
      "time_per_iteration": 2.675445556640625
    },
    {
      "auxiliary_loss_clip": 0.01270675,
      "auxiliary_loss_mlp": 0.01026773,
      "balance_loss_clip": 1.04571128,
      "balance_loss_mlp": 1.01954246,
      "epoch": 0.803823723922323,
      "flos": 22854807838080.0,
      "grad_norm": 1.6904390241527318,
      "language_loss": 0.79710734,
      "learning_rate": 3.901548819273179e-07,
      "loss": 0.82008183,
      "num_input_tokens_seen": 144239035,
      "step": 6685,
      "time_per_iteration": 3.7043263912200928
    },
    {
      "auxiliary_loss_clip": 0.01224365,
      "auxiliary_loss_mlp": 0.01025597,
      "balance_loss_clip": 1.05022514,
      "balance_loss_mlp": 1.0189178,
      "epoch": 0.8039439668129622,
      "flos": 21362145235200.0,
      "grad_norm": 1.8765713946932534,
      "language_loss": 0.69552732,
      "learning_rate": 3.896927779477881e-07,
      "loss": 0.718027,
      "num_input_tokens_seen": 144258295,
      "step": 6686,
      "time_per_iteration": 3.559210777282715
    },
    {
      "auxiliary_loss_clip": 0.01323243,
      "auxiliary_loss_mlp": 0.01029568,
      "balance_loss_clip": 1.04458654,
      "balance_loss_mlp": 1.02230477,
      "epoch": 0.8040642097036013,
      "flos": 23802382575360.0,
      "grad_norm": 2.642549311986958,
      "language_loss": 0.66864729,
      "learning_rate": 3.892309182505833e-07,
      "loss": 0.69217539,
      "num_input_tokens_seen": 144276110,
      "step": 6687,
      "time_per_iteration": 2.7106645107269287
    },
    {
      "auxiliary_loss_clip": 0.01173367,
      "auxiliary_loss_mlp": 0.01030695,
      "balance_loss_clip": 1.04816151,
      "balance_loss_mlp": 1.02390933,
      "epoch": 0.8041844525942403,
      "flos": 25922046009600.0,
      "grad_norm": 2.21332321602218,
      "language_loss": 0.86257958,
      "learning_rate": 3.887693029057675e-07,
      "loss": 0.88462019,
      "num_input_tokens_seen": 144295620,
      "step": 6688,
      "time_per_iteration": 2.6855993270874023
    },
    {
      "auxiliary_loss_clip": 0.01273692,
      "auxiliary_loss_mlp": 0.010224,
      "balance_loss_clip": 1.04731083,
      "balance_loss_mlp": 1.01625168,
      "epoch": 0.8043046954848795,
      "flos": 25191120153600.0,
      "grad_norm": 2.064286687604893,
      "language_loss": 0.81299508,
      "learning_rate": 3.8830793198336684e-07,
      "loss": 0.83595604,
      "num_input_tokens_seen": 144315210,
      "step": 6689,
      "time_per_iteration": 2.6787054538726807
    },
    {
      "auxiliary_loss_clip": 0.01129374,
      "auxiliary_loss_mlp": 0.01025021,
      "balance_loss_clip": 1.0483166,
      "balance_loss_mlp": 1.01749003,
      "epoch": 0.8044249383755185,
      "flos": 41719185123840.0,
      "grad_norm": 1.6428371407147357,
      "language_loss": 0.70542252,
      "learning_rate": 3.878468055533721e-07,
      "loss": 0.7269665,
      "num_input_tokens_seen": 144337750,
      "step": 6690,
      "time_per_iteration": 2.7593507766723633
    },
    {
      "auxiliary_loss_clip": 0.01330424,
      "auxiliary_loss_mlp": 0.01029805,
      "balance_loss_clip": 1.04935503,
      "balance_loss_mlp": 1.02224064,
      "epoch": 0.8045451812661576,
      "flos": 20631434860800.0,
      "grad_norm": 3.2166834552961774,
      "language_loss": 0.84676993,
      "learning_rate": 3.8738592368573464e-07,
      "loss": 0.87037224,
      "num_input_tokens_seen": 144355305,
      "step": 6691,
      "time_per_iteration": 2.6781511306762695
    },
    {
      "auxiliary_loss_clip": 0.01318175,
      "auxiliary_loss_mlp": 0.01031225,
      "balance_loss_clip": 1.04557252,
      "balance_loss_mlp": 1.02364302,
      "epoch": 0.8046654241567968,
      "flos": 29711806254720.0,
      "grad_norm": 2.1436425266807313,
      "language_loss": 0.87982678,
      "learning_rate": 3.8692528645037137e-07,
      "loss": 0.90332079,
      "num_input_tokens_seen": 144374485,
      "step": 6692,
      "time_per_iteration": 3.6279428005218506
    },
    {
      "auxiliary_loss_clip": 0.01176137,
      "auxiliary_loss_mlp": 0.01026955,
      "balance_loss_clip": 1.05085754,
      "balance_loss_mlp": 1.01982331,
      "epoch": 0.8047856670474358,
      "flos": 17671389851520.0,
      "grad_norm": 2.2961588188690465,
      "language_loss": 0.7761085,
      "learning_rate": 3.8646489391715907e-07,
      "loss": 0.79813945,
      "num_input_tokens_seen": 144388780,
      "step": 6693,
      "time_per_iteration": 2.540644645690918
    },
    {
      "auxiliary_loss_clip": 0.01274589,
      "auxiliary_loss_mlp": 0.01026049,
      "balance_loss_clip": 1.04770219,
      "balance_loss_mlp": 1.01862228,
      "epoch": 0.8049059099380749,
      "flos": 17120699464320.0,
      "grad_norm": 2.556491785688585,
      "language_loss": 0.87671244,
      "learning_rate": 3.8600474615593903e-07,
      "loss": 0.89971888,
      "num_input_tokens_seen": 144403395,
      "step": 6694,
      "time_per_iteration": 2.658412456512451
    },
    {
      "auxiliary_loss_clip": 0.01224899,
      "auxiliary_loss_mlp": 0.00997848,
      "balance_loss_clip": 1.00913525,
      "balance_loss_mlp": 0.99694234,
      "epoch": 0.805026152828714,
      "flos": 62212903240320.0,
      "grad_norm": 0.7855348627458713,
      "language_loss": 0.59695101,
      "learning_rate": 3.8554484323651605e-07,
      "loss": 0.61917847,
      "num_input_tokens_seen": 144465265,
      "step": 6695,
      "time_per_iteration": 3.3200016021728516
    },
    {
      "auxiliary_loss_clip": 0.01221909,
      "auxiliary_loss_mlp": 0.02567393,
      "balance_loss_clip": 1.04903507,
      "balance_loss_mlp": 0.99991369,
      "epoch": 0.8051463957193531,
      "flos": 21688608971520.0,
      "grad_norm": 2.1219496683415513,
      "language_loss": 0.79087651,
      "learning_rate": 3.85085185228657e-07,
      "loss": 0.82876951,
      "num_input_tokens_seen": 144484235,
      "step": 6696,
      "time_per_iteration": 2.650770664215088
    },
    {
      "auxiliary_loss_clip": 0.01271407,
      "auxiliary_loss_mlp": 0.01037562,
      "balance_loss_clip": 1.04669213,
      "balance_loss_mlp": 1.02970862,
      "epoch": 0.8052666386099921,
      "flos": 32051458535040.0,
      "grad_norm": 1.8084300153142356,
      "language_loss": 0.7331264,
      "learning_rate": 3.8462577220209114e-07,
      "loss": 0.75621611,
      "num_input_tokens_seen": 144504610,
      "step": 6697,
      "time_per_iteration": 2.775339126586914
    },
    {
      "auxiliary_loss_clip": 0.01061798,
      "auxiliary_loss_mlp": 0.01001119,
      "balance_loss_clip": 1.00746858,
      "balance_loss_mlp": 1.00027835,
      "epoch": 0.8053868815006313,
      "flos": 67157875768320.0,
      "grad_norm": 0.7028846796817646,
      "language_loss": 0.58892626,
      "learning_rate": 3.8416660422651127e-07,
      "loss": 0.60955542,
      "num_input_tokens_seen": 144574260,
      "step": 6698,
      "time_per_iteration": 3.3232975006103516
    },
    {
      "auxiliary_loss_clip": 0.01323274,
      "auxiliary_loss_mlp": 0.01025342,
      "balance_loss_clip": 1.04471254,
      "balance_loss_mlp": 1.01827288,
      "epoch": 0.8055071243912704,
      "flos": 23837000307840.0,
      "grad_norm": 2.4568489614314855,
      "language_loss": 0.68362087,
      "learning_rate": 3.837076813715723e-07,
      "loss": 0.70710707,
      "num_input_tokens_seen": 144594145,
      "step": 6699,
      "time_per_iteration": 2.692138910293579
    },
    {
      "auxiliary_loss_clip": 0.01326135,
      "auxiliary_loss_mlp": 0.01027438,
      "balance_loss_clip": 1.044052,
      "balance_loss_mlp": 1.0196147,
      "epoch": 0.8056273672819094,
      "flos": 21324510760320.0,
      "grad_norm": 1.9673640699031927,
      "language_loss": 0.75169855,
      "learning_rate": 3.832490037068941e-07,
      "loss": 0.77523434,
      "num_input_tokens_seen": 144612935,
      "step": 6700,
      "time_per_iteration": 2.7322816848754883
    },
    {
      "auxiliary_loss_clip": 0.01416093,
      "auxiliary_loss_mlp": 0.01023069,
      "balance_loss_clip": 1.0406673,
      "balance_loss_mlp": 1.01616716,
      "epoch": 0.8057476101725486,
      "flos": 25768383626880.0,
      "grad_norm": 1.9310553554004837,
      "language_loss": 0.75962663,
      "learning_rate": 3.827905713020554e-07,
      "loss": 0.78401822,
      "num_input_tokens_seen": 144630580,
      "step": 6701,
      "time_per_iteration": 2.783207893371582
    },
    {
      "auxiliary_loss_clip": 0.01329968,
      "auxiliary_loss_mlp": 0.01027328,
      "balance_loss_clip": 1.04395461,
      "balance_loss_mlp": 1.01940644,
      "epoch": 0.8058678530631876,
      "flos": 24535283679360.0,
      "grad_norm": 2.063306038362226,
      "language_loss": 0.68780315,
      "learning_rate": 3.823323842266017e-07,
      "loss": 0.71137607,
      "num_input_tokens_seen": 144649975,
      "step": 6702,
      "time_per_iteration": 2.7427124977111816
    },
    {
      "auxiliary_loss_clip": 0.01224649,
      "auxiliary_loss_mlp": 0.01026523,
      "balance_loss_clip": 1.04568636,
      "balance_loss_mlp": 1.01974297,
      "epoch": 0.8059880959538267,
      "flos": 24753728240640.0,
      "grad_norm": 3.2559413595575486,
      "language_loss": 0.72751713,
      "learning_rate": 3.818744425500393e-07,
      "loss": 0.75002885,
      "num_input_tokens_seen": 144667990,
      "step": 6703,
      "time_per_iteration": 2.658201217651367
    },
    {
      "auxiliary_loss_clip": 0.0132028,
      "auxiliary_loss_mlp": 0.0102536,
      "balance_loss_clip": 1.04381084,
      "balance_loss_mlp": 1.01916373,
      "epoch": 0.8061083388444659,
      "flos": 22196349671040.0,
      "grad_norm": 2.205149920412109,
      "language_loss": 0.80629587,
      "learning_rate": 3.8141674634183675e-07,
      "loss": 0.82975227,
      "num_input_tokens_seen": 144687020,
      "step": 6704,
      "time_per_iteration": 2.719902276992798
    },
    {
      "auxiliary_loss_clip": 0.0136823,
      "auxiliary_loss_mlp": 0.01024599,
      "balance_loss_clip": 1.04568398,
      "balance_loss_mlp": 1.01826036,
      "epoch": 0.8062285817351049,
      "flos": 30044195735040.0,
      "grad_norm": 2.2045857598090572,
      "language_loss": 0.6652922,
      "learning_rate": 3.809592956714278e-07,
      "loss": 0.68922049,
      "num_input_tokens_seen": 144710255,
      "step": 6705,
      "time_per_iteration": 2.7683217525482178
    },
    {
      "auxiliary_loss_clip": 0.01230733,
      "auxiliary_loss_mlp": 0.0102895,
      "balance_loss_clip": 1.05055904,
      "balance_loss_mlp": 1.02209878,
      "epoch": 0.806348824625744,
      "flos": 22782591544320.0,
      "grad_norm": 2.2554455792002024,
      "language_loss": 0.74552441,
      "learning_rate": 3.805020906082057e-07,
      "loss": 0.76812124,
      "num_input_tokens_seen": 144728830,
      "step": 6706,
      "time_per_iteration": 2.6473777294158936
    },
    {
      "auxiliary_loss_clip": 0.0127927,
      "auxiliary_loss_mlp": 0.01023683,
      "balance_loss_clip": 1.04907823,
      "balance_loss_mlp": 1.01662302,
      "epoch": 0.8064690675163831,
      "flos": 23404600385280.0,
      "grad_norm": 2.3776203582344544,
      "language_loss": 0.81115985,
      "learning_rate": 3.8004513122152917e-07,
      "loss": 0.83418941,
      "num_input_tokens_seen": 144747140,
      "step": 6707,
      "time_per_iteration": 2.6786985397338867
    },
    {
      "auxiliary_loss_clip": 0.01275706,
      "auxiliary_loss_mlp": 0.01027311,
      "balance_loss_clip": 1.05118132,
      "balance_loss_mlp": 1.02062082,
      "epoch": 0.8065893104070222,
      "flos": 24060903736320.0,
      "grad_norm": 1.8299307235722482,
      "language_loss": 0.67455614,
      "learning_rate": 3.79588417580718e-07,
      "loss": 0.6975863,
      "num_input_tokens_seen": 144765250,
      "step": 6708,
      "time_per_iteration": 2.6912784576416016
    },
    {
      "auxiliary_loss_clip": 0.01227523,
      "auxiliary_loss_mlp": 0.010299,
      "balance_loss_clip": 1.05048537,
      "balance_loss_mlp": 1.02324533,
      "epoch": 0.8067095532976613,
      "flos": 22305410340480.0,
      "grad_norm": 2.110510093263405,
      "language_loss": 0.7641902,
      "learning_rate": 3.791319497550558e-07,
      "loss": 0.78676438,
      "num_input_tokens_seen": 144783080,
      "step": 6709,
      "time_per_iteration": 2.6142446994781494
    },
    {
      "auxiliary_loss_clip": 0.01229974,
      "auxiliary_loss_mlp": 0.02565778,
      "balance_loss_clip": 1.0488354,
      "balance_loss_mlp": 0.99990618,
      "epoch": 0.8068297961883004,
      "flos": 17129498296320.0,
      "grad_norm": 2.7496591662259684,
      "language_loss": 0.7089771,
      "learning_rate": 3.78675727813788e-07,
      "loss": 0.74693465,
      "num_input_tokens_seen": 144800645,
      "step": 6710,
      "time_per_iteration": 3.608246088027954
    },
    {
      "auxiliary_loss_clip": 0.01271765,
      "auxiliary_loss_mlp": 0.01031183,
      "balance_loss_clip": 1.04735303,
      "balance_loss_mlp": 1.02321351,
      "epoch": 0.8069500390789395,
      "flos": 22018843635840.0,
      "grad_norm": 1.6852887430258563,
      "language_loss": 0.73441982,
      "learning_rate": 3.782197518261225e-07,
      "loss": 0.75744933,
      "num_input_tokens_seen": 144820085,
      "step": 6711,
      "time_per_iteration": 3.707763433456421
    },
    {
      "auxiliary_loss_clip": 0.0128033,
      "auxiliary_loss_mlp": 0.01023946,
      "balance_loss_clip": 1.04869843,
      "balance_loss_mlp": 1.0174849,
      "epoch": 0.8070702819695785,
      "flos": 19244241567360.0,
      "grad_norm": 3.2294617272704333,
      "language_loss": 0.95659709,
      "learning_rate": 3.777640218612319e-07,
      "loss": 0.97963989,
      "num_input_tokens_seen": 144838070,
      "step": 6712,
      "time_per_iteration": 2.6053528785705566
    },
    {
      "auxiliary_loss_clip": 0.01221033,
      "auxiliary_loss_mlp": 0.0102539,
      "balance_loss_clip": 1.04915273,
      "balance_loss_mlp": 1.01867819,
      "epoch": 0.8071905248602176,
      "flos": 21544320038400.0,
      "grad_norm": 2.189462378581128,
      "language_loss": 0.72319853,
      "learning_rate": 3.773085379882488e-07,
      "loss": 0.74566275,
      "num_input_tokens_seen": 144857125,
      "step": 6713,
      "time_per_iteration": 3.5437817573547363
    },
    {
      "auxiliary_loss_clip": 0.01224904,
      "auxiliary_loss_mlp": 0.02570352,
      "balance_loss_clip": 1.04626989,
      "balance_loss_mlp": 0.9999367,
      "epoch": 0.8073107677508568,
      "flos": 37268309105280.0,
      "grad_norm": 1.8790625951769118,
      "language_loss": 0.75948626,
      "learning_rate": 3.768533002762715e-07,
      "loss": 0.79743886,
      "num_input_tokens_seen": 144880660,
      "step": 6714,
      "time_per_iteration": 2.7446651458740234
    },
    {
      "auxiliary_loss_clip": 0.01274258,
      "auxiliary_loss_mlp": 0.01022432,
      "balance_loss_clip": 1.04436898,
      "balance_loss_mlp": 1.01528287,
      "epoch": 0.8074310106414958,
      "flos": 28366269759360.0,
      "grad_norm": 1.8039785923485971,
      "language_loss": 0.77228302,
      "learning_rate": 3.763983087943572e-07,
      "loss": 0.79524994,
      "num_input_tokens_seen": 144900050,
      "step": 6715,
      "time_per_iteration": 2.665480375289917
    },
    {
      "auxiliary_loss_clip": 0.01217919,
      "auxiliary_loss_mlp": 0.02565395,
      "balance_loss_clip": 1.0449475,
      "balance_loss_mlp": 0.99993658,
      "epoch": 0.8075512535321349,
      "flos": 24281646768000.0,
      "grad_norm": 1.8089576610851728,
      "language_loss": 0.81270599,
      "learning_rate": 3.759435636115282e-07,
      "loss": 0.85053915,
      "num_input_tokens_seen": 144920835,
      "step": 6716,
      "time_per_iteration": 2.7549901008605957
    },
    {
      "auxiliary_loss_clip": 0.01422111,
      "auxiliary_loss_mlp": 0.02567801,
      "balance_loss_clip": 1.04571116,
      "balance_loss_mlp": 0.99993819,
      "epoch": 0.807671496422774,
      "flos": 26030855283840.0,
      "grad_norm": 1.829771567291944,
      "language_loss": 0.72980082,
      "learning_rate": 3.7548906479676967e-07,
      "loss": 0.76969993,
      "num_input_tokens_seen": 144940430,
      "step": 6717,
      "time_per_iteration": 2.8344368934631348
    },
    {
      "auxiliary_loss_clip": 0.01226854,
      "auxiliary_loss_mlp": 0.01027566,
      "balance_loss_clip": 1.04662347,
      "balance_loss_mlp": 1.02005005,
      "epoch": 0.8077917393134131,
      "flos": 23730740899200.0,
      "grad_norm": 2.198076224325224,
      "language_loss": 0.71411264,
      "learning_rate": 3.7503481241902855e-07,
      "loss": 0.73665679,
      "num_input_tokens_seen": 144960405,
      "step": 6718,
      "time_per_iteration": 3.6452670097351074
    },
    {
      "auxiliary_loss_clip": 0.01272586,
      "auxiliary_loss_mlp": 0.02565141,
      "balance_loss_clip": 1.0452038,
      "balance_loss_mlp": 0.999897,
      "epoch": 0.8079119822040521,
      "flos": 18402028398720.0,
      "grad_norm": 2.2536268465887908,
      "language_loss": 0.80728453,
      "learning_rate": 3.745808065472145e-07,
      "loss": 0.84566182,
      "num_input_tokens_seen": 144977700,
      "step": 6719,
      "time_per_iteration": 2.6567273139953613
    },
    {
      "auxiliary_loss_clip": 0.01228662,
      "auxiliary_loss_mlp": 0.01026664,
      "balance_loss_clip": 1.0551542,
      "balance_loss_mlp": 1.02007425,
      "epoch": 0.8080322250946913,
      "flos": 23621787970560.0,
      "grad_norm": 1.9059482356878545,
      "language_loss": 0.76347792,
      "learning_rate": 3.741270472501994e-07,
      "loss": 0.78603125,
      "num_input_tokens_seen": 144998340,
      "step": 6720,
      "time_per_iteration": 2.7569518089294434
    },
    {
      "auxiliary_loss_clip": 0.01271179,
      "auxiliary_loss_mlp": 0.01025403,
      "balance_loss_clip": 1.0498842,
      "balance_loss_mlp": 1.01844168,
      "epoch": 0.8081524679853304,
      "flos": 22820692896000.0,
      "grad_norm": 1.7012332403831656,
      "language_loss": 0.72625244,
      "learning_rate": 3.736735345968183e-07,
      "loss": 0.74921834,
      "num_input_tokens_seen": 145017950,
      "step": 6721,
      "time_per_iteration": 2.679767608642578
    },
    {
      "auxiliary_loss_clip": 0.012269,
      "auxiliary_loss_mlp": 0.01027015,
      "balance_loss_clip": 1.05004907,
      "balance_loss_mlp": 1.0201633,
      "epoch": 0.8082727108759694,
      "flos": 17640004343040.0,
      "grad_norm": 1.54592283445129,
      "language_loss": 0.78853136,
      "learning_rate": 3.7322026865586986e-07,
      "loss": 0.81107044,
      "num_input_tokens_seen": 145036985,
      "step": 6722,
      "time_per_iteration": 2.600980281829834
    },
    {
      "auxiliary_loss_clip": 0.01233877,
      "auxiliary_loss_mlp": 0.0102075,
      "balance_loss_clip": 1.05298734,
      "balance_loss_mlp": 1.0137229,
      "epoch": 0.8083929537666086,
      "flos": 25958172113280.0,
      "grad_norm": 2.2704576563158314,
      "language_loss": 0.73497331,
      "learning_rate": 3.7276724949611206e-07,
      "loss": 0.7575196,
      "num_input_tokens_seen": 145057095,
      "step": 6723,
      "time_per_iteration": 2.6731224060058594
    },
    {
      "auxiliary_loss_clip": 0.01276017,
      "auxiliary_loss_mlp": 0.01029627,
      "balance_loss_clip": 1.04843092,
      "balance_loss_mlp": 1.02255774,
      "epoch": 0.8085131966572476,
      "flos": 27089178629760.0,
      "grad_norm": 2.0831425121011407,
      "language_loss": 0.75040948,
      "learning_rate": 3.723144771862694e-07,
      "loss": 0.77346599,
      "num_input_tokens_seen": 145077735,
      "step": 6724,
      "time_per_iteration": 2.6592373847961426
    },
    {
      "auxiliary_loss_clip": 0.01323559,
      "auxiliary_loss_mlp": 0.01026763,
      "balance_loss_clip": 1.04463053,
      "balance_loss_mlp": 1.01997972,
      "epoch": 0.8086334395478867,
      "flos": 23988543788160.0,
      "grad_norm": 3.036168602826041,
      "language_loss": 0.76976347,
      "learning_rate": 3.718619517950263e-07,
      "loss": 0.79326665,
      "num_input_tokens_seen": 145098330,
      "step": 6725,
      "time_per_iteration": 2.7119388580322266
    },
    {
      "auxiliary_loss_clip": 0.0117636,
      "auxiliary_loss_mlp": 0.01024558,
      "balance_loss_clip": 1.05155003,
      "balance_loss_mlp": 1.01761985,
      "epoch": 0.8087536824385259,
      "flos": 20405879406720.0,
      "grad_norm": 1.8445856520251995,
      "language_loss": 0.76755941,
      "learning_rate": 3.714096733910301e-07,
      "loss": 0.7895686,
      "num_input_tokens_seen": 145115855,
      "step": 6726,
      "time_per_iteration": 2.5486254692077637
    },
    {
      "auxiliary_loss_clip": 0.01133788,
      "auxiliary_loss_mlp": 0.01028834,
      "balance_loss_clip": 1.05199838,
      "balance_loss_mlp": 1.02103138,
      "epoch": 0.8088739253291649,
      "flos": 25919639798400.0,
      "grad_norm": 2.380040016646791,
      "language_loss": 0.70386159,
      "learning_rate": 3.709576420428926e-07,
      "loss": 0.72548783,
      "num_input_tokens_seen": 145136655,
      "step": 6727,
      "time_per_iteration": 2.6482186317443848
    },
    {
      "auxiliary_loss_clip": 0.01275954,
      "auxiliary_loss_mlp": 0.01021044,
      "balance_loss_clip": 1.04539704,
      "balance_loss_mlp": 1.01404309,
      "epoch": 0.808994168219804,
      "flos": 28402072640640.0,
      "grad_norm": 2.0900462055796827,
      "language_loss": 0.73420751,
      "learning_rate": 3.7050585781918463e-07,
      "loss": 0.75717747,
      "num_input_tokens_seen": 145156955,
      "step": 6728,
      "time_per_iteration": 2.676119327545166
    },
    {
      "auxiliary_loss_clip": 0.01228424,
      "auxiliary_loss_mlp": 0.01026739,
      "balance_loss_clip": 1.04943275,
      "balance_loss_mlp": 1.01964939,
      "epoch": 0.8091144111104431,
      "flos": 17421056991360.0,
      "grad_norm": 4.907000450009968,
      "language_loss": 0.68959165,
      "learning_rate": 3.700543207884428e-07,
      "loss": 0.71214336,
      "num_input_tokens_seen": 145173865,
      "step": 6729,
      "time_per_iteration": 2.596876621246338
    },
    {
      "auxiliary_loss_clip": 0.01219812,
      "auxiliary_loss_mlp": 0.01032576,
      "balance_loss_clip": 1.04890442,
      "balance_loss_mlp": 1.02579319,
      "epoch": 0.8092346540010822,
      "flos": 32153803361280.0,
      "grad_norm": 1.9138602362149857,
      "language_loss": 0.71070206,
      "learning_rate": 3.6960303101916466e-07,
      "loss": 0.73322594,
      "num_input_tokens_seen": 145193780,
      "step": 6730,
      "time_per_iteration": 2.7082972526550293
    },
    {
      "auxiliary_loss_clip": 0.01061364,
      "auxiliary_loss_mlp": 0.0250596,
      "balance_loss_clip": 1.00711143,
      "balance_loss_mlp": 0.99993342,
      "epoch": 0.8093548968917212,
      "flos": 58035093390720.0,
      "grad_norm": 0.7440441532809146,
      "language_loss": 0.55494654,
      "learning_rate": 3.6915198857981047e-07,
      "loss": 0.59061974,
      "num_input_tokens_seen": 145258980,
      "step": 6731,
      "time_per_iteration": 3.237215757369995
    },
    {
      "auxiliary_loss_clip": 0.01331164,
      "auxiliary_loss_mlp": 0.01032832,
      "balance_loss_clip": 1.04801464,
      "balance_loss_mlp": 1.0252142,
      "epoch": 0.8094751397823604,
      "flos": 27381599251200.0,
      "grad_norm": 11.660067049715586,
      "language_loss": 0.68193805,
      "learning_rate": 3.687011935388027e-07,
      "loss": 0.70557797,
      "num_input_tokens_seen": 145281875,
      "step": 6732,
      "time_per_iteration": 2.737178087234497
    },
    {
      "auxiliary_loss_clip": 0.0122341,
      "auxiliary_loss_mlp": 0.01024853,
      "balance_loss_clip": 1.04936385,
      "balance_loss_mlp": 1.01804936,
      "epoch": 0.8095953826729995,
      "flos": 24061083304320.0,
      "grad_norm": 2.1463714415549346,
      "language_loss": 0.72860175,
      "learning_rate": 3.6825064596452646e-07,
      "loss": 0.75108439,
      "num_input_tokens_seen": 145302220,
      "step": 6733,
      "time_per_iteration": 2.7720930576324463
    },
    {
      "auxiliary_loss_clip": 0.01226349,
      "auxiliary_loss_mlp": 0.01021994,
      "balance_loss_clip": 1.04792213,
      "balance_loss_mlp": 1.01528859,
      "epoch": 0.8097156255636385,
      "flos": 23951412103680.0,
      "grad_norm": 1.7058825741225743,
      "language_loss": 0.70394045,
      "learning_rate": 3.678003459253305e-07,
      "loss": 0.72642386,
      "num_input_tokens_seen": 145323070,
      "step": 6734,
      "time_per_iteration": 2.691985845565796
    },
    {
      "auxiliary_loss_clip": 0.0132025,
      "auxiliary_loss_mlp": 0.01025805,
      "balance_loss_clip": 1.04545355,
      "balance_loss_mlp": 1.01810384,
      "epoch": 0.8098358684542777,
      "flos": 21799142098560.0,
      "grad_norm": 2.9283684038263935,
      "language_loss": 0.74295235,
      "learning_rate": 3.673502934895236e-07,
      "loss": 0.76641291,
      "num_input_tokens_seen": 145342575,
      "step": 6735,
      "time_per_iteration": 2.674635410308838
    },
    {
      "auxiliary_loss_clip": 0.01061279,
      "auxiliary_loss_mlp": 0.01002343,
      "balance_loss_clip": 1.00694728,
      "balance_loss_mlp": 1.00147271,
      "epoch": 0.8099561113449167,
      "flos": 68809515966720.0,
      "grad_norm": 0.6832412664277885,
      "language_loss": 0.57925606,
      "learning_rate": 3.669004887253802e-07,
      "loss": 0.59989226,
      "num_input_tokens_seen": 145408865,
      "step": 6736,
      "time_per_iteration": 4.16372013092041
    },
    {
      "auxiliary_loss_clip": 0.01277691,
      "auxiliary_loss_mlp": 0.0102586,
      "balance_loss_clip": 1.04905748,
      "balance_loss_mlp": 1.01908851,
      "epoch": 0.8100763542355558,
      "flos": 23586056916480.0,
      "grad_norm": 1.8593513781166882,
      "language_loss": 0.79294515,
      "learning_rate": 3.664509317011335e-07,
      "loss": 0.81598067,
      "num_input_tokens_seen": 145429200,
      "step": 6737,
      "time_per_iteration": 3.621014356613159
    },
    {
      "auxiliary_loss_clip": 0.01223917,
      "auxiliary_loss_mlp": 0.01031142,
      "balance_loss_clip": 1.04985642,
      "balance_loss_mlp": 1.02354264,
      "epoch": 0.810196597126195,
      "flos": 31650408207360.0,
      "grad_norm": 1.7921676433308114,
      "language_loss": 0.73825002,
      "learning_rate": 3.6600162248498134e-07,
      "loss": 0.7608006,
      "num_input_tokens_seen": 145452830,
      "step": 6738,
      "time_per_iteration": 2.724585771560669
    },
    {
      "auxiliary_loss_clip": 0.0140432,
      "auxiliary_loss_mlp": 0.01026944,
      "balance_loss_clip": 1.03788924,
      "balance_loss_mlp": 1.02036047,
      "epoch": 0.810316840016834,
      "flos": 24900459298560.0,
      "grad_norm": 1.715926960072817,
      "language_loss": 0.76316071,
      "learning_rate": 3.6555256114508426e-07,
      "loss": 0.78747332,
      "num_input_tokens_seen": 145472625,
      "step": 6739,
      "time_per_iteration": 3.713599443435669
    },
    {
      "auxiliary_loss_clip": 0.01274728,
      "auxiliary_loss_mlp": 0.01027094,
      "balance_loss_clip": 1.04331934,
      "balance_loss_mlp": 1.02019453,
      "epoch": 0.8104370829074731,
      "flos": 27965003950080.0,
      "grad_norm": 1.9466318322342349,
      "language_loss": 0.72983217,
      "learning_rate": 3.651037477495642e-07,
      "loss": 0.75285035,
      "num_input_tokens_seen": 145494075,
      "step": 6740,
      "time_per_iteration": 2.716322898864746
    },
    {
      "auxiliary_loss_clip": 0.01172575,
      "auxiliary_loss_mlp": 0.01030649,
      "balance_loss_clip": 1.04741967,
      "balance_loss_mlp": 1.02286482,
      "epoch": 0.8105573257981122,
      "flos": 24640752988800.0,
      "grad_norm": 1.9325223112421896,
      "language_loss": 0.67974895,
      "learning_rate": 3.6465518236650584e-07,
      "loss": 0.70178121,
      "num_input_tokens_seen": 145514220,
      "step": 6741,
      "time_per_iteration": 2.652621269226074
    },
    {
      "auxiliary_loss_clip": 0.01319137,
      "auxiliary_loss_mlp": 0.01029019,
      "balance_loss_clip": 1.04353118,
      "balance_loss_mlp": 1.02242362,
      "epoch": 0.8106775686887513,
      "flos": 26358935132160.0,
      "grad_norm": 1.9586106310060565,
      "language_loss": 0.78429246,
      "learning_rate": 3.642068650639558e-07,
      "loss": 0.80777407,
      "num_input_tokens_seen": 145533965,
      "step": 6742,
      "time_per_iteration": 2.7407236099243164
    },
    {
      "auxiliary_loss_clip": 0.01267964,
      "auxiliary_loss_mlp": 0.01025376,
      "balance_loss_clip": 1.04211521,
      "balance_loss_mlp": 1.01870036,
      "epoch": 0.8107978115793903,
      "flos": 27271892136960.0,
      "grad_norm": 2.196608212390339,
      "language_loss": 0.64747512,
      "learning_rate": 3.6375879590992334e-07,
      "loss": 0.67040849,
      "num_input_tokens_seen": 145554310,
      "step": 6743,
      "time_per_iteration": 2.679072141647339
    },
    {
      "auxiliary_loss_clip": 0.01271897,
      "auxiliary_loss_mlp": 0.01026069,
      "balance_loss_clip": 1.04509425,
      "balance_loss_mlp": 1.01872563,
      "epoch": 0.8109180544700295,
      "flos": 24934322845440.0,
      "grad_norm": 1.8247672243784028,
      "language_loss": 0.81216884,
      "learning_rate": 3.6331097497238173e-07,
      "loss": 0.83514851,
      "num_input_tokens_seen": 145573755,
      "step": 6744,
      "time_per_iteration": 3.632927417755127
    },
    {
      "auxiliary_loss_clip": 0.01320193,
      "auxiliary_loss_mlp": 0.01024252,
      "balance_loss_clip": 1.04497349,
      "balance_loss_mlp": 1.01791859,
      "epoch": 0.8110382973606686,
      "flos": 21105383840640.0,
      "grad_norm": 2.434668965330586,
      "language_loss": 0.79763806,
      "learning_rate": 3.628634023192627e-07,
      "loss": 0.82108247,
      "num_input_tokens_seen": 145594000,
      "step": 6745,
      "time_per_iteration": 2.705348491668701
    },
    {
      "auxiliary_loss_clip": 0.01226865,
      "auxiliary_loss_mlp": 0.0102524,
      "balance_loss_clip": 1.04823291,
      "balance_loss_mlp": 1.01790226,
      "epoch": 0.8111585402513076,
      "flos": 15414081500160.0,
      "grad_norm": 3.8627257766080967,
      "language_loss": 0.75413805,
      "learning_rate": 3.624160780184644e-07,
      "loss": 0.77665913,
      "num_input_tokens_seen": 145611215,
      "step": 6746,
      "time_per_iteration": 2.5655324459075928
    },
    {
      "auxiliary_loss_clip": 0.01268915,
      "auxiliary_loss_mlp": 0.01028313,
      "balance_loss_clip": 1.04510093,
      "balance_loss_mlp": 1.02095222,
      "epoch": 0.8112787831419467,
      "flos": 24095736950400.0,
      "grad_norm": 1.9674641290776718,
      "language_loss": 0.74755454,
      "learning_rate": 3.6196900213784496e-07,
      "loss": 0.77052689,
      "num_input_tokens_seen": 145630530,
      "step": 6747,
      "time_per_iteration": 2.700800895690918
    },
    {
      "auxiliary_loss_clip": 0.01224173,
      "auxiliary_loss_mlp": 0.01024636,
      "balance_loss_clip": 1.04814315,
      "balance_loss_mlp": 1.01770735,
      "epoch": 0.8113990260325858,
      "flos": 20483374999680.0,
      "grad_norm": 2.567541989536935,
      "language_loss": 0.86882198,
      "learning_rate": 3.6152217474522527e-07,
      "loss": 0.8913101,
      "num_input_tokens_seen": 145647345,
      "step": 6748,
      "time_per_iteration": 2.613940715789795
    },
    {
      "auxiliary_loss_clip": 0.01227,
      "auxiliary_loss_mlp": 0.01025948,
      "balance_loss_clip": 1.05261719,
      "balance_loss_mlp": 1.01894462,
      "epoch": 0.8115192689232249,
      "flos": 24901141656960.0,
      "grad_norm": 1.7327727734602034,
      "language_loss": 0.72991067,
      "learning_rate": 3.6107559590838975e-07,
      "loss": 0.75244015,
      "num_input_tokens_seen": 145666330,
      "step": 6749,
      "time_per_iteration": 2.679319143295288
    },
    {
      "auxiliary_loss_clip": 0.01424747,
      "auxiliary_loss_mlp": 0.01028718,
      "balance_loss_clip": 1.04302263,
      "balance_loss_mlp": 1.02189589,
      "epoch": 0.811639511813864,
      "flos": 24057204635520.0,
      "grad_norm": 2.448550034621646,
      "language_loss": 0.66031092,
      "learning_rate": 3.606292656950822e-07,
      "loss": 0.68484557,
      "num_input_tokens_seen": 145684740,
      "step": 6750,
      "time_per_iteration": 2.7487778663635254
    },
    {
      "auxiliary_loss_clip": 0.01271503,
      "auxiliary_loss_mlp": 0.01029429,
      "balance_loss_clip": 1.04533494,
      "balance_loss_mlp": 1.02217829,
      "epoch": 0.8117597547045031,
      "flos": 23185150243200.0,
      "grad_norm": 3.0294185429645695,
      "language_loss": 0.8636533,
      "learning_rate": 3.601831841730121e-07,
      "loss": 0.8866626,
      "num_input_tokens_seen": 145702660,
      "step": 6751,
      "time_per_iteration": 2.668571710586548
    },
    {
      "auxiliary_loss_clip": 0.01220679,
      "auxiliary_loss_mlp": 0.0102885,
      "balance_loss_clip": 1.04835272,
      "balance_loss_mlp": 1.02214158,
      "epoch": 0.8118799975951422,
      "flos": 23040250778880.0,
      "grad_norm": 2.035044792363082,
      "language_loss": 0.72689068,
      "learning_rate": 3.5973735140984916e-07,
      "loss": 0.74938589,
      "num_input_tokens_seen": 145722830,
      "step": 6752,
      "time_per_iteration": 2.6420114040374756
    },
    {
      "auxiliary_loss_clip": 0.01268326,
      "auxiliary_loss_mlp": 0.02564357,
      "balance_loss_clip": 1.04353929,
      "balance_loss_mlp": 0.99990356,
      "epoch": 0.8120002404857812,
      "flos": 24639962889600.0,
      "grad_norm": 2.2564622032197534,
      "language_loss": 0.79740095,
      "learning_rate": 3.5929176747322607e-07,
      "loss": 0.83572769,
      "num_input_tokens_seen": 145741935,
      "step": 6753,
      "time_per_iteration": 2.8492257595062256
    },
    {
      "auxiliary_loss_clip": 0.01172196,
      "auxiliary_loss_mlp": 0.01002435,
      "balance_loss_clip": 1.00695169,
      "balance_loss_mlp": 1.00159454,
      "epoch": 0.8121204833764204,
      "flos": 57415742156160.0,
      "grad_norm": 0.807768289300806,
      "language_loss": 0.5614382,
      "learning_rate": 3.588464324307372e-07,
      "loss": 0.58318454,
      "num_input_tokens_seen": 145805560,
      "step": 6754,
      "time_per_iteration": 3.258282423019409
    },
    {
      "auxiliary_loss_clip": 0.01227258,
      "auxiliary_loss_mlp": 0.01024945,
      "balance_loss_clip": 1.04734242,
      "balance_loss_mlp": 1.0187614,
      "epoch": 0.8122407262670595,
      "flos": 19464589549440.0,
      "grad_norm": 1.723160496097561,
      "language_loss": 0.75894356,
      "learning_rate": 3.584013463499391e-07,
      "loss": 0.78146553,
      "num_input_tokens_seen": 145824180,
      "step": 6755,
      "time_per_iteration": 2.623723268508911
    },
    {
      "auxiliary_loss_clip": 0.01166135,
      "auxiliary_loss_mlp": 0.01002634,
      "balance_loss_clip": 1.00810313,
      "balance_loss_mlp": 1.00174034,
      "epoch": 0.8123609691576985,
      "flos": 56425325472000.0,
      "grad_norm": 0.7311105471600292,
      "language_loss": 0.64351213,
      "learning_rate": 3.579565092983521e-07,
      "loss": 0.66519982,
      "num_input_tokens_seen": 145885300,
      "step": 6756,
      "time_per_iteration": 3.0971193313598633
    },
    {
      "auxiliary_loss_clip": 0.01175015,
      "auxiliary_loss_mlp": 0.01027046,
      "balance_loss_clip": 1.0506289,
      "balance_loss_mlp": 1.01995277,
      "epoch": 0.8124812120483377,
      "flos": 20631973564800.0,
      "grad_norm": 1.855919411370266,
      "language_loss": 0.83887732,
      "learning_rate": 3.575119213434565e-07,
      "loss": 0.8608979,
      "num_input_tokens_seen": 145903815,
      "step": 6757,
      "time_per_iteration": 2.617391347885132
    },
    {
      "auxiliary_loss_clip": 0.01221061,
      "auxiliary_loss_mlp": 0.01023527,
      "balance_loss_clip": 1.04905581,
      "balance_loss_mlp": 1.01699114,
      "epoch": 0.8126014549389767,
      "flos": 22492397566080.0,
      "grad_norm": 1.8079067583546653,
      "language_loss": 0.81859404,
      "learning_rate": 3.5706758255269765e-07,
      "loss": 0.8410399,
      "num_input_tokens_seen": 145922270,
      "step": 6758,
      "time_per_iteration": 2.6401748657226562
    },
    {
      "auxiliary_loss_clip": 0.01276002,
      "auxiliary_loss_mlp": 0.01027844,
      "balance_loss_clip": 1.04784179,
      "balance_loss_mlp": 1.02078974,
      "epoch": 0.8127216978296158,
      "flos": 23287961946240.0,
      "grad_norm": 1.7255443293689678,
      "language_loss": 0.69684076,
      "learning_rate": 3.566234929934795e-07,
      "loss": 0.71987915,
      "num_input_tokens_seen": 145941470,
      "step": 6759,
      "time_per_iteration": 2.6376864910125732
    },
    {
      "auxiliary_loss_clip": 0.01229396,
      "auxiliary_loss_mlp": 0.01023961,
      "balance_loss_clip": 1.05413699,
      "balance_loss_mlp": 1.01736927,
      "epoch": 0.812841940720255,
      "flos": 25154994049920.0,
      "grad_norm": 1.475941635379703,
      "language_loss": 0.71704096,
      "learning_rate": 3.561796527331706e-07,
      "loss": 0.73957455,
      "num_input_tokens_seen": 145963145,
      "step": 6760,
      "time_per_iteration": 2.7162322998046875
    },
    {
      "auxiliary_loss_clip": 0.01323051,
      "auxiliary_loss_mlp": 0.01025414,
      "balance_loss_clip": 1.0453918,
      "balance_loss_mlp": 1.01790643,
      "epoch": 0.812962183610894,
      "flos": 26648446752000.0,
      "grad_norm": 1.9524382514215117,
      "language_loss": 0.77078235,
      "learning_rate": 3.5573606183910163e-07,
      "loss": 0.79426694,
      "num_input_tokens_seen": 145983150,
      "step": 6761,
      "time_per_iteration": 2.704862594604492
    },
    {
      "auxiliary_loss_clip": 0.01229688,
      "auxiliary_loss_mlp": 0.01027471,
      "balance_loss_clip": 1.04747331,
      "balance_loss_mlp": 1.01946282,
      "epoch": 0.8130824265015331,
      "flos": 24966965329920.0,
      "grad_norm": 2.0340169689240732,
      "language_loss": 0.79046547,
      "learning_rate": 3.5529272037856493e-07,
      "loss": 0.81303704,
      "num_input_tokens_seen": 146001365,
      "step": 6762,
      "time_per_iteration": 4.584100723266602
    },
    {
      "auxiliary_loss_clip": 0.01339189,
      "auxiliary_loss_mlp": 0.00998141,
      "balance_loss_clip": 1.00686836,
      "balance_loss_mlp": 0.99707395,
      "epoch": 0.8132026693921722,
      "flos": 67622918175360.0,
      "grad_norm": 0.7220809470047134,
      "language_loss": 0.53794348,
      "learning_rate": 3.548496284188149e-07,
      "loss": 0.56131673,
      "num_input_tokens_seen": 146061570,
      "step": 6763,
      "time_per_iteration": 3.3679988384246826
    },
    {
      "auxiliary_loss_clip": 0.01357641,
      "auxiliary_loss_mlp": 0.01019865,
      "balance_loss_clip": 1.04290926,
      "balance_loss_mlp": 1.01352894,
      "epoch": 0.8133229122828113,
      "flos": 19495149045120.0,
      "grad_norm": 2.2158868537263596,
      "language_loss": 0.79254532,
      "learning_rate": 3.544067860270681e-07,
      "loss": 0.81632042,
      "num_input_tokens_seen": 146079145,
      "step": 6764,
      "time_per_iteration": 2.740636110305786
    },
    {
      "auxiliary_loss_clip": 0.01327374,
      "auxiliary_loss_mlp": 0.01026795,
      "balance_loss_clip": 1.04626763,
      "balance_loss_mlp": 1.01936221,
      "epoch": 0.8134431551734503,
      "flos": 20668135582080.0,
      "grad_norm": 1.812218395000123,
      "language_loss": 0.71080744,
      "learning_rate": 3.539641932705029e-07,
      "loss": 0.73434913,
      "num_input_tokens_seen": 146097625,
      "step": 6765,
      "time_per_iteration": 3.635033369064331
    },
    {
      "auxiliary_loss_clip": 0.01177891,
      "auxiliary_loss_mlp": 0.01027066,
      "balance_loss_clip": 1.05015099,
      "balance_loss_mlp": 1.01927018,
      "epoch": 0.8135633980640895,
      "flos": 21507332008320.0,
      "grad_norm": 2.1131955589591396,
      "language_loss": 0.77307045,
      "learning_rate": 3.53521850216262e-07,
      "loss": 0.79512,
      "num_input_tokens_seen": 146117195,
      "step": 6766,
      "time_per_iteration": 2.637220621109009
    },
    {
      "auxiliary_loss_clip": 0.01172677,
      "auxiliary_loss_mlp": 0.0102872,
      "balance_loss_clip": 1.04847062,
      "balance_loss_mlp": 1.02167511,
      "epoch": 0.8136836409547286,
      "flos": 20554442058240.0,
      "grad_norm": 2.486708687033299,
      "language_loss": 0.77097189,
      "learning_rate": 3.530797569314461e-07,
      "loss": 0.79298592,
      "num_input_tokens_seen": 146136220,
      "step": 6767,
      "time_per_iteration": 2.5604825019836426
    },
    {
      "auxiliary_loss_clip": 0.01176135,
      "auxiliary_loss_mlp": 0.01028084,
      "balance_loss_clip": 1.05061698,
      "balance_loss_mlp": 1.0201447,
      "epoch": 0.8138038838453676,
      "flos": 20299045380480.0,
      "grad_norm": 1.9298000522681296,
      "language_loss": 0.77664232,
      "learning_rate": 3.5263791348312235e-07,
      "loss": 0.79868454,
      "num_input_tokens_seen": 146155415,
      "step": 6768,
      "time_per_iteration": 3.047980785369873
    },
    {
      "auxiliary_loss_clip": 0.01272442,
      "auxiliary_loss_mlp": 0.01025625,
      "balance_loss_clip": 1.04658723,
      "balance_loss_mlp": 1.01870179,
      "epoch": 0.8139241267360068,
      "flos": 29789840551680.0,
      "grad_norm": 2.0407594339944835,
      "language_loss": 0.709113,
      "learning_rate": 3.521963199383171e-07,
      "loss": 0.73209363,
      "num_input_tokens_seen": 146178370,
      "step": 6769,
      "time_per_iteration": 2.71441650390625
    },
    {
      "auxiliary_loss_clip": 0.01376707,
      "auxiliary_loss_mlp": 0.01030388,
      "balance_loss_clip": 1.04448032,
      "balance_loss_mlp": 1.02231705,
      "epoch": 0.8140443696266458,
      "flos": 19713270384000.0,
      "grad_norm": 2.8768263922829584,
      "language_loss": 0.77067089,
      "learning_rate": 3.517549763640197e-07,
      "loss": 0.79474175,
      "num_input_tokens_seen": 146196010,
      "step": 6770,
      "time_per_iteration": 3.617910385131836
    },
    {
      "auxiliary_loss_clip": 0.01220125,
      "auxiliary_loss_mlp": 0.02563986,
      "balance_loss_clip": 1.04991007,
      "balance_loss_mlp": 0.99990332,
      "epoch": 0.8141646125172849,
      "flos": 27160568910720.0,
      "grad_norm": 1.8053192726570528,
      "language_loss": 0.71464515,
      "learning_rate": 3.513138828271829e-07,
      "loss": 0.75248629,
      "num_input_tokens_seen": 146215880,
      "step": 6771,
      "time_per_iteration": 2.6727659702301025
    },
    {
      "auxiliary_loss_clip": 0.01324474,
      "auxiliary_loss_mlp": 0.01025463,
      "balance_loss_clip": 1.04746485,
      "balance_loss_mlp": 1.01878715,
      "epoch": 0.8142848554079241,
      "flos": 39673102700160.0,
      "grad_norm": 1.837374695719706,
      "language_loss": 0.70008832,
      "learning_rate": 3.508730393947179e-07,
      "loss": 0.72358775,
      "num_input_tokens_seen": 146239135,
      "step": 6772,
      "time_per_iteration": 2.890540838241577
    },
    {
      "auxiliary_loss_clip": 0.01316292,
      "auxiliary_loss_mlp": 0.0102359,
      "balance_loss_clip": 1.04476047,
      "balance_loss_mlp": 1.01617217,
      "epoch": 0.8144050982985631,
      "flos": 22237288197120.0,
      "grad_norm": 2.7333969594044487,
      "language_loss": 0.72511154,
      "learning_rate": 3.504324461335024e-07,
      "loss": 0.74851036,
      "num_input_tokens_seen": 146259245,
      "step": 6773,
      "time_per_iteration": 2.730473756790161
    },
    {
      "auxiliary_loss_clip": 0.0137639,
      "auxiliary_loss_mlp": 0.01032456,
      "balance_loss_clip": 1.04496098,
      "balance_loss_mlp": 1.02487707,
      "epoch": 0.8145253411892022,
      "flos": 23038239617280.0,
      "grad_norm": 3.2124667549613712,
      "language_loss": 0.88146311,
      "learning_rate": 3.499921031103732e-07,
      "loss": 0.90555155,
      "num_input_tokens_seen": 146280015,
      "step": 6774,
      "time_per_iteration": 2.7828454971313477
    },
    {
      "auxiliary_loss_clip": 0.01233169,
      "auxiliary_loss_mlp": 0.0102625,
      "balance_loss_clip": 1.04377079,
      "balance_loss_mlp": 1.01950288,
      "epoch": 0.8146455840798413,
      "flos": 24827668387200.0,
      "grad_norm": 1.7568928587914643,
      "language_loss": 0.78569055,
      "learning_rate": 3.4955201039212987e-07,
      "loss": 0.80828476,
      "num_input_tokens_seen": 146300935,
      "step": 6775,
      "time_per_iteration": 2.770789861679077
    },
    {
      "auxiliary_loss_clip": 0.01129257,
      "auxiliary_loss_mlp": 0.01023544,
      "balance_loss_clip": 1.04987967,
      "balance_loss_mlp": 1.01675749,
      "epoch": 0.8147658269704804,
      "flos": 19974520978560.0,
      "grad_norm": 10.62729516117895,
      "language_loss": 0.65713966,
      "learning_rate": 3.4911216804553465e-07,
      "loss": 0.67866766,
      "num_input_tokens_seen": 146319835,
      "step": 6776,
      "time_per_iteration": 2.596358299255371
    },
    {
      "auxiliary_loss_clip": 0.01273434,
      "auxiliary_loss_mlp": 0.01028116,
      "balance_loss_clip": 1.04493022,
      "balance_loss_mlp": 1.0205698,
      "epoch": 0.8148860698611194,
      "flos": 21178031097600.0,
      "grad_norm": 2.205118681827262,
      "language_loss": 0.70167303,
      "learning_rate": 3.4867257613731017e-07,
      "loss": 0.72468853,
      "num_input_tokens_seen": 146339030,
      "step": 6777,
      "time_per_iteration": 2.6332342624664307
    },
    {
      "auxiliary_loss_clip": 0.01275443,
      "auxiliary_loss_mlp": 0.01025489,
      "balance_loss_clip": 1.04757369,
      "balance_loss_mlp": 1.01841116,
      "epoch": 0.8150063127517585,
      "flos": 19606903234560.0,
      "grad_norm": 1.9236986557132842,
      "language_loss": 0.85620487,
      "learning_rate": 3.4823323473414343e-07,
      "loss": 0.87921423,
      "num_input_tokens_seen": 146358550,
      "step": 6778,
      "time_per_iteration": 2.6682279109954834
    },
    {
      "auxiliary_loss_clip": 0.01228616,
      "auxiliary_loss_mlp": 0.01022119,
      "balance_loss_clip": 1.04500151,
      "balance_loss_mlp": 1.01377106,
      "epoch": 0.8151265556423977,
      "flos": 22638374438400.0,
      "grad_norm": 1.9159517263033041,
      "language_loss": 0.75863838,
      "learning_rate": 3.477941439026812e-07,
      "loss": 0.78114569,
      "num_input_tokens_seen": 146376770,
      "step": 6779,
      "time_per_iteration": 2.7625417709350586
    },
    {
      "auxiliary_loss_clip": 0.01174154,
      "auxiliary_loss_mlp": 0.01026441,
      "balance_loss_clip": 1.04895735,
      "balance_loss_mlp": 1.02025151,
      "epoch": 0.8152467985330367,
      "flos": 17968048277760.0,
      "grad_norm": 2.6744062831103066,
      "language_loss": 0.73271513,
      "learning_rate": 3.473553037095349e-07,
      "loss": 0.75472105,
      "num_input_tokens_seen": 146395795,
      "step": 6780,
      "time_per_iteration": 2.6668953895568848
    },
    {
      "auxiliary_loss_clip": 0.01271108,
      "auxiliary_loss_mlp": 0.01026805,
      "balance_loss_clip": 1.04453123,
      "balance_loss_mlp": 1.02053463,
      "epoch": 0.8153670414236758,
      "flos": 24969012405120.0,
      "grad_norm": 1.682322580232965,
      "language_loss": 0.83466876,
      "learning_rate": 3.469167142212743e-07,
      "loss": 0.8576479,
      "num_input_tokens_seen": 146417640,
      "step": 6781,
      "time_per_iteration": 2.6714727878570557
    },
    {
      "auxiliary_loss_clip": 0.01225497,
      "auxiliary_loss_mlp": 0.0102851,
      "balance_loss_clip": 1.04896235,
      "balance_loss_mlp": 1.02073145,
      "epoch": 0.8154872843143149,
      "flos": 31066069754880.0,
      "grad_norm": 16.30964690890461,
      "language_loss": 0.63178509,
      "learning_rate": 3.4647837550443337e-07,
      "loss": 0.65432513,
      "num_input_tokens_seen": 146436205,
      "step": 6782,
      "time_per_iteration": 2.7289083003997803
    },
    {
      "auxiliary_loss_clip": 0.01322387,
      "auxiliary_loss_mlp": 0.01021485,
      "balance_loss_clip": 1.04607844,
      "balance_loss_mlp": 1.01466334,
      "epoch": 0.815607527204954,
      "flos": 19391654983680.0,
      "grad_norm": 2.1156903241986766,
      "language_loss": 0.74699688,
      "learning_rate": 3.460402876255086e-07,
      "loss": 0.77043557,
      "num_input_tokens_seen": 146453595,
      "step": 6783,
      "time_per_iteration": 2.650433301925659
    },
    {
      "auxiliary_loss_clip": 0.01225776,
      "auxiliary_loss_mlp": 0.01027111,
      "balance_loss_clip": 1.04773831,
      "balance_loss_mlp": 1.01989841,
      "epoch": 0.815727770095593,
      "flos": 26140418743680.0,
      "grad_norm": 2.279355468814845,
      "language_loss": 0.72410667,
      "learning_rate": 3.456024506509574e-07,
      "loss": 0.74663556,
      "num_input_tokens_seen": 146474515,
      "step": 6784,
      "time_per_iteration": 2.6543045043945312
    },
    {
      "auxiliary_loss_clip": 0.01224836,
      "auxiliary_loss_mlp": 0.02565564,
      "balance_loss_clip": 1.05012035,
      "balance_loss_mlp": 0.99992394,
      "epoch": 0.8158480129862322,
      "flos": 25337527989120.0,
      "grad_norm": 2.5142419479259144,
      "language_loss": 0.74116451,
      "learning_rate": 3.4516486464719873e-07,
      "loss": 0.77906847,
      "num_input_tokens_seen": 146493905,
      "step": 6785,
      "time_per_iteration": 2.654092788696289
    },
    {
      "auxiliary_loss_clip": 0.01367786,
      "auxiliary_loss_mlp": 0.01027355,
      "balance_loss_clip": 1.04124808,
      "balance_loss_mlp": 1.02034545,
      "epoch": 0.8159682558768713,
      "flos": 34423645559040.0,
      "grad_norm": 1.606137512856906,
      "language_loss": 0.62451601,
      "learning_rate": 3.4472752968061445e-07,
      "loss": 0.64846742,
      "num_input_tokens_seen": 146518335,
      "step": 6786,
      "time_per_iteration": 2.856773614883423
    },
    {
      "auxiliary_loss_clip": 0.01223805,
      "auxiliary_loss_mlp": 0.01023869,
      "balance_loss_clip": 1.04695678,
      "balance_loss_mlp": 1.01684153,
      "epoch": 0.8160884987675103,
      "flos": 18653223185280.0,
      "grad_norm": 1.9431368776640814,
      "language_loss": 0.74163115,
      "learning_rate": 3.442904458175475e-07,
      "loss": 0.76410794,
      "num_input_tokens_seen": 146535655,
      "step": 6787,
      "time_per_iteration": 2.5840935707092285
    },
    {
      "auxiliary_loss_clip": 0.0122376,
      "auxiliary_loss_mlp": 0.01026586,
      "balance_loss_clip": 1.04739952,
      "balance_loss_mlp": 1.01911712,
      "epoch": 0.8162087416581495,
      "flos": 31430527102080.0,
      "grad_norm": 20.961183858067592,
      "language_loss": 0.76250541,
      "learning_rate": 3.438536131243044e-07,
      "loss": 0.78500891,
      "num_input_tokens_seen": 146556815,
      "step": 6788,
      "time_per_iteration": 4.518171072006226
    },
    {
      "auxiliary_loss_clip": 0.0127263,
      "auxiliary_loss_mlp": 0.01026374,
      "balance_loss_clip": 1.04675889,
      "balance_loss_mlp": 1.01836872,
      "epoch": 0.8163289845487885,
      "flos": 37593910915200.0,
      "grad_norm": 2.19901717228257,
      "language_loss": 0.61776161,
      "learning_rate": 3.434170316671503e-07,
      "loss": 0.6407516,
      "num_input_tokens_seen": 146581845,
      "step": 6789,
      "time_per_iteration": 2.8035991191864014
    },
    {
      "auxiliary_loss_clip": 0.01317582,
      "auxiliary_loss_mlp": 0.01026682,
      "balance_loss_clip": 1.04978657,
      "balance_loss_mlp": 1.01963401,
      "epoch": 0.8164492274394276,
      "flos": 13953989554560.0,
      "grad_norm": 3.4084233023498456,
      "language_loss": 0.89938104,
      "learning_rate": 3.4298070151231583e-07,
      "loss": 0.92282373,
      "num_input_tokens_seen": 146597245,
      "step": 6790,
      "time_per_iteration": 2.6464617252349854
    },
    {
      "auxiliary_loss_clip": 0.01277738,
      "auxiliary_loss_mlp": 0.01028409,
      "balance_loss_clip": 1.04813957,
      "balance_loss_mlp": 1.02120626,
      "epoch": 0.8165694703300668,
      "flos": 28986554747520.0,
      "grad_norm": 2.009506368508766,
      "language_loss": 0.60220379,
      "learning_rate": 3.425446227259916e-07,
      "loss": 0.62526524,
      "num_input_tokens_seen": 146618210,
      "step": 6791,
      "time_per_iteration": 3.6455626487731934
    },
    {
      "auxiliary_loss_clip": 0.01274824,
      "auxiliary_loss_mlp": 0.01027022,
      "balance_loss_clip": 1.04717767,
      "balance_loss_mlp": 1.0200038,
      "epoch": 0.8166897132207058,
      "flos": 25118365155840.0,
      "grad_norm": 2.4247251860351695,
      "language_loss": 0.82351649,
      "learning_rate": 3.421087953743296e-07,
      "loss": 0.84653497,
      "num_input_tokens_seen": 146637975,
      "step": 6792,
      "time_per_iteration": 2.6589529514312744
    },
    {
      "auxiliary_loss_clip": 0.01226161,
      "auxiliary_loss_mlp": 0.01023945,
      "balance_loss_clip": 1.04737294,
      "balance_loss_mlp": 1.01658046,
      "epoch": 0.8168099561113449,
      "flos": 23148593176320.0,
      "grad_norm": 6.560661246336411,
      "language_loss": 0.80141181,
      "learning_rate": 3.416732195234464e-07,
      "loss": 0.82391286,
      "num_input_tokens_seen": 146658030,
      "step": 6793,
      "time_per_iteration": 2.6033036708831787
    },
    {
      "auxiliary_loss_clip": 0.01225632,
      "auxiliary_loss_mlp": 0.01026054,
      "balance_loss_clip": 1.04827166,
      "balance_loss_mlp": 1.01916957,
      "epoch": 0.816930199001984,
      "flos": 18407666833920.0,
      "grad_norm": 1.5221281931778758,
      "language_loss": 0.79397523,
      "learning_rate": 3.4123789523941613e-07,
      "loss": 0.81649208,
      "num_input_tokens_seen": 146677855,
      "step": 6794,
      "time_per_iteration": 2.6432907581329346
    },
    {
      "auxiliary_loss_clip": 0.01222049,
      "auxiliary_loss_mlp": 0.0102605,
      "balance_loss_clip": 1.04554021,
      "balance_loss_mlp": 1.01898098,
      "epoch": 0.8170504418926231,
      "flos": 21251324799360.0,
      "grad_norm": 1.74745415733074,
      "language_loss": 0.63974035,
      "learning_rate": 3.4080282258827884e-07,
      "loss": 0.66222131,
      "num_input_tokens_seen": 146696230,
      "step": 6795,
      "time_per_iteration": 2.6281282901763916
    },
    {
      "auxiliary_loss_clip": 0.01227275,
      "auxiliary_loss_mlp": 0.0102519,
      "balance_loss_clip": 1.04847121,
      "balance_loss_mlp": 1.01852643,
      "epoch": 0.8171706847832622,
      "flos": 19099234362240.0,
      "grad_norm": 2.2706528478772263,
      "language_loss": 0.72810698,
      "learning_rate": 3.403680016360342e-07,
      "loss": 0.75063163,
      "num_input_tokens_seen": 146714835,
      "step": 6796,
      "time_per_iteration": 3.5340683460235596
    },
    {
      "auxiliary_loss_clip": 0.0122388,
      "auxiliary_loss_mlp": 0.01026657,
      "balance_loss_clip": 1.04910564,
      "balance_loss_mlp": 1.01962924,
      "epoch": 0.8172909276739013,
      "flos": 21470128496640.0,
      "grad_norm": 1.5396615459760938,
      "language_loss": 0.67535949,
      "learning_rate": 3.3993343244864403e-07,
      "loss": 0.69786489,
      "num_input_tokens_seen": 146734425,
      "step": 6797,
      "time_per_iteration": 2.6217939853668213
    },
    {
      "auxiliary_loss_clip": 0.01221229,
      "auxiliary_loss_mlp": 0.0102453,
      "balance_loss_clip": 1.04750466,
      "balance_loss_mlp": 1.01778913,
      "epoch": 0.8174111705645404,
      "flos": 27599792417280.0,
      "grad_norm": 1.6272879764520989,
      "language_loss": 0.72938049,
      "learning_rate": 3.394991150920323e-07,
      "loss": 0.75183809,
      "num_input_tokens_seen": 146757545,
      "step": 6798,
      "time_per_iteration": 2.655233860015869
    },
    {
      "auxiliary_loss_clip": 0.01375698,
      "auxiliary_loss_mlp": 0.02572054,
      "balance_loss_clip": 1.04243433,
      "balance_loss_mlp": 0.99993306,
      "epoch": 0.8175314134551794,
      "flos": 14064594508800.0,
      "grad_norm": 4.446636363375664,
      "language_loss": 0.74556893,
      "learning_rate": 3.3906504963208396e-07,
      "loss": 0.78504646,
      "num_input_tokens_seen": 146774240,
      "step": 6799,
      "time_per_iteration": 2.671595335006714
    },
    {
      "auxiliary_loss_clip": 0.01363273,
      "auxiliary_loss_mlp": 0.01026792,
      "balance_loss_clip": 1.04581821,
      "balance_loss_mlp": 1.01900733,
      "epoch": 0.8176516563458186,
      "flos": 22708076780160.0,
      "grad_norm": 1.870835195413321,
      "language_loss": 0.66908175,
      "learning_rate": 3.3863123613464774e-07,
      "loss": 0.69298238,
      "num_input_tokens_seen": 146793140,
      "step": 6800,
      "time_per_iteration": 2.741248846054077
    },
    {
      "auxiliary_loss_clip": 0.01274806,
      "auxiliary_loss_mlp": 0.01025317,
      "balance_loss_clip": 1.04195452,
      "balance_loss_mlp": 1.01830184,
      "epoch": 0.8177718992364577,
      "flos": 21945406279680.0,
      "grad_norm": 1.8666250342444926,
      "language_loss": 0.75072581,
      "learning_rate": 3.381976746655317e-07,
      "loss": 0.77372706,
      "num_input_tokens_seen": 146812895,
      "step": 6801,
      "time_per_iteration": 2.676837205886841
    },
    {
      "auxiliary_loss_clip": 0.01361235,
      "auxiliary_loss_mlp": 0.01023074,
      "balance_loss_clip": 1.04555225,
      "balance_loss_mlp": 1.01586509,
      "epoch": 0.8178921421270967,
      "flos": 22017443005440.0,
      "grad_norm": 2.035059172315375,
      "language_loss": 0.67749858,
      "learning_rate": 3.3776436529050756e-07,
      "loss": 0.70134169,
      "num_input_tokens_seen": 146832445,
      "step": 6802,
      "time_per_iteration": 2.781282663345337
    },
    {
      "auxiliary_loss_clip": 0.01174296,
      "auxiliary_loss_mlp": 0.01026177,
      "balance_loss_clip": 1.04983675,
      "balance_loss_mlp": 1.01943326,
      "epoch": 0.8180123850177359,
      "flos": 33183111496320.0,
      "grad_norm": 1.7482533883068616,
      "language_loss": 0.72428042,
      "learning_rate": 3.373313080753073e-07,
      "loss": 0.7462852,
      "num_input_tokens_seen": 146856505,
      "step": 6803,
      "time_per_iteration": 2.723553419113159
    },
    {
      "auxiliary_loss_clip": 0.0122077,
      "auxiliary_loss_mlp": 0.01024984,
      "balance_loss_clip": 1.04530334,
      "balance_loss_mlp": 1.01808178,
      "epoch": 0.8181326279083749,
      "flos": 22091167670400.0,
      "grad_norm": 1.6908780355046258,
      "language_loss": 0.77552247,
      "learning_rate": 3.3689850308562527e-07,
      "loss": 0.79798007,
      "num_input_tokens_seen": 146876950,
      "step": 6804,
      "time_per_iteration": 2.6377220153808594
    },
    {
      "auxiliary_loss_clip": 0.01372886,
      "auxiliary_loss_mlp": 0.01029015,
      "balance_loss_clip": 1.04908872,
      "balance_loss_mlp": 1.02192545,
      "epoch": 0.818252870799014,
      "flos": 15705747936000.0,
      "grad_norm": 3.2593315133909226,
      "language_loss": 0.77699137,
      "learning_rate": 3.364659503871183e-07,
      "loss": 0.80101037,
      "num_input_tokens_seen": 146894885,
      "step": 6805,
      "time_per_iteration": 2.7260968685150146
    },
    {
      "auxiliary_loss_clip": 0.01319414,
      "auxiliary_loss_mlp": 0.01023965,
      "balance_loss_clip": 1.04208398,
      "balance_loss_mlp": 1.01751888,
      "epoch": 0.8183731136896532,
      "flos": 18770687637120.0,
      "grad_norm": 2.2754121420906173,
      "language_loss": 0.84084195,
      "learning_rate": 3.3603365004540417e-07,
      "loss": 0.86427575,
      "num_input_tokens_seen": 146913180,
      "step": 6806,
      "time_per_iteration": 2.6681227684020996
    },
    {
      "auxiliary_loss_clip": 0.01174257,
      "auxiliary_loss_mlp": 0.01024883,
      "balance_loss_clip": 1.05035496,
      "balance_loss_mlp": 1.01839471,
      "epoch": 0.8184933565802922,
      "flos": 26541792293760.0,
      "grad_norm": 2.0197955488118593,
      "language_loss": 0.77595568,
      "learning_rate": 3.356016021260624e-07,
      "loss": 0.79794711,
      "num_input_tokens_seen": 146933510,
      "step": 6807,
      "time_per_iteration": 2.6272759437561035
    },
    {
      "auxiliary_loss_clip": 0.01223714,
      "auxiliary_loss_mlp": 0.01029072,
      "balance_loss_clip": 1.05079091,
      "balance_loss_mlp": 1.02203894,
      "epoch": 0.8186135994709313,
      "flos": 17530117660800.0,
      "grad_norm": 2.4758110003360554,
      "language_loss": 0.65707105,
      "learning_rate": 3.35169806694634e-07,
      "loss": 0.67959887,
      "num_input_tokens_seen": 146951760,
      "step": 6808,
      "time_per_iteration": 2.7163734436035156
    },
    {
      "auxiliary_loss_clip": 0.01220108,
      "auxiliary_loss_mlp": 0.01003133,
      "balance_loss_clip": 1.00771642,
      "balance_loss_mlp": 1.00220287,
      "epoch": 0.8187338423615703,
      "flos": 63480300675840.0,
      "grad_norm": 0.7153392003753575,
      "language_loss": 0.60622025,
      "learning_rate": 3.3473826381662186e-07,
      "loss": 0.62845266,
      "num_input_tokens_seen": 147022900,
      "step": 6809,
      "time_per_iteration": 3.3499295711517334
    },
    {
      "auxiliary_loss_clip": 0.01218652,
      "auxiliary_loss_mlp": 0.01020013,
      "balance_loss_clip": 1.04824758,
      "balance_loss_mlp": 1.01328683,
      "epoch": 0.8188540852522095,
      "flos": 17529974006400.0,
      "grad_norm": 2.665375156264429,
      "language_loss": 0.81808937,
      "learning_rate": 3.3430697355749216e-07,
      "loss": 0.84047604,
      "num_input_tokens_seen": 147040590,
      "step": 6810,
      "time_per_iteration": 2.615480661392212
    },
    {
      "auxiliary_loss_clip": 0.01366513,
      "auxiliary_loss_mlp": 0.01024313,
      "balance_loss_clip": 1.04131091,
      "balance_loss_mlp": 1.01725626,
      "epoch": 0.8189743281428485,
      "flos": 14392530702720.0,
      "grad_norm": 2.3433862222857904,
      "language_loss": 0.75363958,
      "learning_rate": 3.3387593598266907e-07,
      "loss": 0.77754784,
      "num_input_tokens_seen": 147057200,
      "step": 6811,
      "time_per_iteration": 2.6746647357940674
    },
    {
      "auxiliary_loss_clip": 0.01320129,
      "auxiliary_loss_mlp": 0.01027959,
      "balance_loss_clip": 1.04301047,
      "balance_loss_mlp": 1.02085066,
      "epoch": 0.8190945710334876,
      "flos": 25080479285760.0,
      "grad_norm": 2.4699324224975157,
      "language_loss": 0.78427529,
      "learning_rate": 3.3344515115754225e-07,
      "loss": 0.80775619,
      "num_input_tokens_seen": 147076180,
      "step": 6812,
      "time_per_iteration": 2.7727279663085938
    },
    {
      "auxiliary_loss_clip": 0.01223051,
      "auxiliary_loss_mlp": 0.01027101,
      "balance_loss_clip": 1.04339862,
      "balance_loss_mlp": 1.02033234,
      "epoch": 0.8192148139241268,
      "flos": 21507152440320.0,
      "grad_norm": 2.293578688059005,
      "language_loss": 0.79727095,
      "learning_rate": 3.33014619147461e-07,
      "loss": 0.81977242,
      "num_input_tokens_seen": 147094205,
      "step": 6813,
      "time_per_iteration": 2.791146755218506
    },
    {
      "auxiliary_loss_clip": 0.01273024,
      "auxiliary_loss_mlp": 0.01027225,
      "balance_loss_clip": 1.04976964,
      "balance_loss_mlp": 1.01986122,
      "epoch": 0.8193350568147658,
      "flos": 23952166289280.0,
      "grad_norm": 9.443954181959024,
      "language_loss": 0.71739572,
      "learning_rate": 3.325843400177362e-07,
      "loss": 0.74039817,
      "num_input_tokens_seen": 147115545,
      "step": 6814,
      "time_per_iteration": 3.6216471195220947
    },
    {
      "auxiliary_loss_clip": 0.0122708,
      "auxiliary_loss_mlp": 0.0256735,
      "balance_loss_clip": 1.04750752,
      "balance_loss_mlp": 0.99994195,
      "epoch": 0.8194552997054049,
      "flos": 20559469962240.0,
      "grad_norm": 2.6601201763279203,
      "language_loss": 0.73972458,
      "learning_rate": 3.32154313833642e-07,
      "loss": 0.77766889,
      "num_input_tokens_seen": 147135700,
      "step": 6815,
      "time_per_iteration": 3.622901678085327
    },
    {
      "auxiliary_loss_clip": 0.01176362,
      "auxiliary_loss_mlp": 0.01026413,
      "balance_loss_clip": 1.049752,
      "balance_loss_mlp": 1.01912355,
      "epoch": 0.819575542596044,
      "flos": 26031753123840.0,
      "grad_norm": 2.2421491513023977,
      "language_loss": 0.59749335,
      "learning_rate": 3.3172454066041164e-07,
      "loss": 0.61952108,
      "num_input_tokens_seen": 147155205,
      "step": 6816,
      "time_per_iteration": 2.5803370475769043
    },
    {
      "auxiliary_loss_clip": 0.01425771,
      "auxiliary_loss_mlp": 0.02561587,
      "balance_loss_clip": 1.0483706,
      "balance_loss_mlp": 0.99994195,
      "epoch": 0.8196957854866831,
      "flos": 29096944220160.0,
      "grad_norm": 1.801701723011235,
      "language_loss": 0.76230073,
      "learning_rate": 3.3129502056324234e-07,
      "loss": 0.80217433,
      "num_input_tokens_seen": 147176570,
      "step": 6817,
      "time_per_iteration": 3.7387492656707764
    },
    {
      "auxiliary_loss_clip": 0.01385019,
      "auxiliary_loss_mlp": 0.01000379,
      "balance_loss_clip": 1.01020372,
      "balance_loss_mlp": 0.99937803,
      "epoch": 0.8198160283773221,
      "flos": 69033631898880.0,
      "grad_norm": 0.8152012679793276,
      "language_loss": 0.59710503,
      "learning_rate": 3.3086575360729165e-07,
      "loss": 0.62095904,
      "num_input_tokens_seen": 147234105,
      "step": 6818,
      "time_per_iteration": 3.359842300415039
    },
    {
      "auxiliary_loss_clip": 0.01272739,
      "auxiliary_loss_mlp": 0.01028031,
      "balance_loss_clip": 1.04679203,
      "balance_loss_mlp": 1.02122986,
      "epoch": 0.8199362712679613,
      "flos": 16618058496000.0,
      "grad_norm": 1.963748550422414,
      "language_loss": 0.71592933,
      "learning_rate": 3.3043673985767906e-07,
      "loss": 0.73893702,
      "num_input_tokens_seen": 147253170,
      "step": 6819,
      "time_per_iteration": 3.238403797149658
    },
    {
      "auxiliary_loss_clip": 0.01314234,
      "auxiliary_loss_mlp": 0.01024311,
      "balance_loss_clip": 1.03999782,
      "balance_loss_mlp": 1.01759923,
      "epoch": 0.8200565141586004,
      "flos": 21757664868480.0,
      "grad_norm": 1.9878521255193269,
      "language_loss": 0.77782738,
      "learning_rate": 3.3000797937948564e-07,
      "loss": 0.80121279,
      "num_input_tokens_seen": 147271465,
      "step": 6820,
      "time_per_iteration": 2.8147151470184326
    },
    {
      "auxiliary_loss_clip": 0.01120008,
      "auxiliary_loss_mlp": 0.01001261,
      "balance_loss_clip": 1.00787652,
      "balance_loss_mlp": 1.00036705,
      "epoch": 0.8201767570492394,
      "flos": 69807112392960.0,
      "grad_norm": 0.9419398410811813,
      "language_loss": 0.65048718,
      "learning_rate": 3.295794722377534e-07,
      "loss": 0.67169988,
      "num_input_tokens_seen": 147335070,
      "step": 6821,
      "time_per_iteration": 3.290086507797241
    },
    {
      "auxiliary_loss_clip": 0.01171578,
      "auxiliary_loss_mlp": 0.01026096,
      "balance_loss_clip": 1.04950261,
      "balance_loss_mlp": 1.01959324,
      "epoch": 0.8202969999398786,
      "flos": 23111892455040.0,
      "grad_norm": 2.265964099936706,
      "language_loss": 0.80027509,
      "learning_rate": 3.291512184974876e-07,
      "loss": 0.8222518,
      "num_input_tokens_seen": 147355460,
      "step": 6822,
      "time_per_iteration": 3.4942917823791504
    },
    {
      "auxiliary_loss_clip": 0.01274027,
      "auxiliary_loss_mlp": 0.01027818,
      "balance_loss_clip": 1.04331994,
      "balance_loss_mlp": 1.02041209,
      "epoch": 0.8204172428305176,
      "flos": 28220616109440.0,
      "grad_norm": 2.097961677723926,
      "language_loss": 0.66896379,
      "learning_rate": 3.2872321822365346e-07,
      "loss": 0.69198227,
      "num_input_tokens_seen": 147375675,
      "step": 6823,
      "time_per_iteration": 2.769864559173584
    },
    {
      "auxiliary_loss_clip": 0.01223771,
      "auxiliary_loss_mlp": 0.01020583,
      "balance_loss_clip": 1.05002213,
      "balance_loss_mlp": 1.01387787,
      "epoch": 0.8205374857211567,
      "flos": 20887011106560.0,
      "grad_norm": 1.927936502782618,
      "language_loss": 0.73409432,
      "learning_rate": 3.282954714811783e-07,
      "loss": 0.75653791,
      "num_input_tokens_seen": 147394580,
      "step": 6824,
      "time_per_iteration": 2.620063304901123
    },
    {
      "auxiliary_loss_clip": 0.01266431,
      "auxiliary_loss_mlp": 0.01025047,
      "balance_loss_clip": 1.04190135,
      "balance_loss_mlp": 1.01819801,
      "epoch": 0.8206577286117959,
      "flos": 13152140294400.0,
      "grad_norm": 2.469487354915222,
      "language_loss": 0.71117258,
      "learning_rate": 3.2786797833495093e-07,
      "loss": 0.73408735,
      "num_input_tokens_seen": 147409935,
      "step": 6825,
      "time_per_iteration": 2.6245388984680176
    },
    {
      "auxiliary_loss_clip": 0.01172411,
      "auxiliary_loss_mlp": 0.01028825,
      "balance_loss_clip": 1.04804993,
      "balance_loss_mlp": 1.02211046,
      "epoch": 0.8207779715024349,
      "flos": 25265634917760.0,
      "grad_norm": 2.6451725821433567,
      "language_loss": 0.72601295,
      "learning_rate": 3.274407388498213e-07,
      "loss": 0.7480253,
      "num_input_tokens_seen": 147428065,
      "step": 6826,
      "time_per_iteration": 2.60349702835083
    },
    {
      "auxiliary_loss_clip": 0.01320064,
      "auxiliary_loss_mlp": 0.01020841,
      "balance_loss_clip": 1.04345679,
      "balance_loss_mlp": 1.01403975,
      "epoch": 0.820898214393074,
      "flos": 19610243199360.0,
      "grad_norm": 1.9305486509705465,
      "language_loss": 0.74278504,
      "learning_rate": 3.270137530906021e-07,
      "loss": 0.76619411,
      "num_input_tokens_seen": 147447300,
      "step": 6827,
      "time_per_iteration": 2.693319797515869
    },
    {
      "auxiliary_loss_clip": 0.01367033,
      "auxiliary_loss_mlp": 0.01022458,
      "balance_loss_clip": 1.04760289,
      "balance_loss_mlp": 1.01598513,
      "epoch": 0.8210184572837131,
      "flos": 15596615439360.0,
      "grad_norm": 1.73747645972751,
      "language_loss": 0.83749592,
      "learning_rate": 3.265870211220665e-07,
      "loss": 0.86139083,
      "num_input_tokens_seen": 147465135,
      "step": 6828,
      "time_per_iteration": 2.7035157680511475
    },
    {
      "auxiliary_loss_clip": 0.01324889,
      "auxiliary_loss_mlp": 0.01027296,
      "balance_loss_clip": 1.04627705,
      "balance_loss_mlp": 1.01985121,
      "epoch": 0.8211387001743522,
      "flos": 20813932886400.0,
      "grad_norm": 2.011892085174537,
      "language_loss": 0.82086265,
      "learning_rate": 3.2616054300894934e-07,
      "loss": 0.84438443,
      "num_input_tokens_seen": 147484585,
      "step": 6829,
      "time_per_iteration": 2.848741292953491
    },
    {
      "auxiliary_loss_clip": 0.01272316,
      "auxiliary_loss_mlp": 0.01026747,
      "balance_loss_clip": 1.04804623,
      "balance_loss_mlp": 1.02016711,
      "epoch": 0.8212589430649913,
      "flos": 27704579368320.0,
      "grad_norm": 2.1383824555635105,
      "language_loss": 0.84155977,
      "learning_rate": 3.2573431881594693e-07,
      "loss": 0.86455041,
      "num_input_tokens_seen": 147504130,
      "step": 6830,
      "time_per_iteration": 2.6674158573150635
    },
    {
      "auxiliary_loss_clip": 0.0141777,
      "auxiliary_loss_mlp": 0.01024098,
      "balance_loss_clip": 1.03908801,
      "balance_loss_mlp": 1.01694512,
      "epoch": 0.8213791859556304,
      "flos": 22455625017600.0,
      "grad_norm": 2.1631205754067,
      "language_loss": 0.66306305,
      "learning_rate": 3.2530834860771663e-07,
      "loss": 0.6874817,
      "num_input_tokens_seen": 147523510,
      "step": 6831,
      "time_per_iteration": 2.786813974380493
    },
    {
      "auxiliary_loss_clip": 0.01225471,
      "auxiliary_loss_mlp": 0.0102804,
      "balance_loss_clip": 1.04687643,
      "balance_loss_mlp": 1.02097416,
      "epoch": 0.8214994288462695,
      "flos": 16654471908480.0,
      "grad_norm": 5.221646466420648,
      "language_loss": 0.74223304,
      "learning_rate": 3.248826324488794e-07,
      "loss": 0.76476812,
      "num_input_tokens_seen": 147540805,
      "step": 6832,
      "time_per_iteration": 2.6033363342285156
    },
    {
      "auxiliary_loss_clip": 0.01174922,
      "auxiliary_loss_mlp": 0.01024772,
      "balance_loss_clip": 1.05252028,
      "balance_loss_mlp": 1.01777744,
      "epoch": 0.8216196717369085,
      "flos": 25221787390080.0,
      "grad_norm": 1.9299328126259223,
      "language_loss": 0.88158536,
      "learning_rate": 3.244571704040138e-07,
      "loss": 0.90358233,
      "num_input_tokens_seen": 147560965,
      "step": 6833,
      "time_per_iteration": 2.5997700691223145
    },
    {
      "auxiliary_loss_clip": 0.01220468,
      "auxiliary_loss_mlp": 0.01027397,
      "balance_loss_clip": 1.04419231,
      "balance_loss_mlp": 1.01999998,
      "epoch": 0.8217399146275477,
      "flos": 25371930240000.0,
      "grad_norm": 1.9563481517546748,
      "language_loss": 0.73705041,
      "learning_rate": 3.2403196253766374e-07,
      "loss": 0.75952911,
      "num_input_tokens_seen": 147580045,
      "step": 6834,
      "time_per_iteration": 2.6733803749084473
    },
    {
      "auxiliary_loss_clip": 0.01226038,
      "auxiliary_loss_mlp": 0.01027093,
      "balance_loss_clip": 1.04856908,
      "balance_loss_mlp": 1.01926112,
      "epoch": 0.8218601575181868,
      "flos": 25629625388160.0,
      "grad_norm": 2.263090122009148,
      "language_loss": 0.790797,
      "learning_rate": 3.2360700891433254e-07,
      "loss": 0.81332833,
      "num_input_tokens_seen": 147599070,
      "step": 6835,
      "time_per_iteration": 2.6097846031188965
    },
    {
      "auxiliary_loss_clip": 0.01273342,
      "auxiliary_loss_mlp": 0.00998796,
      "balance_loss_clip": 1.0101676,
      "balance_loss_mlp": 0.99788439,
      "epoch": 0.8219804004088258,
      "flos": 67660229427840.0,
      "grad_norm": 0.7949293193217626,
      "language_loss": 0.57270622,
      "learning_rate": 3.231823095984847e-07,
      "loss": 0.59542751,
      "num_input_tokens_seen": 147653710,
      "step": 6836,
      "time_per_iteration": 3.218120813369751
    },
    {
      "auxiliary_loss_clip": 0.01271136,
      "auxiliary_loss_mlp": 0.01019548,
      "balance_loss_clip": 1.04736269,
      "balance_loss_mlp": 1.01284218,
      "epoch": 0.822100643299465,
      "flos": 19464266327040.0,
      "grad_norm": 2.284721406289287,
      "language_loss": 0.75987053,
      "learning_rate": 3.2275786465454814e-07,
      "loss": 0.78277731,
      "num_input_tokens_seen": 147670360,
      "step": 6837,
      "time_per_iteration": 2.605473041534424
    },
    {
      "auxiliary_loss_clip": 0.01322909,
      "auxiliary_loss_mlp": 0.01022756,
      "balance_loss_clip": 1.04372644,
      "balance_loss_mlp": 1.0158596,
      "epoch": 0.822220886190104,
      "flos": 24681368292480.0,
      "grad_norm": 5.567301041693867,
      "language_loss": 0.75799859,
      "learning_rate": 3.2233367414690917e-07,
      "loss": 0.78145528,
      "num_input_tokens_seen": 147692550,
      "step": 6838,
      "time_per_iteration": 2.766899824142456
    },
    {
      "auxiliary_loss_clip": 0.01318223,
      "auxiliary_loss_mlp": 0.01028142,
      "balance_loss_clip": 1.04197538,
      "balance_loss_mlp": 1.02180648,
      "epoch": 0.8223411290807431,
      "flos": 27819062991360.0,
      "grad_norm": 2.198407304856372,
      "language_loss": 0.85300118,
      "learning_rate": 3.219097381399183e-07,
      "loss": 0.87646484,
      "num_input_tokens_seen": 147709725,
      "step": 6839,
      "time_per_iteration": 2.7385849952697754
    },
    {
      "auxiliary_loss_clip": 0.01283092,
      "auxiliary_loss_mlp": 0.01027356,
      "balance_loss_clip": 1.05001962,
      "balance_loss_mlp": 1.01977992,
      "epoch": 0.8224613719713821,
      "flos": 23218546913280.0,
      "grad_norm": 1.8160193015676551,
      "language_loss": 0.81195027,
      "learning_rate": 3.2148605669788584e-07,
      "loss": 0.83505476,
      "num_input_tokens_seen": 147729615,
      "step": 6840,
      "time_per_iteration": 4.65327000617981
    },
    {
      "auxiliary_loss_clip": 0.01274214,
      "auxiliary_loss_mlp": 0.01020944,
      "balance_loss_clip": 1.04883552,
      "balance_loss_mlp": 1.01434922,
      "epoch": 0.8225816148620213,
      "flos": 15706250726400.0,
      "grad_norm": 2.7425572295645755,
      "language_loss": 0.77566087,
      "learning_rate": 3.2106262988508405e-07,
      "loss": 0.79861248,
      "num_input_tokens_seen": 147747665,
      "step": 6841,
      "time_per_iteration": 2.6476659774780273
    },
    {
      "auxiliary_loss_clip": 0.01270844,
      "auxiliary_loss_mlp": 0.01025068,
      "balance_loss_clip": 1.04674363,
      "balance_loss_mlp": 1.0180999,
      "epoch": 0.8227018577526604,
      "flos": 18515111391360.0,
      "grad_norm": 1.7551219264902935,
      "language_loss": 0.74287027,
      "learning_rate": 3.206394577657465e-07,
      "loss": 0.76582944,
      "num_input_tokens_seen": 147765445,
      "step": 6842,
      "time_per_iteration": 2.647963285446167
    },
    {
      "auxiliary_loss_clip": 0.01228053,
      "auxiliary_loss_mlp": 0.01025749,
      "balance_loss_clip": 1.04963219,
      "balance_loss_mlp": 1.0179441,
      "epoch": 0.8228221006432994,
      "flos": 22236785406720.0,
      "grad_norm": 2.3375453393173586,
      "language_loss": 0.72553831,
      "learning_rate": 3.202165404040675e-07,
      "loss": 0.74807632,
      "num_input_tokens_seen": 147783365,
      "step": 6843,
      "time_per_iteration": 3.4711837768554688
    },
    {
      "auxiliary_loss_clip": 0.01425889,
      "auxiliary_loss_mlp": 0.01026672,
      "balance_loss_clip": 1.044873,
      "balance_loss_mlp": 1.01925421,
      "epoch": 0.8229423435339386,
      "flos": 24097532630400.0,
      "grad_norm": 10.714833472333588,
      "language_loss": 0.74591839,
      "learning_rate": 3.1979387786420396e-07,
      "loss": 0.77044404,
      "num_input_tokens_seen": 147803605,
      "step": 6844,
      "time_per_iteration": 2.813122510910034
    },
    {
      "auxiliary_loss_clip": 0.01272895,
      "auxiliary_loss_mlp": 0.0102664,
      "balance_loss_clip": 1.04277492,
      "balance_loss_mlp": 1.01959229,
      "epoch": 0.8230625864245776,
      "flos": 23878549365120.0,
      "grad_norm": 1.9671351431648865,
      "language_loss": 0.82231522,
      "learning_rate": 3.1937147021027346e-07,
      "loss": 0.84531057,
      "num_input_tokens_seen": 147822060,
      "step": 6845,
      "time_per_iteration": 2.700087785720825
    },
    {
      "auxiliary_loss_clip": 0.01219281,
      "auxiliary_loss_mlp": 0.01031073,
      "balance_loss_clip": 1.04779088,
      "balance_loss_mlp": 1.02432609,
      "epoch": 0.8231828293152167,
      "flos": 16581106379520.0,
      "grad_norm": 2.313914094379665,
      "language_loss": 0.76776212,
      "learning_rate": 3.189493175063547e-07,
      "loss": 0.79026568,
      "num_input_tokens_seen": 147839295,
      "step": 6846,
      "time_per_iteration": 2.6102447509765625
    },
    {
      "auxiliary_loss_clip": 0.01275697,
      "auxiliary_loss_mlp": 0.0102597,
      "balance_loss_clip": 1.04941297,
      "balance_loss_mlp": 1.01877022,
      "epoch": 0.8233030722058559,
      "flos": 18880071528960.0,
      "grad_norm": 2.59520600950338,
      "language_loss": 0.67486167,
      "learning_rate": 3.1852741981648776e-07,
      "loss": 0.69787836,
      "num_input_tokens_seen": 147857945,
      "step": 6847,
      "time_per_iteration": 2.68234920501709
    },
    {
      "auxiliary_loss_clip": 0.01320787,
      "auxiliary_loss_mlp": 0.01034071,
      "balance_loss_clip": 1.0471549,
      "balance_loss_mlp": 1.02747858,
      "epoch": 0.8234233150964949,
      "flos": 28439024757120.0,
      "grad_norm": 2.407267921916861,
      "language_loss": 0.69722742,
      "learning_rate": 3.1810577720467404e-07,
      "loss": 0.72077596,
      "num_input_tokens_seen": 147879675,
      "step": 6848,
      "time_per_iteration": 3.6897475719451904
    },
    {
      "auxiliary_loss_clip": 0.01276104,
      "auxiliary_loss_mlp": 0.01027068,
      "balance_loss_clip": 1.04800296,
      "balance_loss_mlp": 1.01996338,
      "epoch": 0.823543557987134,
      "flos": 33765941577600.0,
      "grad_norm": 1.5949055119751934,
      "language_loss": 0.56731194,
      "learning_rate": 3.176843897348769e-07,
      "loss": 0.59034365,
      "num_input_tokens_seen": 147902870,
      "step": 6849,
      "time_per_iteration": 2.7959048748016357
    },
    {
      "auxiliary_loss_clip": 0.01271123,
      "auxiliary_loss_mlp": 0.01026204,
      "balance_loss_clip": 1.04624665,
      "balance_loss_mlp": 1.01837754,
      "epoch": 0.8236638008777731,
      "flos": 17092366611840.0,
      "grad_norm": 2.9578485315626253,
      "language_loss": 0.75892758,
      "learning_rate": 3.1726325747102034e-07,
      "loss": 0.78190082,
      "num_input_tokens_seen": 147921245,
      "step": 6850,
      "time_per_iteration": 2.637136220932007
    },
    {
      "auxiliary_loss_clip": 0.01365849,
      "auxiliary_loss_mlp": 0.01027705,
      "balance_loss_clip": 1.03689027,
      "balance_loss_mlp": 1.02076697,
      "epoch": 0.8237840437684122,
      "flos": 61639982334720.0,
      "grad_norm": 1.6857020507723248,
      "language_loss": 0.64200616,
      "learning_rate": 3.1684238047698974e-07,
      "loss": 0.66594177,
      "num_input_tokens_seen": 147949515,
      "step": 6851,
      "time_per_iteration": 3.0977232456207275
    },
    {
      "auxiliary_loss_clip": 0.01275643,
      "auxiliary_loss_mlp": 0.0102512,
      "balance_loss_clip": 1.0477922,
      "balance_loss_mlp": 1.01808095,
      "epoch": 0.8239042866590512,
      "flos": 27309023821440.0,
      "grad_norm": 2.1975193477272636,
      "language_loss": 0.52991784,
      "learning_rate": 3.1642175881663155e-07,
      "loss": 0.55292547,
      "num_input_tokens_seen": 147969245,
      "step": 6852,
      "time_per_iteration": 2.746411085128784
    },
    {
      "auxiliary_loss_clip": 0.01171394,
      "auxiliary_loss_mlp": 0.01025384,
      "balance_loss_clip": 1.04751396,
      "balance_loss_mlp": 1.01847291,
      "epoch": 0.8240245295496904,
      "flos": 21726351187200.0,
      "grad_norm": 2.247745445393768,
      "language_loss": 0.8394829,
      "learning_rate": 3.160013925537537e-07,
      "loss": 0.86145067,
      "num_input_tokens_seen": 147990080,
      "step": 6853,
      "time_per_iteration": 2.591095447540283
    },
    {
      "auxiliary_loss_clip": 0.01225123,
      "auxiliary_loss_mlp": 0.01025763,
      "balance_loss_clip": 1.04652369,
      "balance_loss_mlp": 1.01860404,
      "epoch": 0.8241447724403295,
      "flos": 20009318279040.0,
      "grad_norm": 2.1398394458436596,
      "language_loss": 0.75658637,
      "learning_rate": 3.155812817521266e-07,
      "loss": 0.77909523,
      "num_input_tokens_seen": 148010455,
      "step": 6854,
      "time_per_iteration": 2.7101058959960938
    },
    {
      "auxiliary_loss_clip": 0.01278693,
      "auxiliary_loss_mlp": 0.01025822,
      "balance_loss_clip": 1.05091679,
      "balance_loss_mlp": 1.01813269,
      "epoch": 0.8242650153309685,
      "flos": 22272983337600.0,
      "grad_norm": 2.3610709135328127,
      "language_loss": 0.77975512,
      "learning_rate": 3.151614264754787e-07,
      "loss": 0.8028003,
      "num_input_tokens_seen": 148028400,
      "step": 6855,
      "time_per_iteration": 2.672050714492798
    },
    {
      "auxiliary_loss_clip": 0.0117422,
      "auxiliary_loss_mlp": 0.0102073,
      "balance_loss_clip": 1.0477196,
      "balance_loss_mlp": 1.01386356,
      "epoch": 0.8243852582216077,
      "flos": 22309971367680.0,
      "grad_norm": 2.5368371574171875,
      "language_loss": 0.79629368,
      "learning_rate": 3.147418267875035e-07,
      "loss": 0.81824321,
      "num_input_tokens_seen": 148046530,
      "step": 6856,
      "time_per_iteration": 2.5626449584960938
    },
    {
      "auxiliary_loss_clip": 0.01421087,
      "auxiliary_loss_mlp": 0.02565082,
      "balance_loss_clip": 1.03961778,
      "balance_loss_mlp": 0.99991339,
      "epoch": 0.8245055011122467,
      "flos": 24645421756800.0,
      "grad_norm": 3.36237442034114,
      "language_loss": 0.65553236,
      "learning_rate": 3.1432248275185315e-07,
      "loss": 0.69539404,
      "num_input_tokens_seen": 148067040,
      "step": 6857,
      "time_per_iteration": 2.8278796672821045
    },
    {
      "auxiliary_loss_clip": 0.0122101,
      "auxiliary_loss_mlp": 0.01024196,
      "balance_loss_clip": 1.04795933,
      "balance_loss_mlp": 1.01692462,
      "epoch": 0.8246257440028858,
      "flos": 17487275713920.0,
      "grad_norm": 3.440686790732016,
      "language_loss": 0.76549423,
      "learning_rate": 3.139033944321412e-07,
      "loss": 0.78794628,
      "num_input_tokens_seen": 148084400,
      "step": 6858,
      "time_per_iteration": 2.5445709228515625
    },
    {
      "auxiliary_loss_clip": 0.01225878,
      "auxiliary_loss_mlp": 0.01025086,
      "balance_loss_clip": 1.04779267,
      "balance_loss_mlp": 1.01816273,
      "epoch": 0.824745986893525,
      "flos": 25010130499200.0,
      "grad_norm": 1.712750835300007,
      "language_loss": 0.7897166,
      "learning_rate": 3.1348456189194507e-07,
      "loss": 0.81222618,
      "num_input_tokens_seen": 148104860,
      "step": 6859,
      "time_per_iteration": 2.6594972610473633
    },
    {
      "auxiliary_loss_clip": 0.01318249,
      "auxiliary_loss_mlp": 0.01022861,
      "balance_loss_clip": 1.04137492,
      "balance_loss_mlp": 1.0156461,
      "epoch": 0.824866229784164,
      "flos": 18772698798720.0,
      "grad_norm": 1.8749240135180056,
      "language_loss": 0.83091676,
      "learning_rate": 3.1306598519479876e-07,
      "loss": 0.8543278,
      "num_input_tokens_seen": 148124680,
      "step": 6860,
      "time_per_iteration": 2.689815044403076
    },
    {
      "auxiliary_loss_clip": 0.0126999,
      "auxiliary_loss_mlp": 0.01030412,
      "balance_loss_clip": 1.04668331,
      "balance_loss_mlp": 1.02334833,
      "epoch": 0.8249864726748031,
      "flos": 23842171866240.0,
      "grad_norm": 2.5909489018878675,
      "language_loss": 0.78245074,
      "learning_rate": 3.1264766440420177e-07,
      "loss": 0.80545479,
      "num_input_tokens_seen": 148147150,
      "step": 6861,
      "time_per_iteration": 2.710061550140381
    },
    {
      "auxiliary_loss_clip": 0.01219358,
      "auxiliary_loss_mlp": 0.01025681,
      "balance_loss_clip": 1.04784036,
      "balance_loss_mlp": 1.01899004,
      "epoch": 0.8251067155654422,
      "flos": 20303103617280.0,
      "grad_norm": 1.9410662852094713,
      "language_loss": 0.6940161,
      "learning_rate": 3.122295995836124e-07,
      "loss": 0.71646649,
      "num_input_tokens_seen": 148167020,
      "step": 6862,
      "time_per_iteration": 2.676382064819336
    },
    {
      "auxiliary_loss_clip": 0.01226099,
      "auxiliary_loss_mlp": 0.01023195,
      "balance_loss_clip": 1.04635835,
      "balance_loss_mlp": 1.01626873,
      "epoch": 0.8252269584560813,
      "flos": 25009699536000.0,
      "grad_norm": 1.8677406558935064,
      "language_loss": 0.77654016,
      "learning_rate": 3.118117907964508e-07,
      "loss": 0.79903316,
      "num_input_tokens_seen": 148188965,
      "step": 6863,
      "time_per_iteration": 2.6928765773773193
    },
    {
      "auxiliary_loss_clip": 0.01329912,
      "auxiliary_loss_mlp": 0.0102467,
      "balance_loss_clip": 1.0451231,
      "balance_loss_mlp": 1.01829553,
      "epoch": 0.8253472013467203,
      "flos": 17128564542720.0,
      "grad_norm": 1.9527667043482129,
      "language_loss": 0.80109549,
      "learning_rate": 3.1139423810609856e-07,
      "loss": 0.82464129,
      "num_input_tokens_seen": 148205660,
      "step": 6864,
      "time_per_iteration": 2.728998899459839
    },
    {
      "auxiliary_loss_clip": 0.01171812,
      "auxiliary_loss_mlp": 0.01023633,
      "balance_loss_clip": 1.04689765,
      "balance_loss_mlp": 1.01691234,
      "epoch": 0.8254674442373595,
      "flos": 22414794232320.0,
      "grad_norm": 2.1099261657712596,
      "language_loss": 0.75353622,
      "learning_rate": 3.1097694157589714e-07,
      "loss": 0.77549064,
      "num_input_tokens_seen": 148225545,
      "step": 6865,
      "time_per_iteration": 3.6986193656921387
    },
    {
      "auxiliary_loss_clip": 0.01219436,
      "auxiliary_loss_mlp": 0.01021832,
      "balance_loss_clip": 1.04851162,
      "balance_loss_mlp": 1.01484334,
      "epoch": 0.8255876871279986,
      "flos": 24786765774720.0,
      "grad_norm": 2.4843214968512757,
      "language_loss": 0.76084709,
      "learning_rate": 3.105599012691511e-07,
      "loss": 0.78325975,
      "num_input_tokens_seen": 148243975,
      "step": 6866,
      "time_per_iteration": 3.482978105545044
    },
    {
      "auxiliary_loss_clip": 0.01219724,
      "auxiliary_loss_mlp": 0.01026893,
      "balance_loss_clip": 1.04812264,
      "balance_loss_mlp": 1.0197134,
      "epoch": 0.8257079300186376,
      "flos": 27455431656960.0,
      "grad_norm": 1.4375205107180464,
      "language_loss": 0.82595432,
      "learning_rate": 3.101431172491249e-07,
      "loss": 0.84842044,
      "num_input_tokens_seen": 148265520,
      "step": 6867,
      "time_per_iteration": 2.66448974609375
    },
    {
      "auxiliary_loss_clip": 0.0132513,
      "auxiliary_loss_mlp": 0.02568002,
      "balance_loss_clip": 1.04253078,
      "balance_loss_mlp": 0.99991709,
      "epoch": 0.8258281729092768,
      "flos": 16471866142080.0,
      "grad_norm": 2.1983982420043207,
      "language_loss": 0.72141808,
      "learning_rate": 3.097265895790444e-07,
      "loss": 0.76034939,
      "num_input_tokens_seen": 148283730,
      "step": 6868,
      "time_per_iteration": 2.6679491996765137
    },
    {
      "auxiliary_loss_clip": 0.01320125,
      "auxiliary_loss_mlp": 0.01029626,
      "balance_loss_clip": 1.04262793,
      "balance_loss_mlp": 1.02218127,
      "epoch": 0.8259484157999158,
      "flos": 21433822824960.0,
      "grad_norm": 1.8982764641070924,
      "language_loss": 0.83476067,
      "learning_rate": 3.093103183220962e-07,
      "loss": 0.85825813,
      "num_input_tokens_seen": 148303775,
      "step": 6869,
      "time_per_iteration": 3.517441987991333
    },
    {
      "auxiliary_loss_clip": 0.01115105,
      "auxiliary_loss_mlp": 0.01003852,
      "balance_loss_clip": 1.00747228,
      "balance_loss_mlp": 1.00298786,
      "epoch": 0.8260686586905549,
      "flos": 58322342453760.0,
      "grad_norm": 0.8242082274866178,
      "language_loss": 0.59290105,
      "learning_rate": 3.0889430354142796e-07,
      "loss": 0.61409062,
      "num_input_tokens_seen": 148365285,
      "step": 6870,
      "time_per_iteration": 3.16646409034729
    },
    {
      "auxiliary_loss_clip": 0.01321984,
      "auxiliary_loss_mlp": 0.01025521,
      "balance_loss_clip": 1.04286766,
      "balance_loss_mlp": 1.01850581,
      "epoch": 0.826188901581194,
      "flos": 27527288814720.0,
      "grad_norm": 3.432511771096845,
      "language_loss": 0.69942486,
      "learning_rate": 3.084785453001497e-07,
      "loss": 0.72289991,
      "num_input_tokens_seen": 148386200,
      "step": 6871,
      "time_per_iteration": 2.758803129196167
    },
    {
      "auxiliary_loss_clip": 0.01271699,
      "auxiliary_loss_mlp": 0.02567009,
      "balance_loss_clip": 1.0485481,
      "balance_loss_mlp": 0.99992251,
      "epoch": 0.8263091444718331,
      "flos": 23696051339520.0,
      "grad_norm": 4.5130259671257615,
      "language_loss": 0.82297146,
      "learning_rate": 3.080630436613314e-07,
      "loss": 0.86135858,
      "num_input_tokens_seen": 148403970,
      "step": 6872,
      "time_per_iteration": 2.705799102783203
    },
    {
      "auxiliary_loss_clip": 0.01218928,
      "auxiliary_loss_mlp": 0.01025535,
      "balance_loss_clip": 1.04606891,
      "balance_loss_mlp": 1.01865959,
      "epoch": 0.8264293873624722,
      "flos": 17165157523200.0,
      "grad_norm": 2.1044213496194586,
      "language_loss": 0.85895318,
      "learning_rate": 3.076477986880039e-07,
      "loss": 0.88139772,
      "num_input_tokens_seen": 148421765,
      "step": 6873,
      "time_per_iteration": 2.5802536010742188
    },
    {
      "auxiliary_loss_clip": 0.01269005,
      "auxiliary_loss_mlp": 0.01023583,
      "balance_loss_clip": 1.04666543,
      "balance_loss_mlp": 1.01634645,
      "epoch": 0.8265496302531112,
      "flos": 24098645952000.0,
      "grad_norm": 2.3465040842235787,
      "language_loss": 0.69720829,
      "learning_rate": 3.0723281044315986e-07,
      "loss": 0.72013414,
      "num_input_tokens_seen": 148443720,
      "step": 6874,
      "time_per_iteration": 3.6080055236816406
    },
    {
      "auxiliary_loss_clip": 0.01167114,
      "auxiliary_loss_mlp": 0.01026703,
      "balance_loss_clip": 1.04572749,
      "balance_loss_mlp": 1.02020621,
      "epoch": 0.8266698731437504,
      "flos": 14099894599680.0,
      "grad_norm": 2.138434155816014,
      "language_loss": 0.76391959,
      "learning_rate": 3.068180789897521e-07,
      "loss": 0.78585774,
      "num_input_tokens_seen": 148462130,
      "step": 6875,
      "time_per_iteration": 2.550772190093994
    },
    {
      "auxiliary_loss_clip": 0.01227974,
      "auxiliary_loss_mlp": 0.01027892,
      "balance_loss_clip": 1.0490284,
      "balance_loss_mlp": 1.02012002,
      "epoch": 0.8267901160343895,
      "flos": 30777563715840.0,
      "grad_norm": 2.3755490906661616,
      "language_loss": 0.81513262,
      "learning_rate": 3.064036043906966e-07,
      "loss": 0.83769131,
      "num_input_tokens_seen": 148485570,
      "step": 6876,
      "time_per_iteration": 2.717819929122925
    },
    {
      "auxiliary_loss_clip": 0.01329701,
      "auxiliary_loss_mlp": 0.01028927,
      "balance_loss_clip": 1.04533184,
      "balance_loss_mlp": 1.02167642,
      "epoch": 0.8269103589250285,
      "flos": 40624915242240.0,
      "grad_norm": 2.2345010911581205,
      "language_loss": 0.67955077,
      "learning_rate": 3.059893867088668e-07,
      "loss": 0.70313704,
      "num_input_tokens_seen": 148509715,
      "step": 6877,
      "time_per_iteration": 2.869713544845581
    },
    {
      "auxiliary_loss_clip": 0.01220809,
      "auxiliary_loss_mlp": 0.01024507,
      "balance_loss_clip": 1.04707682,
      "balance_loss_mlp": 1.01750362,
      "epoch": 0.8270306018156677,
      "flos": 30263645877120.0,
      "grad_norm": 1.8153693788476575,
      "language_loss": 0.67087936,
      "learning_rate": 3.055754260071004e-07,
      "loss": 0.69333255,
      "num_input_tokens_seen": 148532010,
      "step": 6878,
      "time_per_iteration": 2.7003121376037598
    },
    {
      "auxiliary_loss_clip": 0.01222426,
      "auxiliary_loss_mlp": 0.01022875,
      "balance_loss_clip": 1.04742026,
      "balance_loss_mlp": 1.01607108,
      "epoch": 0.8271508447063067,
      "flos": 25226599812480.0,
      "grad_norm": 2.189257792299753,
      "language_loss": 0.73829222,
      "learning_rate": 3.051617223481948e-07,
      "loss": 0.76074523,
      "num_input_tokens_seen": 148553330,
      "step": 6879,
      "time_per_iteration": 2.687650442123413
    },
    {
      "auxiliary_loss_clip": 0.01231427,
      "auxiliary_loss_mlp": 0.01030753,
      "balance_loss_clip": 1.04456353,
      "balance_loss_mlp": 1.02334762,
      "epoch": 0.8272710875969458,
      "flos": 17566602900480.0,
      "grad_norm": 4.83644842740187,
      "language_loss": 0.75127506,
      "learning_rate": 3.047482757949078e-07,
      "loss": 0.77389693,
      "num_input_tokens_seen": 148570960,
      "step": 6880,
      "time_per_iteration": 2.650857925415039
    },
    {
      "auxiliary_loss_clip": 0.0131674,
      "auxiliary_loss_mlp": 0.02560515,
      "balance_loss_clip": 1.04256737,
      "balance_loss_mlp": 0.99991429,
      "epoch": 0.827391330487585,
      "flos": 19755465886080.0,
      "grad_norm": 2.334594954328534,
      "language_loss": 0.85829675,
      "learning_rate": 3.043350864099605e-07,
      "loss": 0.89706928,
      "num_input_tokens_seen": 148589520,
      "step": 6881,
      "time_per_iteration": 2.6686511039733887
    },
    {
      "auxiliary_loss_clip": 0.01226131,
      "auxiliary_loss_mlp": 0.01027968,
      "balance_loss_clip": 1.04612112,
      "balance_loss_mlp": 1.02077425,
      "epoch": 0.827511573378224,
      "flos": 16835174254080.0,
      "grad_norm": 2.699538401908338,
      "language_loss": 0.80987948,
      "learning_rate": 3.039221542560315e-07,
      "loss": 0.83242053,
      "num_input_tokens_seen": 148606085,
      "step": 6882,
      "time_per_iteration": 2.608227491378784
    },
    {
      "auxiliary_loss_clip": 0.01220124,
      "auxiliary_loss_mlp": 0.01024495,
      "balance_loss_clip": 1.04760647,
      "balance_loss_mlp": 1.01789129,
      "epoch": 0.8276318162688631,
      "flos": 18369242259840.0,
      "grad_norm": 1.9365515716355337,
      "language_loss": 0.73757762,
      "learning_rate": 3.0350947939576356e-07,
      "loss": 0.76002383,
      "num_input_tokens_seen": 148625240,
      "step": 6883,
      "time_per_iteration": 2.610100507736206
    },
    {
      "auxiliary_loss_clip": 0.0123023,
      "auxiliary_loss_mlp": 0.01026995,
      "balance_loss_clip": 1.04899955,
      "balance_loss_mlp": 1.01968443,
      "epoch": 0.8277520591595022,
      "flos": 19352691705600.0,
      "grad_norm": 1.8015335880173156,
      "language_loss": 0.72372264,
      "learning_rate": 3.0309706189175876e-07,
      "loss": 0.74629486,
      "num_input_tokens_seen": 148645075,
      "step": 6884,
      "time_per_iteration": 2.654231071472168
    },
    {
      "auxiliary_loss_clip": 0.01171309,
      "auxiliary_loss_mlp": 0.01000011,
      "balance_loss_clip": 1.00764942,
      "balance_loss_mlp": 0.99911708,
      "epoch": 0.8278723020501413,
      "flos": 67918858329600.0,
      "grad_norm": 0.754428159258695,
      "language_loss": 0.57309759,
      "learning_rate": 3.0268490180658045e-07,
      "loss": 0.59481078,
      "num_input_tokens_seen": 148707855,
      "step": 6885,
      "time_per_iteration": 3.2540221214294434
    },
    {
      "auxiliary_loss_clip": 0.01177862,
      "auxiliary_loss_mlp": 0.01030356,
      "balance_loss_clip": 1.05160117,
      "balance_loss_mlp": 1.0232991,
      "epoch": 0.8279925449407803,
      "flos": 18185738653440.0,
      "grad_norm": 2.9549567070758442,
      "language_loss": 0.79450917,
      "learning_rate": 3.0227299920275305e-07,
      "loss": 0.81659126,
      "num_input_tokens_seen": 148724170,
      "step": 6886,
      "time_per_iteration": 2.5398073196411133
    },
    {
      "auxiliary_loss_clip": 0.013223,
      "auxiliary_loss_mlp": 0.01031158,
      "balance_loss_clip": 1.04641509,
      "balance_loss_mlp": 1.02360654,
      "epoch": 0.8281127878314195,
      "flos": 20631434860800.0,
      "grad_norm": 2.0539805425995206,
      "language_loss": 0.85616624,
      "learning_rate": 3.018613541427613e-07,
      "loss": 0.87970084,
      "num_input_tokens_seen": 148743690,
      "step": 6887,
      "time_per_iteration": 2.710864782333374
    },
    {
      "auxiliary_loss_clip": 0.01172566,
      "auxiliary_loss_mlp": 0.01031217,
      "balance_loss_clip": 1.04773808,
      "balance_loss_mlp": 1.02432036,
      "epoch": 0.8282330307220586,
      "flos": 18004282122240.0,
      "grad_norm": 1.732597852631137,
      "language_loss": 0.74185508,
      "learning_rate": 3.0144996668905243e-07,
      "loss": 0.76389289,
      "num_input_tokens_seen": 148761070,
      "step": 6888,
      "time_per_iteration": 2.5511906147003174
    },
    {
      "auxiliary_loss_clip": 0.01423831,
      "auxiliary_loss_mlp": 0.02566032,
      "balance_loss_clip": 1.04100037,
      "balance_loss_mlp": 0.99991691,
      "epoch": 0.8283532736126976,
      "flos": 20084120352000.0,
      "grad_norm": 2.3880817732015105,
      "language_loss": 0.82215619,
      "learning_rate": 3.010388369040331e-07,
      "loss": 0.86205488,
      "num_input_tokens_seen": 148779730,
      "step": 6889,
      "time_per_iteration": 2.778477907180786
    },
    {
      "auxiliary_loss_clip": 0.01219392,
      "auxiliary_loss_mlp": 0.01030525,
      "balance_loss_clip": 1.04685056,
      "balance_loss_mlp": 1.02372146,
      "epoch": 0.8284735165033368,
      "flos": 31868421805440.0,
      "grad_norm": 1.7171590567151225,
      "language_loss": 0.82859206,
      "learning_rate": 3.0062796485007156e-07,
      "loss": 0.85109121,
      "num_input_tokens_seen": 148800670,
      "step": 6890,
      "time_per_iteration": 2.689568042755127
    },
    {
      "auxiliary_loss_clip": 0.01173606,
      "auxiliary_loss_mlp": 0.02568006,
      "balance_loss_clip": 1.04887938,
      "balance_loss_mlp": 0.99991995,
      "epoch": 0.8285937593939758,
      "flos": 26651319840000.0,
      "grad_norm": 3.716724106018293,
      "language_loss": 0.65730619,
      "learning_rate": 3.002173505894965e-07,
      "loss": 0.69472229,
      "num_input_tokens_seen": 148819820,
      "step": 6891,
      "time_per_iteration": 3.4912078380584717
    },
    {
      "auxiliary_loss_clip": 0.01229057,
      "auxiliary_loss_mlp": 0.01026808,
      "balance_loss_clip": 1.04766464,
      "balance_loss_mlp": 1.01895237,
      "epoch": 0.8287140022846149,
      "flos": 20193683811840.0,
      "grad_norm": 4.305123467572963,
      "language_loss": 0.63304007,
      "learning_rate": 2.998069941845973e-07,
      "loss": 0.6555987,
      "num_input_tokens_seen": 148838890,
      "step": 6892,
      "time_per_iteration": 3.599705457687378
    },
    {
      "auxiliary_loss_clip": 0.01061277,
      "auxiliary_loss_mlp": 0.01002674,
      "balance_loss_clip": 1.00710082,
      "balance_loss_mlp": 1.00179768,
      "epoch": 0.8288342451752541,
      "flos": 70755980019840.0,
      "grad_norm": 0.7080599710986688,
      "language_loss": 0.57366323,
      "learning_rate": 2.993968956976258e-07,
      "loss": 0.59430277,
      "num_input_tokens_seen": 148906635,
      "step": 6893,
      "time_per_iteration": 3.3059377670288086
    },
    {
      "auxiliary_loss_clip": 0.0117985,
      "auxiliary_loss_mlp": 0.01030783,
      "balance_loss_clip": 1.05061102,
      "balance_loss_mlp": 1.02327919,
      "epoch": 0.8289544880658931,
      "flos": 24572235795840.0,
      "grad_norm": 1.8023642303324676,
      "language_loss": 0.70186508,
      "learning_rate": 2.9898705519079313e-07,
      "loss": 0.72397149,
      "num_input_tokens_seen": 148925740,
      "step": 6894,
      "time_per_iteration": 2.6171798706054688
    },
    {
      "auxiliary_loss_clip": 0.01267052,
      "auxiliary_loss_mlp": 0.01023802,
      "balance_loss_clip": 1.04360986,
      "balance_loss_mlp": 1.01665235,
      "epoch": 0.8290747309565322,
      "flos": 22273378387200.0,
      "grad_norm": 2.2601270752193963,
      "language_loss": 0.7492407,
      "learning_rate": 2.985774727262715e-07,
      "loss": 0.77214921,
      "num_input_tokens_seen": 148944585,
      "step": 6895,
      "time_per_iteration": 3.6169822216033936
    },
    {
      "auxiliary_loss_clip": 0.01171789,
      "auxiliary_loss_mlp": 0.01023062,
      "balance_loss_clip": 1.04810953,
      "balance_loss_mlp": 1.01690805,
      "epoch": 0.8291949738471713,
      "flos": 23255570856960.0,
      "grad_norm": 4.04637680392429,
      "language_loss": 0.81520879,
      "learning_rate": 2.981681483661949e-07,
      "loss": 0.83715737,
      "num_input_tokens_seen": 148964170,
      "step": 6896,
      "time_per_iteration": 2.5873208045959473
    },
    {
      "auxiliary_loss_clip": 0.01225365,
      "auxiliary_loss_mlp": 0.01026211,
      "balance_loss_clip": 1.05174756,
      "balance_loss_mlp": 1.01887965,
      "epoch": 0.8293152167378104,
      "flos": 52555768185600.0,
      "grad_norm": 2.1166483665659124,
      "language_loss": 0.71185768,
      "learning_rate": 2.9775908217265633e-07,
      "loss": 0.73437345,
      "num_input_tokens_seen": 148989405,
      "step": 6897,
      "time_per_iteration": 2.8773345947265625
    },
    {
      "auxiliary_loss_clip": 0.01321636,
      "auxiliary_loss_mlp": 0.01000298,
      "balance_loss_clip": 1.00676465,
      "balance_loss_mlp": 0.99936849,
      "epoch": 0.8294354596284494,
      "flos": 63356156294400.0,
      "grad_norm": 0.8298432984683878,
      "language_loss": 0.50343776,
      "learning_rate": 2.9735027420771253e-07,
      "loss": 0.5266571,
      "num_input_tokens_seen": 149049740,
      "step": 6898,
      "time_per_iteration": 3.4237570762634277
    },
    {
      "auxiliary_loss_clip": 0.01266142,
      "auxiliary_loss_mlp": 0.0102501,
      "balance_loss_clip": 1.04746723,
      "balance_loss_mlp": 1.01834881,
      "epoch": 0.8295557025190886,
      "flos": 24827021942400.0,
      "grad_norm": 2.0710057178447725,
      "language_loss": 0.71275198,
      "learning_rate": 2.969417245333774e-07,
      "loss": 0.73566353,
      "num_input_tokens_seen": 149069120,
      "step": 6899,
      "time_per_iteration": 3.8427321910858154
    },
    {
      "auxiliary_loss_clip": 0.01313415,
      "auxiliary_loss_mlp": 0.01025011,
      "balance_loss_clip": 1.04549789,
      "balance_loss_mlp": 1.01857114,
      "epoch": 0.8296759454097277,
      "flos": 25118580637440.0,
      "grad_norm": 2.0846447782156305,
      "language_loss": 0.77815878,
      "learning_rate": 2.9653343321162915e-07,
      "loss": 0.80154312,
      "num_input_tokens_seen": 149088630,
      "step": 6900,
      "time_per_iteration": 2.7146289348602295
    },
    {
      "auxiliary_loss_clip": 0.01321096,
      "auxiliary_loss_mlp": 0.0102438,
      "balance_loss_clip": 1.04769826,
      "balance_loss_mlp": 1.01717103,
      "epoch": 0.8297961883003667,
      "flos": 24132581326080.0,
      "grad_norm": 2.0959099167835644,
      "language_loss": 0.64879942,
      "learning_rate": 2.9612540030440446e-07,
      "loss": 0.67225415,
      "num_input_tokens_seen": 149109175,
      "step": 6901,
      "time_per_iteration": 2.6854827404022217
    },
    {
      "auxiliary_loss_clip": 0.0116763,
      "auxiliary_loss_mlp": 0.01004305,
      "balance_loss_clip": 1.00599253,
      "balance_loss_mlp": 1.00343513,
      "epoch": 0.8299164311910058,
      "flos": 67446561375360.0,
      "grad_norm": 0.843773906398358,
      "language_loss": 0.64030319,
      "learning_rate": 2.9571762587360206e-07,
      "loss": 0.66202253,
      "num_input_tokens_seen": 149165560,
      "step": 6902,
      "time_per_iteration": 3.209599733352661
    },
    {
      "auxiliary_loss_clip": 0.01372472,
      "auxiliary_loss_mlp": 0.01023368,
      "balance_loss_clip": 1.0383141,
      "balance_loss_mlp": 1.01674867,
      "epoch": 0.8300366740816449,
      "flos": 25228682801280.0,
      "grad_norm": 1.743456936098702,
      "language_loss": 0.74433464,
      "learning_rate": 2.953101099810806e-07,
      "loss": 0.76829302,
      "num_input_tokens_seen": 149185165,
      "step": 6903,
      "time_per_iteration": 2.882227897644043
    },
    {
      "auxiliary_loss_clip": 0.01218288,
      "auxiliary_loss_mlp": 0.01025557,
      "balance_loss_clip": 1.04908109,
      "balance_loss_mlp": 1.01924181,
      "epoch": 0.830156916972284,
      "flos": 18041018757120.0,
      "grad_norm": 2.103764252669273,
      "language_loss": 0.82990605,
      "learning_rate": 2.9490285268865965e-07,
      "loss": 0.85234451,
      "num_input_tokens_seen": 149202655,
      "step": 6904,
      "time_per_iteration": 2.5901689529418945
    },
    {
      "auxiliary_loss_clip": 0.01231503,
      "auxiliary_loss_mlp": 0.0102679,
      "balance_loss_clip": 1.05234265,
      "balance_loss_mlp": 1.01920199,
      "epoch": 0.830277159862923,
      "flos": 26322485806080.0,
      "grad_norm": 2.260151225591108,
      "language_loss": 0.79675686,
      "learning_rate": 2.9449585405812085e-07,
      "loss": 0.81933981,
      "num_input_tokens_seen": 149220035,
      "step": 6905,
      "time_per_iteration": 2.706911087036133
    },
    {
      "auxiliary_loss_clip": 0.01223359,
      "auxiliary_loss_mlp": 0.01026607,
      "balance_loss_clip": 1.04680467,
      "balance_loss_mlp": 1.01976752,
      "epoch": 0.8303974027535622,
      "flos": 19938861751680.0,
      "grad_norm": 1.7866222348310496,
      "language_loss": 0.73795938,
      "learning_rate": 2.940891141512043e-07,
      "loss": 0.76045907,
      "num_input_tokens_seen": 149238055,
      "step": 6906,
      "time_per_iteration": 2.741992235183716
    },
    {
      "auxiliary_loss_clip": 0.01274941,
      "auxiliary_loss_mlp": 0.01026914,
      "balance_loss_clip": 1.0446974,
      "balance_loss_mlp": 1.01944542,
      "epoch": 0.8305176456442013,
      "flos": 17165552572800.0,
      "grad_norm": 2.6318841988185837,
      "language_loss": 0.72064793,
      "learning_rate": 2.9368263302961385e-07,
      "loss": 0.74366647,
      "num_input_tokens_seen": 149256755,
      "step": 6907,
      "time_per_iteration": 2.702859401702881
    },
    {
      "auxiliary_loss_clip": 0.01417694,
      "auxiliary_loss_mlp": 0.01023905,
      "balance_loss_clip": 1.04022312,
      "balance_loss_mlp": 1.0168364,
      "epoch": 0.8306378885348403,
      "flos": 25627614226560.0,
      "grad_norm": 1.777146384346947,
      "language_loss": 0.79932821,
      "learning_rate": 2.9327641075501075e-07,
      "loss": 0.82374418,
      "num_input_tokens_seen": 149275745,
      "step": 6908,
      "time_per_iteration": 2.8070156574249268
    },
    {
      "auxiliary_loss_clip": 0.01267783,
      "auxiliary_loss_mlp": 0.01022479,
      "balance_loss_clip": 1.04194319,
      "balance_loss_mlp": 1.01503754,
      "epoch": 0.8307581314254795,
      "flos": 33947864985600.0,
      "grad_norm": 3.3697048303296215,
      "language_loss": 0.66882253,
      "learning_rate": 2.9287044738901866e-07,
      "loss": 0.69172513,
      "num_input_tokens_seen": 149293730,
      "step": 6909,
      "time_per_iteration": 2.727203607559204
    },
    {
      "auxiliary_loss_clip": 0.01227291,
      "auxiliary_loss_mlp": 0.02565538,
      "balance_loss_clip": 1.04907322,
      "balance_loss_mlp": 0.99989522,
      "epoch": 0.8308783743161186,
      "flos": 17562724231680.0,
      "grad_norm": 2.178967488985534,
      "language_loss": 0.90950489,
      "learning_rate": 2.9246474299322274e-07,
      "loss": 0.94743311,
      "num_input_tokens_seen": 149309290,
      "step": 6910,
      "time_per_iteration": 2.5994694232940674
    },
    {
      "auxiliary_loss_clip": 0.01115615,
      "auxiliary_loss_mlp": 0.01000898,
      "balance_loss_clip": 1.00794625,
      "balance_loss_mlp": 0.99991435,
      "epoch": 0.8309986172067576,
      "flos": 69412885649280.0,
      "grad_norm": 0.8819562740062139,
      "language_loss": 0.63116491,
      "learning_rate": 2.920592976291678e-07,
      "loss": 0.65233004,
      "num_input_tokens_seen": 149366620,
      "step": 6911,
      "time_per_iteration": 3.18475341796875
    },
    {
      "auxiliary_loss_clip": 0.0122439,
      "auxiliary_loss_mlp": 0.01028305,
      "balance_loss_clip": 1.04632008,
      "balance_loss_mlp": 1.02136993,
      "epoch": 0.8311188600973968,
      "flos": 22309755886080.0,
      "grad_norm": 2.0840556827873833,
      "language_loss": 0.81042898,
      "learning_rate": 2.916541113583595e-07,
      "loss": 0.83295584,
      "num_input_tokens_seen": 149385120,
      "step": 6912,
      "time_per_iteration": 2.6249215602874756
    },
    {
      "auxiliary_loss_clip": 0.01223611,
      "auxiliary_loss_mlp": 0.01030025,
      "balance_loss_clip": 1.04759896,
      "balance_loss_mlp": 1.02320576,
      "epoch": 0.8312391029880358,
      "flos": 18770077105920.0,
      "grad_norm": 10.876641066251473,
      "language_loss": 0.66291153,
      "learning_rate": 2.912491842422642e-07,
      "loss": 0.68544793,
      "num_input_tokens_seen": 149402825,
      "step": 6913,
      "time_per_iteration": 2.6369078159332275
    },
    {
      "auxiliary_loss_clip": 0.0122419,
      "auxiliary_loss_mlp": 0.0103349,
      "balance_loss_clip": 1.04759741,
      "balance_loss_mlp": 1.02640629,
      "epoch": 0.8313593458786749,
      "flos": 20376648714240.0,
      "grad_norm": 1.8846834540387183,
      "language_loss": 0.70869315,
      "learning_rate": 2.9084451634230857e-07,
      "loss": 0.73126996,
      "num_input_tokens_seen": 149422125,
      "step": 6914,
      "time_per_iteration": 2.6218020915985107
    },
    {
      "auxiliary_loss_clip": 0.01317841,
      "auxiliary_loss_mlp": 0.01026782,
      "balance_loss_clip": 1.04265618,
      "balance_loss_mlp": 1.01952219,
      "epoch": 0.831479588769314,
      "flos": 32124069878400.0,
      "grad_norm": 2.5457551926742523,
      "language_loss": 0.71609962,
      "learning_rate": 2.9044010771988125e-07,
      "loss": 0.73954588,
      "num_input_tokens_seen": 149441940,
      "step": 6915,
      "time_per_iteration": 2.7740542888641357
    },
    {
      "auxiliary_loss_clip": 0.01269037,
      "auxiliary_loss_mlp": 0.0102876,
      "balance_loss_clip": 1.0454576,
      "balance_loss_mlp": 1.02200985,
      "epoch": 0.8315998316599531,
      "flos": 45185929338240.0,
      "grad_norm": 1.7965839409102575,
      "language_loss": 0.72221893,
      "learning_rate": 2.900359584363303e-07,
      "loss": 0.74519694,
      "num_input_tokens_seen": 149465045,
      "step": 6916,
      "time_per_iteration": 2.8624584674835205
    },
    {
      "auxiliary_loss_clip": 0.01371917,
      "auxiliary_loss_mlp": 0.0103003,
      "balance_loss_clip": 1.04522991,
      "balance_loss_mlp": 1.02226937,
      "epoch": 0.8317200745505922,
      "flos": 18363747479040.0,
      "grad_norm": 2.1755449054169533,
      "language_loss": 0.84518856,
      "learning_rate": 2.8963206855296494e-07,
      "loss": 0.86920798,
      "num_input_tokens_seen": 149481285,
      "step": 6917,
      "time_per_iteration": 4.056193113327026
    },
    {
      "auxiliary_loss_clip": 0.01225695,
      "auxiliary_loss_mlp": 0.01023466,
      "balance_loss_clip": 1.04745221,
      "balance_loss_mlp": 1.01645088,
      "epoch": 0.8318403174412313,
      "flos": 24206557386240.0,
      "grad_norm": 2.0767989667324005,
      "language_loss": 0.77338642,
      "learning_rate": 2.892284381310548e-07,
      "loss": 0.79587799,
      "num_input_tokens_seen": 149502700,
      "step": 6918,
      "time_per_iteration": 3.5085866451263428
    },
    {
      "auxiliary_loss_clip": 0.01267605,
      "auxiliary_loss_mlp": 0.01025588,
      "balance_loss_clip": 1.04560399,
      "balance_loss_mlp": 1.01848888,
      "epoch": 0.8319605603318704,
      "flos": 22418780641920.0,
      "grad_norm": 2.4738552832796663,
      "language_loss": 0.72418672,
      "learning_rate": 2.888250672318302e-07,
      "loss": 0.74711871,
      "num_input_tokens_seen": 149520100,
      "step": 6919,
      "time_per_iteration": 2.6608033180236816
    },
    {
      "auxiliary_loss_clip": 0.01176489,
      "auxiliary_loss_mlp": 0.01023933,
      "balance_loss_clip": 1.05003119,
      "balance_loss_mlp": 1.01721215,
      "epoch": 0.8320808032225094,
      "flos": 37414501459200.0,
      "grad_norm": 1.8353553890799,
      "language_loss": 0.68739438,
      "learning_rate": 2.884219559164831e-07,
      "loss": 0.70939851,
      "num_input_tokens_seen": 149543245,
      "step": 6920,
      "time_per_iteration": 2.6901814937591553
    },
    {
      "auxiliary_loss_clip": 0.01222918,
      "auxiliary_loss_mlp": 0.0103264,
      "balance_loss_clip": 1.04925704,
      "balance_loss_mlp": 1.02595234,
      "epoch": 0.8322010461131486,
      "flos": 12787395638400.0,
      "grad_norm": 2.146992057335875,
      "language_loss": 0.81454015,
      "learning_rate": 2.880191042461635e-07,
      "loss": 0.83709574,
      "num_input_tokens_seen": 149559185,
      "step": 6921,
      "time_per_iteration": 3.522088050842285
    },
    {
      "auxiliary_loss_clip": 0.01377608,
      "auxiliary_loss_mlp": 0.01025756,
      "balance_loss_clip": 1.04302859,
      "balance_loss_mlp": 1.01926458,
      "epoch": 0.8323212890037877,
      "flos": 15815455050240.0,
      "grad_norm": 1.76239463731894,
      "language_loss": 0.79850978,
      "learning_rate": 2.876165122819849e-07,
      "loss": 0.82254338,
      "num_input_tokens_seen": 149577165,
      "step": 6922,
      "time_per_iteration": 2.7109344005584717
    },
    {
      "auxiliary_loss_clip": 0.01171715,
      "auxiliary_loss_mlp": 0.01021712,
      "balance_loss_clip": 1.04855609,
      "balance_loss_mlp": 1.01534379,
      "epoch": 0.8324415318944267,
      "flos": 21719276208000.0,
      "grad_norm": 3.905092301915365,
      "language_loss": 0.79612559,
      "learning_rate": 2.872141800850201e-07,
      "loss": 0.81805986,
      "num_input_tokens_seen": 149594340,
      "step": 6923,
      "time_per_iteration": 2.537079095840454
    },
    {
      "auxiliary_loss_clip": 0.01171982,
      "auxiliary_loss_mlp": 0.01029237,
      "balance_loss_clip": 1.04801202,
      "balance_loss_mlp": 1.02219212,
      "epoch": 0.8325617747850659,
      "flos": 34198700636160.0,
      "grad_norm": 1.7319705342379115,
      "language_loss": 0.73702455,
      "learning_rate": 2.868121077163024e-07,
      "loss": 0.75903678,
      "num_input_tokens_seen": 149613895,
      "step": 6924,
      "time_per_iteration": 2.698092460632324
    },
    {
      "auxiliary_loss_clip": 0.01223901,
      "auxiliary_loss_mlp": 0.01026266,
      "balance_loss_clip": 1.04526591,
      "balance_loss_mlp": 1.01933074,
      "epoch": 0.8326820176757049,
      "flos": 18369457741440.0,
      "grad_norm": 6.458254260913927,
      "language_loss": 0.72386765,
      "learning_rate": 2.864102952368257e-07,
      "loss": 0.74636936,
      "num_input_tokens_seen": 149631820,
      "step": 6925,
      "time_per_iteration": 3.468461513519287
    },
    {
      "auxiliary_loss_clip": 0.013618,
      "auxiliary_loss_mlp": 0.01030798,
      "balance_loss_clip": 1.03579974,
      "balance_loss_mlp": 1.02355576,
      "epoch": 0.832802260566344,
      "flos": 35991325716480.0,
      "grad_norm": 1.5207405727163525,
      "language_loss": 0.59554154,
      "learning_rate": 2.860087427075444e-07,
      "loss": 0.61946756,
      "num_input_tokens_seen": 149656070,
      "step": 6926,
      "time_per_iteration": 2.8231353759765625
    },
    {
      "auxiliary_loss_clip": 0.01268783,
      "auxiliary_loss_mlp": 0.01020624,
      "balance_loss_clip": 1.04382992,
      "balance_loss_mlp": 1.01406479,
      "epoch": 0.8329225034569832,
      "flos": 14244434928000.0,
      "grad_norm": 2.8132246145999726,
      "language_loss": 0.86183989,
      "learning_rate": 2.856074501893744e-07,
      "loss": 0.88473403,
      "num_input_tokens_seen": 149671270,
      "step": 6927,
      "time_per_iteration": 2.639317512512207
    },
    {
      "auxiliary_loss_clip": 0.01224734,
      "auxiliary_loss_mlp": 0.01029542,
      "balance_loss_clip": 1.05088115,
      "balance_loss_mlp": 1.02250862,
      "epoch": 0.8330427463476222,
      "flos": 18077468083200.0,
      "grad_norm": 13.405841674803577,
      "language_loss": 0.81749249,
      "learning_rate": 2.8520641774319054e-07,
      "loss": 0.8400352,
      "num_input_tokens_seen": 149689360,
      "step": 6928,
      "time_per_iteration": 2.607154130935669
    },
    {
      "auxiliary_loss_clip": 0.01274066,
      "auxiliary_loss_mlp": 0.01029051,
      "balance_loss_clip": 1.04186761,
      "balance_loss_mlp": 1.02205396,
      "epoch": 0.8331629892382613,
      "flos": 18040839189120.0,
      "grad_norm": 2.1402604423044727,
      "language_loss": 0.75803304,
      "learning_rate": 2.848056454298309e-07,
      "loss": 0.78106427,
      "num_input_tokens_seen": 149706685,
      "step": 6929,
      "time_per_iteration": 2.6168859004974365
    },
    {
      "auxiliary_loss_clip": 0.01269796,
      "auxiliary_loss_mlp": 0.01024097,
      "balance_loss_clip": 1.04751384,
      "balance_loss_mlp": 1.01698339,
      "epoch": 0.8332832321289004,
      "flos": 17457398576640.0,
      "grad_norm": 2.6997826160034406,
      "language_loss": 0.6552844,
      "learning_rate": 2.844051333100905e-07,
      "loss": 0.67822337,
      "num_input_tokens_seen": 149724230,
      "step": 6930,
      "time_per_iteration": 2.6249876022338867
    },
    {
      "auxiliary_loss_clip": 0.01270562,
      "auxiliary_loss_mlp": 0.01025907,
      "balance_loss_clip": 1.04830158,
      "balance_loss_mlp": 1.01969028,
      "epoch": 0.8334034750195395,
      "flos": 15084852416640.0,
      "grad_norm": 1.9364926147375603,
      "language_loss": 0.83754587,
      "learning_rate": 2.840048814447269e-07,
      "loss": 0.86051059,
      "num_input_tokens_seen": 149742395,
      "step": 6931,
      "time_per_iteration": 2.6255948543548584
    },
    {
      "auxiliary_loss_clip": 0.01269042,
      "auxiliary_loss_mlp": 0.01032956,
      "balance_loss_clip": 1.04314268,
      "balance_loss_mlp": 1.02608085,
      "epoch": 0.8335237179101785,
      "flos": 19427170556160.0,
      "grad_norm": 2.908789640714984,
      "language_loss": 0.74261725,
      "learning_rate": 2.836048898944587e-07,
      "loss": 0.76563722,
      "num_input_tokens_seen": 149760820,
      "step": 6932,
      "time_per_iteration": 2.6810662746429443
    },
    {
      "auxiliary_loss_clip": 0.0127278,
      "auxiliary_loss_mlp": 0.01025932,
      "balance_loss_clip": 1.04660726,
      "balance_loss_mlp": 1.01989412,
      "epoch": 0.8336439608008177,
      "flos": 21762046327680.0,
      "grad_norm": 2.586953748936526,
      "language_loss": 0.72916806,
      "learning_rate": 2.832051587199642e-07,
      "loss": 0.75215513,
      "num_input_tokens_seen": 149778075,
      "step": 6933,
      "time_per_iteration": 2.689176559448242
    },
    {
      "auxiliary_loss_clip": 0.0111602,
      "auxiliary_loss_mlp": 0.01001863,
      "balance_loss_clip": 1.00701022,
      "balance_loss_mlp": 1.00096846,
      "epoch": 0.8337642036914568,
      "flos": 59702783990400.0,
      "grad_norm": 0.8082923431065165,
      "language_loss": 0.57683122,
      "learning_rate": 2.828056879818821e-07,
      "loss": 0.59801006,
      "num_input_tokens_seen": 149837150,
      "step": 6934,
      "time_per_iteration": 3.1253604888916016
    },
    {
      "auxiliary_loss_clip": 0.01321251,
      "auxiliary_loss_mlp": 0.01025717,
      "balance_loss_clip": 1.04091573,
      "balance_loss_mlp": 1.01934218,
      "epoch": 0.8338844465820958,
      "flos": 27162185022720.0,
      "grad_norm": 2.025056036385724,
      "language_loss": 0.83550274,
      "learning_rate": 2.824064777408117e-07,
      "loss": 0.85897243,
      "num_input_tokens_seen": 149856940,
      "step": 6935,
      "time_per_iteration": 2.7809760570526123
    },
    {
      "auxiliary_loss_clip": 0.01222146,
      "auxiliary_loss_mlp": 0.01031078,
      "balance_loss_clip": 1.04869223,
      "balance_loss_mlp": 1.02403593,
      "epoch": 0.8340046894727349,
      "flos": 30481264425600.0,
      "grad_norm": 2.313732438252162,
      "language_loss": 0.76224792,
      "learning_rate": 2.8200752805731263e-07,
      "loss": 0.78478014,
      "num_input_tokens_seen": 149879930,
      "step": 6936,
      "time_per_iteration": 2.6804275512695312
    },
    {
      "auxiliary_loss_clip": 0.0122174,
      "auxiliary_loss_mlp": 0.01021495,
      "balance_loss_clip": 1.04918635,
      "balance_loss_mlp": 1.01445317,
      "epoch": 0.834124932363374,
      "flos": 27126166659840.0,
      "grad_norm": 1.4402777532725128,
      "language_loss": 0.8108725,
      "learning_rate": 2.8160883899190625e-07,
      "loss": 0.83330482,
      "num_input_tokens_seen": 149903200,
      "step": 6937,
      "time_per_iteration": 2.7079339027404785
    },
    {
      "auxiliary_loss_clip": 0.01318246,
      "auxiliary_loss_mlp": 0.01030071,
      "balance_loss_clip": 1.04676509,
      "balance_loss_mlp": 1.02303791,
      "epoch": 0.8342451752540131,
      "flos": 24569865498240.0,
      "grad_norm": 3.0277230091059515,
      "language_loss": 0.73554862,
      "learning_rate": 2.8121041060507234e-07,
      "loss": 0.75903177,
      "num_input_tokens_seen": 149922230,
      "step": 6938,
      "time_per_iteration": 2.731416940689087
    },
    {
      "auxiliary_loss_clip": 0.0122775,
      "auxiliary_loss_mlp": 0.01023163,
      "balance_loss_clip": 1.04737055,
      "balance_loss_mlp": 1.01639771,
      "epoch": 0.8343654181446521,
      "flos": 26615085995520.0,
      "grad_norm": 1.8274513334432378,
      "language_loss": 0.71552277,
      "learning_rate": 2.808122429572528e-07,
      "loss": 0.73803192,
      "num_input_tokens_seen": 149942435,
      "step": 6939,
      "time_per_iteration": 2.7479119300842285
    },
    {
      "auxiliary_loss_clip": 0.01329129,
      "auxiliary_loss_mlp": 0.0102275,
      "balance_loss_clip": 1.04504263,
      "balance_loss_mlp": 1.01606202,
      "epoch": 0.8344856610352913,
      "flos": 20777268078720.0,
      "grad_norm": 2.588692238038136,
      "language_loss": 0.75493282,
      "learning_rate": 2.804143361088489e-07,
      "loss": 0.77845168,
      "num_input_tokens_seen": 149961615,
      "step": 6940,
      "time_per_iteration": 2.6946322917938232
    },
    {
      "auxiliary_loss_clip": 0.01269886,
      "auxiliary_loss_mlp": 0.01026559,
      "balance_loss_clip": 1.04566097,
      "balance_loss_mlp": 1.01954317,
      "epoch": 0.8346059039259304,
      "flos": 26095960684800.0,
      "grad_norm": 2.453199428921057,
      "language_loss": 0.77823657,
      "learning_rate": 2.8001669012022277e-07,
      "loss": 0.80120099,
      "num_input_tokens_seen": 149979585,
      "step": 6941,
      "time_per_iteration": 2.698209762573242
    },
    {
      "auxiliary_loss_clip": 0.01223115,
      "auxiliary_loss_mlp": 0.01027128,
      "balance_loss_clip": 1.0498805,
      "balance_loss_mlp": 1.02012503,
      "epoch": 0.8347261468165694,
      "flos": 29027708755200.0,
      "grad_norm": 4.621431341325348,
      "language_loss": 0.69343984,
      "learning_rate": 2.7961930505169795e-07,
      "loss": 0.7159422,
      "num_input_tokens_seen": 150003830,
      "step": 6942,
      "time_per_iteration": 2.696486711502075
    },
    {
      "auxiliary_loss_clip": 0.01224581,
      "auxiliary_loss_mlp": 0.02566335,
      "balance_loss_clip": 1.04885948,
      "balance_loss_mlp": 0.99991179,
      "epoch": 0.8348463897072086,
      "flos": 26396461866240.0,
      "grad_norm": 3.4003930313250104,
      "language_loss": 0.76281965,
      "learning_rate": 2.792221809635558e-07,
      "loss": 0.8007288,
      "num_input_tokens_seen": 150024460,
      "step": 6943,
      "time_per_iteration": 3.609750509262085
    },
    {
      "auxiliary_loss_clip": 0.01469307,
      "auxiliary_loss_mlp": 0.01028728,
      "balance_loss_clip": 1.04090583,
      "balance_loss_mlp": 1.02133989,
      "epoch": 0.8349666325978476,
      "flos": 23367720096000.0,
      "grad_norm": 2.9563872782296907,
      "language_loss": 0.74685276,
      "learning_rate": 2.788253179160411e-07,
      "loss": 0.77183306,
      "num_input_tokens_seen": 150045620,
      "step": 6944,
      "time_per_iteration": 3.853145122528076
    },
    {
      "auxiliary_loss_clip": 0.01271094,
      "auxiliary_loss_mlp": 0.01027672,
      "balance_loss_clip": 1.04582751,
      "balance_loss_mlp": 1.02071023,
      "epoch": 0.8350868754884867,
      "flos": 12896528135040.0,
      "grad_norm": 2.7946560310935524,
      "language_loss": 0.65156782,
      "learning_rate": 2.7842871596935725e-07,
      "loss": 0.67455554,
      "num_input_tokens_seen": 150064135,
      "step": 6945,
      "time_per_iteration": 2.808785915374756
    },
    {
      "auxiliary_loss_clip": 0.01125583,
      "auxiliary_loss_mlp": 0.01022735,
      "balance_loss_clip": 1.04722857,
      "balance_loss_mlp": 1.01535904,
      "epoch": 0.8352071183791259,
      "flos": 26505522535680.0,
      "grad_norm": 1.7792155423429783,
      "language_loss": 0.69263428,
      "learning_rate": 2.780323751836682e-07,
      "loss": 0.71411753,
      "num_input_tokens_seen": 150085350,
      "step": 6946,
      "time_per_iteration": 2.6427671909332275
    },
    {
      "auxiliary_loss_clip": 0.0127222,
      "auxiliary_loss_mlp": 0.0256467,
      "balance_loss_clip": 1.04401147,
      "balance_loss_mlp": 0.99990654,
      "epoch": 0.8353273612697649,
      "flos": 20668063754880.0,
      "grad_norm": 2.442520966307865,
      "language_loss": 0.79038143,
      "learning_rate": 2.7763629561909876e-07,
      "loss": 0.82875037,
      "num_input_tokens_seen": 150106180,
      "step": 6947,
      "time_per_iteration": 3.9412951469421387
    },
    {
      "auxiliary_loss_clip": 0.01172403,
      "auxiliary_loss_mlp": 0.01026622,
      "balance_loss_clip": 1.0487349,
      "balance_loss_mlp": 1.01995838,
      "epoch": 0.835447604160404,
      "flos": 19754137082880.0,
      "grad_norm": 2.123831448841596,
      "language_loss": 0.76815766,
      "learning_rate": 2.772404773357335e-07,
      "loss": 0.7901479,
      "num_input_tokens_seen": 150125585,
      "step": 6948,
      "time_per_iteration": 2.587252140045166
    },
    {
      "auxiliary_loss_clip": 0.01314831,
      "auxiliary_loss_mlp": 0.01026633,
      "balance_loss_clip": 1.04339015,
      "balance_loss_mlp": 1.01980782,
      "epoch": 0.8355678470510431,
      "flos": 23435842239360.0,
      "grad_norm": 1.834714662085777,
      "language_loss": 0.78457052,
      "learning_rate": 2.7684492039361853e-07,
      "loss": 0.80798519,
      "num_input_tokens_seen": 150144810,
      "step": 6949,
      "time_per_iteration": 2.7391738891601562
    },
    {
      "auxiliary_loss_clip": 0.01176254,
      "auxiliary_loss_mlp": 0.01028102,
      "balance_loss_clip": 1.05027294,
      "balance_loss_mlp": 1.02125072,
      "epoch": 0.8356880899416822,
      "flos": 21214588164480.0,
      "grad_norm": 1.855612512171314,
      "language_loss": 0.8370142,
      "learning_rate": 2.764496248527586e-07,
      "loss": 0.85905778,
      "num_input_tokens_seen": 150163785,
      "step": 6950,
      "time_per_iteration": 2.6354825496673584
    },
    {
      "auxiliary_loss_clip": 0.01228951,
      "auxiliary_loss_mlp": 0.01024851,
      "balance_loss_clip": 1.04546976,
      "balance_loss_mlp": 1.01808596,
      "epoch": 0.8358083328323213,
      "flos": 28037543466240.0,
      "grad_norm": 2.077607041694871,
      "language_loss": 0.78621441,
      "learning_rate": 2.760545907731211e-07,
      "loss": 0.80875242,
      "num_input_tokens_seen": 150184360,
      "step": 6951,
      "time_per_iteration": 2.713735818862915
    },
    {
      "auxiliary_loss_clip": 0.01224838,
      "auxiliary_loss_mlp": 0.01031492,
      "balance_loss_clip": 1.04625809,
      "balance_loss_mlp": 1.0244348,
      "epoch": 0.8359285757229604,
      "flos": 27783655159680.0,
      "grad_norm": 1.6782910810734124,
      "language_loss": 0.67878717,
      "learning_rate": 2.75659818214631e-07,
      "loss": 0.70135045,
      "num_input_tokens_seen": 150205465,
      "step": 6952,
      "time_per_iteration": 3.6054940223693848
    },
    {
      "auxiliary_loss_clip": 0.0127661,
      "auxiliary_loss_mlp": 0.01027421,
      "balance_loss_clip": 1.04798818,
      "balance_loss_mlp": 1.02064991,
      "epoch": 0.8360488186135995,
      "flos": 21435115714560.0,
      "grad_norm": 1.8730782944132267,
      "language_loss": 0.78294599,
      "learning_rate": 2.752653072371749e-07,
      "loss": 0.80598629,
      "num_input_tokens_seen": 150224900,
      "step": 6953,
      "time_per_iteration": 2.6986758708953857
    },
    {
      "auxiliary_loss_clip": 0.01312977,
      "auxiliary_loss_mlp": 0.01025928,
      "balance_loss_clip": 1.04659331,
      "balance_loss_mlp": 1.01928496,
      "epoch": 0.8361690615042385,
      "flos": 27632327160960.0,
      "grad_norm": 2.5207324046993485,
      "language_loss": 0.74864709,
      "learning_rate": 2.7487105790060105e-07,
      "loss": 0.77203619,
      "num_input_tokens_seen": 150244310,
      "step": 6954,
      "time_per_iteration": 2.8156509399414062
    },
    {
      "auxiliary_loss_clip": 0.01222853,
      "auxiliary_loss_mlp": 0.01023426,
      "balance_loss_clip": 1.04591966,
      "balance_loss_mlp": 1.01691413,
      "epoch": 0.8362893043948777,
      "flos": 39202529598720.0,
      "grad_norm": 2.145139650828975,
      "language_loss": 0.69610465,
      "learning_rate": 2.7447707026471587e-07,
      "loss": 0.71856749,
      "num_input_tokens_seen": 150267285,
      "step": 6955,
      "time_per_iteration": 2.788815975189209
    },
    {
      "auxiliary_loss_clip": 0.01320321,
      "auxiliary_loss_mlp": 0.01022918,
      "balance_loss_clip": 1.04487121,
      "balance_loss_mlp": 1.01656425,
      "epoch": 0.8364095472855168,
      "flos": 24785329230720.0,
      "grad_norm": 2.18109656368797,
      "language_loss": 0.80051899,
      "learning_rate": 2.740833443892874e-07,
      "loss": 0.82395136,
      "num_input_tokens_seen": 150285455,
      "step": 6956,
      "time_per_iteration": 2.7479331493377686
    },
    {
      "auxiliary_loss_clip": 0.01272564,
      "auxiliary_loss_mlp": 0.01022702,
      "balance_loss_clip": 1.04563057,
      "balance_loss_mlp": 1.01597822,
      "epoch": 0.8365297901761558,
      "flos": 22743412784640.0,
      "grad_norm": 1.9415049374115332,
      "language_loss": 0.79507583,
      "learning_rate": 2.7368988033404327e-07,
      "loss": 0.81802851,
      "num_input_tokens_seen": 150302970,
      "step": 6957,
      "time_per_iteration": 2.7203803062438965
    },
    {
      "auxiliary_loss_clip": 0.01320995,
      "auxiliary_loss_mlp": 0.01025512,
      "balance_loss_clip": 1.04589963,
      "balance_loss_mlp": 1.01893413,
      "epoch": 0.836650033066795,
      "flos": 28396003242240.0,
      "grad_norm": 1.8329169115275334,
      "language_loss": 0.84572268,
      "learning_rate": 2.732966781586712e-07,
      "loss": 0.86918783,
      "num_input_tokens_seen": 150322715,
      "step": 6958,
      "time_per_iteration": 2.7367193698883057
    },
    {
      "auxiliary_loss_clip": 0.01217872,
      "auxiliary_loss_mlp": 0.01022533,
      "balance_loss_clip": 1.04525709,
      "balance_loss_mlp": 1.0161258,
      "epoch": 0.836770275957434,
      "flos": 22236857233920.0,
      "grad_norm": 3.961298451414,
      "language_loss": 0.66943747,
      "learning_rate": 2.729037379228205e-07,
      "loss": 0.69184154,
      "num_input_tokens_seen": 150342900,
      "step": 6959,
      "time_per_iteration": 2.6965510845184326
    },
    {
      "auxiliary_loss_clip": 0.01271478,
      "auxiliary_loss_mlp": 0.01024728,
      "balance_loss_clip": 1.04782891,
      "balance_loss_mlp": 1.01814818,
      "epoch": 0.8368905188480731,
      "flos": 22491930689280.0,
      "grad_norm": 1.5551924515625721,
      "language_loss": 0.80670381,
      "learning_rate": 2.725110596860998e-07,
      "loss": 0.8296659,
      "num_input_tokens_seen": 150363580,
      "step": 6960,
      "time_per_iteration": 2.6930713653564453
    },
    {
      "auxiliary_loss_clip": 0.01370513,
      "auxiliary_loss_mlp": 0.01022537,
      "balance_loss_clip": 1.04532516,
      "balance_loss_mlp": 1.01567972,
      "epoch": 0.8370107617387123,
      "flos": 13370405287680.0,
      "grad_norm": 7.218823076356944,
      "language_loss": 0.70256764,
      "learning_rate": 2.7211864350807776e-07,
      "loss": 0.72649813,
      "num_input_tokens_seen": 150381780,
      "step": 6961,
      "time_per_iteration": 2.8124196529388428
    },
    {
      "auxiliary_loss_clip": 0.01174923,
      "auxiliary_loss_mlp": 0.01029766,
      "balance_loss_clip": 1.04952347,
      "balance_loss_mlp": 1.0227536,
      "epoch": 0.8371310046293513,
      "flos": 25261289372160.0,
      "grad_norm": 1.7090739433206692,
      "language_loss": 0.73978359,
      "learning_rate": 2.717264894482836e-07,
      "loss": 0.76183045,
      "num_input_tokens_seen": 150402120,
      "step": 6962,
      "time_per_iteration": 2.658801317214966
    },
    {
      "auxiliary_loss_clip": 0.01225653,
      "auxiliary_loss_mlp": 0.0102458,
      "balance_loss_clip": 1.04973149,
      "balance_loss_mlp": 1.01752591,
      "epoch": 0.8372512475199904,
      "flos": 19792705311360.0,
      "grad_norm": 4.465641354311493,
      "language_loss": 0.81336296,
      "learning_rate": 2.7133459756620646e-07,
      "loss": 0.83586526,
      "num_input_tokens_seen": 150419315,
      "step": 6963,
      "time_per_iteration": 2.65259051322937
    },
    {
      "auxiliary_loss_clip": 0.01222332,
      "auxiliary_loss_mlp": 0.01023279,
      "balance_loss_clip": 1.04743218,
      "balance_loss_mlp": 1.01630867,
      "epoch": 0.8373714904106295,
      "flos": 19391224020480.0,
      "grad_norm": 1.6716700607501263,
      "language_loss": 0.73809528,
      "learning_rate": 2.7094296792129733e-07,
      "loss": 0.76055145,
      "num_input_tokens_seen": 150438915,
      "step": 6964,
      "time_per_iteration": 2.636814832687378
    },
    {
      "auxiliary_loss_clip": 0.01223231,
      "auxiliary_loss_mlp": 0.01024486,
      "balance_loss_clip": 1.04788923,
      "balance_loss_mlp": 1.01756597,
      "epoch": 0.8374917333012686,
      "flos": 14975935401600.0,
      "grad_norm": 1.8030330556774168,
      "language_loss": 0.75590527,
      "learning_rate": 2.7055160057296424e-07,
      "loss": 0.77838242,
      "num_input_tokens_seen": 150456155,
      "step": 6965,
      "time_per_iteration": 2.7629599571228027
    },
    {
      "auxiliary_loss_clip": 0.01322368,
      "auxiliary_loss_mlp": 0.0102519,
      "balance_loss_clip": 1.04435384,
      "balance_loss_mlp": 1.01843643,
      "epoch": 0.8376119761919076,
      "flos": 30331839847680.0,
      "grad_norm": 1.822700580963271,
      "language_loss": 0.7256763,
      "learning_rate": 2.7016049558057896e-07,
      "loss": 0.74915189,
      "num_input_tokens_seen": 150478115,
      "step": 6966,
      "time_per_iteration": 2.7857275009155273
    },
    {
      "auxiliary_loss_clip": 0.01221335,
      "auxiliary_loss_mlp": 0.01032072,
      "balance_loss_clip": 1.04792166,
      "balance_loss_mlp": 1.02525365,
      "epoch": 0.8377322190825467,
      "flos": 29423336129280.0,
      "grad_norm": 1.9718118547578198,
      "language_loss": 0.7079103,
      "learning_rate": 2.6976965300347074e-07,
      "loss": 0.73044443,
      "num_input_tokens_seen": 150500725,
      "step": 6967,
      "time_per_iteration": 2.72890305519104
    },
    {
      "auxiliary_loss_clip": 0.01268329,
      "auxiliary_loss_mlp": 0.01024749,
      "balance_loss_clip": 1.04330051,
      "balance_loss_mlp": 1.01791239,
      "epoch": 0.8378524619731859,
      "flos": 26687086807680.0,
      "grad_norm": 2.6973108579610403,
      "language_loss": 0.69376159,
      "learning_rate": 2.693790729009309e-07,
      "loss": 0.71669233,
      "num_input_tokens_seen": 150522335,
      "step": 6968,
      "time_per_iteration": 2.716850757598877
    },
    {
      "auxiliary_loss_clip": 0.0127114,
      "auxiliary_loss_mlp": 0.01026268,
      "balance_loss_clip": 1.04580832,
      "balance_loss_mlp": 1.01918066,
      "epoch": 0.8379727048638249,
      "flos": 20703866636160.0,
      "grad_norm": 1.8717485992238072,
      "language_loss": 0.8841188,
      "learning_rate": 2.6898875533220946e-07,
      "loss": 0.90709287,
      "num_input_tokens_seen": 150541640,
      "step": 6969,
      "time_per_iteration": 3.5974512100219727
    },
    {
      "auxiliary_loss_clip": 0.01171934,
      "auxiliary_loss_mlp": 0.01024739,
      "balance_loss_clip": 1.05088997,
      "balance_loss_mlp": 1.01873064,
      "epoch": 0.838092947754464,
      "flos": 20084084438400.0,
      "grad_norm": 2.1335213990408057,
      "language_loss": 0.81687868,
      "learning_rate": 2.685987003565171e-07,
      "loss": 0.83884537,
      "num_input_tokens_seen": 150559680,
      "step": 6970,
      "time_per_iteration": 3.584362268447876
    },
    {
      "auxiliary_loss_clip": 0.01317565,
      "auxiliary_loss_mlp": 0.0102949,
      "balance_loss_clip": 1.04818892,
      "balance_loss_mlp": 1.02227235,
      "epoch": 0.8382131906451031,
      "flos": 18113270964480.0,
      "grad_norm": 2.804961155343125,
      "language_loss": 0.75315738,
      "learning_rate": 2.6820890803302566e-07,
      "loss": 0.7766279,
      "num_input_tokens_seen": 150575205,
      "step": 6971,
      "time_per_iteration": 2.715165615081787
    },
    {
      "auxiliary_loss_clip": 0.01178488,
      "auxiliary_loss_mlp": 0.01021864,
      "balance_loss_clip": 1.05163372,
      "balance_loss_mlp": 1.01472378,
      "epoch": 0.8383334335357422,
      "flos": 17092653920640.0,
      "grad_norm": 2.9343085191467364,
      "language_loss": 0.81961918,
      "learning_rate": 2.6781937842086557e-07,
      "loss": 0.84162271,
      "num_input_tokens_seen": 150593995,
      "step": 6972,
      "time_per_iteration": 2.6390252113342285
    },
    {
      "auxiliary_loss_clip": 0.01225919,
      "auxiliary_loss_mlp": 0.01026845,
      "balance_loss_clip": 1.04794931,
      "balance_loss_mlp": 1.02010655,
      "epoch": 0.8384536764263812,
      "flos": 20704728562560.0,
      "grad_norm": 1.9318507936087848,
      "language_loss": 0.67846859,
      "learning_rate": 2.6743011157912933e-07,
      "loss": 0.70099622,
      "num_input_tokens_seen": 150613715,
      "step": 6973,
      "time_per_iteration": 2.660090923309326
    },
    {
      "auxiliary_loss_clip": 0.01371437,
      "auxiliary_loss_mlp": 0.0102496,
      "balance_loss_clip": 1.03991938,
      "balance_loss_mlp": 1.01794124,
      "epoch": 0.8385739193170204,
      "flos": 28986842056320.0,
      "grad_norm": 1.7379823836881858,
      "language_loss": 0.65085596,
      "learning_rate": 2.6704110756686725e-07,
      "loss": 0.67481995,
      "num_input_tokens_seen": 150634540,
      "step": 6974,
      "time_per_iteration": 3.7116384506225586
    },
    {
      "auxiliary_loss_clip": 0.0127228,
      "auxiliary_loss_mlp": 0.02566719,
      "balance_loss_clip": 1.04372835,
      "balance_loss_mlp": 0.99988616,
      "epoch": 0.8386941622076595,
      "flos": 23438068882560.0,
      "grad_norm": 1.7516380599222245,
      "language_loss": 0.83848083,
      "learning_rate": 2.6665236644309085e-07,
      "loss": 0.87687087,
      "num_input_tokens_seen": 150654850,
      "step": 6975,
      "time_per_iteration": 2.7332377433776855
    },
    {
      "auxiliary_loss_clip": 0.01224296,
      "auxiliary_loss_mlp": 0.01022032,
      "balance_loss_clip": 1.04757726,
      "balance_loss_mlp": 1.01559138,
      "epoch": 0.8388144050982985,
      "flos": 23002724044800.0,
      "grad_norm": 1.9728370347649666,
      "language_loss": 0.79515129,
      "learning_rate": 2.662638882667727e-07,
      "loss": 0.81761456,
      "num_input_tokens_seen": 150673790,
      "step": 6976,
      "time_per_iteration": 2.625753164291382
    },
    {
      "auxiliary_loss_clip": 0.01177622,
      "auxiliary_loss_mlp": 0.01030087,
      "balance_loss_clip": 1.05081892,
      "balance_loss_mlp": 1.0229013,
      "epoch": 0.8389346479889377,
      "flos": 24280353878400.0,
      "grad_norm": 1.9138655212848097,
      "language_loss": 0.73313999,
      "learning_rate": 2.658756730968443e-07,
      "loss": 0.75521708,
      "num_input_tokens_seen": 150692255,
      "step": 6977,
      "time_per_iteration": 3.565002202987671
    },
    {
      "auxiliary_loss_clip": 0.01275834,
      "auxiliary_loss_mlp": 0.01030059,
      "balance_loss_clip": 1.04945409,
      "balance_loss_mlp": 1.02341938,
      "epoch": 0.8390548908795767,
      "flos": 21215019127680.0,
      "grad_norm": 2.224239056945468,
      "language_loss": 0.88242602,
      "learning_rate": 2.654877209921975e-07,
      "loss": 0.90548497,
      "num_input_tokens_seen": 150709790,
      "step": 6978,
      "time_per_iteration": 2.6831705570220947
    },
    {
      "auxiliary_loss_clip": 0.01380819,
      "auxiliary_loss_mlp": 0.01024394,
      "balance_loss_clip": 1.0418222,
      "balance_loss_mlp": 1.01725888,
      "epoch": 0.8391751337702158,
      "flos": 35627299332480.0,
      "grad_norm": 2.5948364632569363,
      "language_loss": 0.63262105,
      "learning_rate": 2.651000320116843e-07,
      "loss": 0.65667319,
      "num_input_tokens_seen": 150730675,
      "step": 6979,
      "time_per_iteration": 2.8210246562957764
    },
    {
      "auxiliary_loss_clip": 0.01320638,
      "auxiliary_loss_mlp": 0.02569674,
      "balance_loss_clip": 1.04398644,
      "balance_loss_mlp": 0.99988437,
      "epoch": 0.839295376660855,
      "flos": 21325229032320.0,
      "grad_norm": 1.8669515809627832,
      "language_loss": 0.7605021,
      "learning_rate": 2.647126062141163e-07,
      "loss": 0.79940522,
      "num_input_tokens_seen": 150749750,
      "step": 6980,
      "time_per_iteration": 2.7357916831970215
    },
    {
      "auxiliary_loss_clip": 0.01276647,
      "auxiliary_loss_mlp": 0.010265,
      "balance_loss_clip": 1.04478121,
      "balance_loss_mlp": 1.01973808,
      "epoch": 0.839415619551494,
      "flos": 18442535961600.0,
      "grad_norm": 2.1557227280286226,
      "language_loss": 0.84149146,
      "learning_rate": 2.643254436582669e-07,
      "loss": 0.86452293,
      "num_input_tokens_seen": 150769240,
      "step": 6981,
      "time_per_iteration": 2.6301429271698
    },
    {
      "auxiliary_loss_clip": 0.01374276,
      "auxiliary_loss_mlp": 0.01024045,
      "balance_loss_clip": 1.04412675,
      "balance_loss_mlp": 1.01696432,
      "epoch": 0.8395358624421331,
      "flos": 23221958705280.0,
      "grad_norm": 3.7438401411608724,
      "language_loss": 0.82820672,
      "learning_rate": 2.6393854440286743e-07,
      "loss": 0.8521899,
      "num_input_tokens_seen": 150788410,
      "step": 6982,
      "time_per_iteration": 2.765263557434082
    },
    {
      "auxiliary_loss_clip": 0.01173289,
      "auxiliary_loss_mlp": 0.01027072,
      "balance_loss_clip": 1.05044031,
      "balance_loss_mlp": 1.02002728,
      "epoch": 0.8396561053327722,
      "flos": 24381657210240.0,
      "grad_norm": 3.386382888284045,
      "language_loss": 0.70848775,
      "learning_rate": 2.6355190850661045e-07,
      "loss": 0.73049134,
      "num_input_tokens_seen": 150805245,
      "step": 6983,
      "time_per_iteration": 2.663285732269287
    },
    {
      "auxiliary_loss_clip": 0.0127361,
      "auxiliary_loss_mlp": 0.01030608,
      "balance_loss_clip": 1.04834092,
      "balance_loss_mlp": 1.02406931,
      "epoch": 0.8397763482234113,
      "flos": 22237755073920.0,
      "grad_norm": 18.212430697750005,
      "language_loss": 0.86777258,
      "learning_rate": 2.631655360281486e-07,
      "loss": 0.89081478,
      "num_input_tokens_seen": 150824920,
      "step": 6984,
      "time_per_iteration": 2.68686580657959
    },
    {
      "auxiliary_loss_clip": 0.01128741,
      "auxiliary_loss_mlp": 0.02569381,
      "balance_loss_clip": 1.0490551,
      "balance_loss_mlp": 0.99993467,
      "epoch": 0.8398965911140504,
      "flos": 22163743100160.0,
      "grad_norm": 1.9508929227996006,
      "language_loss": 0.6557619,
      "learning_rate": 2.6277942702609323e-07,
      "loss": 0.69274306,
      "num_input_tokens_seen": 150844400,
      "step": 6985,
      "time_per_iteration": 2.6494362354278564
    },
    {
      "auxiliary_loss_clip": 0.01322838,
      "auxiliary_loss_mlp": 0.01026559,
      "balance_loss_clip": 1.0456841,
      "balance_loss_mlp": 1.01970446,
      "epoch": 0.8400168340046895,
      "flos": 21542775753600.0,
      "grad_norm": 1.997301140240795,
      "language_loss": 0.87370908,
      "learning_rate": 2.623935815590186e-07,
      "loss": 0.89720309,
      "num_input_tokens_seen": 150862780,
      "step": 6986,
      "time_per_iteration": 2.733858108520508
    },
    {
      "auxiliary_loss_clip": 0.01273917,
      "auxiliary_loss_mlp": 0.01031303,
      "balance_loss_clip": 1.04770994,
      "balance_loss_mlp": 1.02384686,
      "epoch": 0.8401370768953286,
      "flos": 22491966602880.0,
      "grad_norm": 1.8685444116564762,
      "language_loss": 0.81142294,
      "learning_rate": 2.6200799968545516e-07,
      "loss": 0.83447516,
      "num_input_tokens_seen": 150883075,
      "step": 6987,
      "time_per_iteration": 2.6888091564178467
    },
    {
      "auxiliary_loss_clip": 0.01162284,
      "auxiliary_loss_mlp": 0.01002319,
      "balance_loss_clip": 1.00679183,
      "balance_loss_mlp": 1.0013535,
      "epoch": 0.8402573197859676,
      "flos": 59238890818560.0,
      "grad_norm": 0.788387543996549,
      "language_loss": 0.56391138,
      "learning_rate": 2.616226814638969e-07,
      "loss": 0.58555746,
      "num_input_tokens_seen": 150948180,
      "step": 6988,
      "time_per_iteration": 3.226602554321289
    },
    {
      "auxiliary_loss_clip": 0.01172568,
      "auxiliary_loss_mlp": 0.01028722,
      "balance_loss_clip": 1.04778385,
      "balance_loss_mlp": 1.02190351,
      "epoch": 0.8403775626766068,
      "flos": 22674608282880.0,
      "grad_norm": 1.9946322185106191,
      "language_loss": 0.77450222,
      "learning_rate": 2.612376269527954e-07,
      "loss": 0.79651511,
      "num_input_tokens_seen": 150967885,
      "step": 6989,
      "time_per_iteration": 2.6563477516174316
    },
    {
      "auxiliary_loss_clip": 0.01269543,
      "auxiliary_loss_mlp": 0.01027851,
      "balance_loss_clip": 1.04787612,
      "balance_loss_mlp": 1.02077842,
      "epoch": 0.8404978055672458,
      "flos": 19609704495360.0,
      "grad_norm": 1.9611520227239005,
      "language_loss": 0.67613661,
      "learning_rate": 2.608528362105635e-07,
      "loss": 0.69911057,
      "num_input_tokens_seen": 150987255,
      "step": 6990,
      "time_per_iteration": 2.6791694164276123
    },
    {
      "auxiliary_loss_clip": 0.01324208,
      "auxiliary_loss_mlp": 0.01025482,
      "balance_loss_clip": 1.04320812,
      "balance_loss_mlp": 1.0190537,
      "epoch": 0.8406180484578849,
      "flos": 27526929678720.0,
      "grad_norm": 1.919641763794121,
      "language_loss": 0.73652941,
      "learning_rate": 2.6046830929557374e-07,
      "loss": 0.76002634,
      "num_input_tokens_seen": 151006905,
      "step": 6991,
      "time_per_iteration": 2.753127098083496
    },
    {
      "auxiliary_loss_clip": 0.01320521,
      "auxiliary_loss_mlp": 0.01027666,
      "balance_loss_clip": 1.04544115,
      "balance_loss_mlp": 1.02032852,
      "epoch": 0.8407382913485241,
      "flos": 22127473342080.0,
      "grad_norm": 1.8723503984206455,
      "language_loss": 0.85150683,
      "learning_rate": 2.6008404626615776e-07,
      "loss": 0.87498868,
      "num_input_tokens_seen": 151025405,
      "step": 6992,
      "time_per_iteration": 2.6967387199401855
    },
    {
      "auxiliary_loss_clip": 0.01227167,
      "auxiliary_loss_mlp": 0.01018731,
      "balance_loss_clip": 1.04997849,
      "balance_loss_mlp": 1.01199913,
      "epoch": 0.8408585342391631,
      "flos": 13918473982080.0,
      "grad_norm": 2.4606987387828854,
      "language_loss": 0.73846138,
      "learning_rate": 2.597000471806092e-07,
      "loss": 0.76092041,
      "num_input_tokens_seen": 151041970,
      "step": 6993,
      "time_per_iteration": 2.6547951698303223
    },
    {
      "auxiliary_loss_clip": 0.01268935,
      "auxiliary_loss_mlp": 0.01026642,
      "balance_loss_clip": 1.04833055,
      "balance_loss_mlp": 1.01925683,
      "epoch": 0.8409787771298022,
      "flos": 20187865808640.0,
      "grad_norm": 2.3541202433929103,
      "language_loss": 0.73067451,
      "learning_rate": 2.593163120971793e-07,
      "loss": 0.75363022,
      "num_input_tokens_seen": 151060835,
      "step": 6994,
      "time_per_iteration": 2.66286301612854
    },
    {
      "auxiliary_loss_clip": 0.01363245,
      "auxiliary_loss_mlp": 0.01032122,
      "balance_loss_clip": 1.03809285,
      "balance_loss_mlp": 1.02571726,
      "epoch": 0.8410990200204413,
      "flos": 23142523777920.0,
      "grad_norm": 2.1363827360811265,
      "language_loss": 0.69088113,
      "learning_rate": 2.5893284107408165e-07,
      "loss": 0.71483481,
      "num_input_tokens_seen": 151078205,
      "step": 6995,
      "time_per_iteration": 4.578260660171509
    },
    {
      "auxiliary_loss_clip": 0.01377785,
      "auxiliary_loss_mlp": 0.01025701,
      "balance_loss_clip": 1.04744816,
      "balance_loss_mlp": 1.01901639,
      "epoch": 0.8412192629110804,
      "flos": 24027219757440.0,
      "grad_norm": 1.763646527505341,
      "language_loss": 0.78293908,
      "learning_rate": 2.5854963416948726e-07,
      "loss": 0.80697393,
      "num_input_tokens_seen": 151100470,
      "step": 6996,
      "time_per_iteration": 2.7925093173980713
    },
    {
      "auxiliary_loss_clip": 0.0127255,
      "auxiliary_loss_mlp": 0.01028352,
      "balance_loss_clip": 1.03877425,
      "balance_loss_mlp": 1.02161694,
      "epoch": 0.8413395058017195,
      "flos": 25591703604480.0,
      "grad_norm": 1.7662825373072963,
      "language_loss": 0.69668353,
      "learning_rate": 2.5816669144152816e-07,
      "loss": 0.71969259,
      "num_input_tokens_seen": 151121650,
      "step": 6997,
      "time_per_iteration": 2.7456624507904053
    },
    {
      "auxiliary_loss_clip": 0.01060813,
      "auxiliary_loss_mlp": 0.01001732,
      "balance_loss_clip": 1.00646877,
      "balance_loss_mlp": 1.00085545,
      "epoch": 0.8414597486923585,
      "flos": 63635396624640.0,
      "grad_norm": 0.8439460086025359,
      "language_loss": 0.6630246,
      "learning_rate": 2.5778401294829777e-07,
      "loss": 0.68365008,
      "num_input_tokens_seen": 151180390,
      "step": 6998,
      "time_per_iteration": 3.2494678497314453
    },
    {
      "auxiliary_loss_clip": 0.01220516,
      "auxiliary_loss_mlp": 0.02565063,
      "balance_loss_clip": 1.04838943,
      "balance_loss_mlp": 0.99988496,
      "epoch": 0.8415799915829977,
      "flos": 19098731571840.0,
      "grad_norm": 2.3949006516302718,
      "language_loss": 0.65442485,
      "learning_rate": 2.574015987478473e-07,
      "loss": 0.69228059,
      "num_input_tokens_seen": 151198520,
      "step": 6999,
      "time_per_iteration": 3.5540902614593506
    },
    {
      "auxiliary_loss_clip": 0.01277568,
      "auxiliary_loss_mlp": 0.0102818,
      "balance_loss_clip": 1.04695594,
      "balance_loss_mlp": 1.02024662,
      "epoch": 0.8417002344736367,
      "flos": 19821612781440.0,
      "grad_norm": 3.9319219911072585,
      "language_loss": 0.87255764,
      "learning_rate": 2.570194488981887e-07,
      "loss": 0.8956151,
      "num_input_tokens_seen": 151215065,
      "step": 7000,
      "time_per_iteration": 2.6505143642425537
    },
    {
      "auxiliary_loss_clip": 0.01061144,
      "auxiliary_loss_mlp": 0.01000913,
      "balance_loss_clip": 1.00694895,
      "balance_loss_mlp": 1.00001907,
      "epoch": 0.8418204773642758,
      "flos": 62161516834560.0,
      "grad_norm": 0.8515036852540623,
      "language_loss": 0.60252786,
      "learning_rate": 2.566375634572939e-07,
      "loss": 0.62314832,
      "num_input_tokens_seen": 151275705,
      "step": 7001,
      "time_per_iteration": 3.1394131183624268
    },
    {
      "auxiliary_loss_clip": 0.01324581,
      "auxiliary_loss_mlp": 0.01026405,
      "balance_loss_clip": 1.04245007,
      "balance_loss_mlp": 1.01885605,
      "epoch": 0.841940720254915,
      "flos": 17092905315840.0,
      "grad_norm": 2.2057648486642303,
      "language_loss": 0.762918,
      "learning_rate": 2.562559424830943e-07,
      "loss": 0.78642786,
      "num_input_tokens_seen": 151293665,
      "step": 7002,
      "time_per_iteration": 2.678239107131958
    },
    {
      "auxiliary_loss_clip": 0.01272933,
      "auxiliary_loss_mlp": 0.01024618,
      "balance_loss_clip": 1.04425132,
      "balance_loss_mlp": 1.01763868,
      "epoch": 0.842060963145554,
      "flos": 16283586026880.0,
      "grad_norm": 2.214067949972543,
      "language_loss": 0.70792389,
      "learning_rate": 2.5587458603348256e-07,
      "loss": 0.73089939,
      "num_input_tokens_seen": 151310955,
      "step": 7003,
      "time_per_iteration": 3.5713565349578857
    },
    {
      "auxiliary_loss_clip": 0.0132046,
      "auxiliary_loss_mlp": 0.01024659,
      "balance_loss_clip": 1.04487681,
      "balance_loss_mlp": 1.01766145,
      "epoch": 0.8421812060361931,
      "flos": 21908238681600.0,
      "grad_norm": 2.759303016300212,
      "language_loss": 0.84170234,
      "learning_rate": 2.554934941663085e-07,
      "loss": 0.86515355,
      "num_input_tokens_seen": 151328490,
      "step": 7004,
      "time_per_iteration": 2.777838706970215
    },
    {
      "auxiliary_loss_clip": 0.01322814,
      "auxiliary_loss_mlp": 0.01030847,
      "balance_loss_clip": 1.04545295,
      "balance_loss_mlp": 1.02286649,
      "epoch": 0.8423014489268322,
      "flos": 27777693502080.0,
      "grad_norm": 2.525990181291804,
      "language_loss": 0.73490113,
      "learning_rate": 2.5511266693938484e-07,
      "loss": 0.75843775,
      "num_input_tokens_seen": 151346950,
      "step": 7005,
      "time_per_iteration": 2.7402217388153076
    },
    {
      "auxiliary_loss_clip": 0.01270017,
      "auxiliary_loss_mlp": 0.01023077,
      "balance_loss_clip": 1.04753709,
      "balance_loss_mlp": 1.01537573,
      "epoch": 0.8424216918174713,
      "flos": 25117610970240.0,
      "grad_norm": 1.487879343494548,
      "language_loss": 0.77618456,
      "learning_rate": 2.547321044104822e-07,
      "loss": 0.79911542,
      "num_input_tokens_seen": 151368445,
      "step": 7006,
      "time_per_iteration": 2.672760009765625
    },
    {
      "auxiliary_loss_clip": 0.01176931,
      "auxiliary_loss_mlp": 0.01026483,
      "balance_loss_clip": 1.05105698,
      "balance_loss_mlp": 1.01955342,
      "epoch": 0.8425419347081103,
      "flos": 24748448941440.0,
      "grad_norm": 1.9254224989353974,
      "language_loss": 0.76935679,
      "learning_rate": 2.5435180663733113e-07,
      "loss": 0.7913909,
      "num_input_tokens_seen": 151388745,
      "step": 7007,
      "time_per_iteration": 2.6327967643737793
    },
    {
      "auxiliary_loss_clip": 0.0128045,
      "auxiliary_loss_mlp": 0.01026707,
      "balance_loss_clip": 1.04305148,
      "balance_loss_mlp": 1.01950336,
      "epoch": 0.8426621775987495,
      "flos": 24820916630400.0,
      "grad_norm": 2.4991797143518806,
      "language_loss": 0.71477592,
      "learning_rate": 2.539717736776241e-07,
      "loss": 0.73784751,
      "num_input_tokens_seen": 151404970,
      "step": 7008,
      "time_per_iteration": 2.759664297103882
    },
    {
      "auxiliary_loss_clip": 0.01219556,
      "auxiliary_loss_mlp": 0.01025179,
      "balance_loss_clip": 1.04795396,
      "balance_loss_mlp": 1.01851535,
      "epoch": 0.8427824204893886,
      "flos": 23550074467200.0,
      "grad_norm": 1.4331295574071072,
      "language_loss": 0.76660585,
      "learning_rate": 2.535920055890097e-07,
      "loss": 0.7890532,
      "num_input_tokens_seen": 151426265,
      "step": 7009,
      "time_per_iteration": 2.6285464763641357
    },
    {
      "auxiliary_loss_clip": 0.01370439,
      "auxiliary_loss_mlp": 0.0102396,
      "balance_loss_clip": 1.04150259,
      "balance_loss_mlp": 1.01620245,
      "epoch": 0.8429026633800276,
      "flos": 16143858120960.0,
      "grad_norm": 4.372176812756397,
      "language_loss": 0.64381343,
      "learning_rate": 2.5321250242910006e-07,
      "loss": 0.66775739,
      "num_input_tokens_seen": 151444180,
      "step": 7010,
      "time_per_iteration": 2.724247932434082
    },
    {
      "auxiliary_loss_clip": 0.01174377,
      "auxiliary_loss_mlp": 0.01026423,
      "balance_loss_clip": 1.05065119,
      "balance_loss_mlp": 1.01952624,
      "epoch": 0.8430229062706668,
      "flos": 22198540400640.0,
      "grad_norm": 1.6764383112069592,
      "language_loss": 0.86284482,
      "learning_rate": 2.5283326425546493e-07,
      "loss": 0.88485289,
      "num_input_tokens_seen": 151463290,
      "step": 7011,
      "time_per_iteration": 2.6184706687927246
    },
    {
      "auxiliary_loss_clip": 0.01313915,
      "auxiliary_loss_mlp": 0.01024831,
      "balance_loss_clip": 1.04704273,
      "balance_loss_mlp": 1.01826251,
      "epoch": 0.8431431491613058,
      "flos": 35330317683840.0,
      "grad_norm": 2.01018445198061,
      "language_loss": 0.69590378,
      "learning_rate": 2.5245429112563443e-07,
      "loss": 0.71929121,
      "num_input_tokens_seen": 151483965,
      "step": 7012,
      "time_per_iteration": 2.8739497661590576
    },
    {
      "auxiliary_loss_clip": 0.01222251,
      "auxiliary_loss_mlp": 0.01029287,
      "balance_loss_clip": 1.04909396,
      "balance_loss_mlp": 1.02220035,
      "epoch": 0.8432633920519449,
      "flos": 25812374808960.0,
      "grad_norm": 2.08715583637686,
      "language_loss": 0.82368481,
      "learning_rate": 2.5207558309709865e-07,
      "loss": 0.84620023,
      "num_input_tokens_seen": 151503700,
      "step": 7013,
      "time_per_iteration": 2.7245538234710693
    },
    {
      "auxiliary_loss_clip": 0.01228937,
      "auxiliary_loss_mlp": 0.02505906,
      "balance_loss_clip": 1.00650144,
      "balance_loss_mlp": 0.99985486,
      "epoch": 0.8433836349425841,
      "flos": 64959531592320.0,
      "grad_norm": 0.6554455748902237,
      "language_loss": 0.56227279,
      "learning_rate": 2.516971402273065e-07,
      "loss": 0.59962118,
      "num_input_tokens_seen": 151569765,
      "step": 7014,
      "time_per_iteration": 3.2818844318389893
    },
    {
      "auxiliary_loss_clip": 0.01269873,
      "auxiliary_loss_mlp": 0.01026662,
      "balance_loss_clip": 1.04389524,
      "balance_loss_mlp": 1.02048063,
      "epoch": 0.8435038778332231,
      "flos": 20229989483520.0,
      "grad_norm": 1.751948238260273,
      "language_loss": 0.67777121,
      "learning_rate": 2.513189625736687e-07,
      "loss": 0.70073658,
      "num_input_tokens_seen": 151586660,
      "step": 7015,
      "time_per_iteration": 2.6959545612335205
    },
    {
      "auxiliary_loss_clip": 0.01324923,
      "auxiliary_loss_mlp": 0.01029155,
      "balance_loss_clip": 1.04430807,
      "balance_loss_mlp": 1.02191949,
      "epoch": 0.8436241207238622,
      "flos": 20992229020800.0,
      "grad_norm": 19.661450113071776,
      "language_loss": 0.71724874,
      "learning_rate": 2.509410501935534e-07,
      "loss": 0.74078953,
      "num_input_tokens_seen": 151602295,
      "step": 7016,
      "time_per_iteration": 2.7230887413024902
    },
    {
      "auxiliary_loss_clip": 0.01276142,
      "auxiliary_loss_mlp": 0.01029303,
      "balance_loss_clip": 1.04824543,
      "balance_loss_mlp": 1.02185524,
      "epoch": 0.8437443636145013,
      "flos": 14682257804160.0,
      "grad_norm": 3.269475351381643,
      "language_loss": 0.75883532,
      "learning_rate": 2.5056340314429116e-07,
      "loss": 0.7818898,
      "num_input_tokens_seen": 151619760,
      "step": 7017,
      "time_per_iteration": 2.6632111072540283
    },
    {
      "auxiliary_loss_clip": 0.0137501,
      "auxiliary_loss_mlp": 0.01026721,
      "balance_loss_clip": 1.04142976,
      "balance_loss_mlp": 1.01942503,
      "epoch": 0.8438646065051404,
      "flos": 21608814908160.0,
      "grad_norm": 2.567824047552669,
      "language_loss": 0.80383009,
      "learning_rate": 2.5018602148316904e-07,
      "loss": 0.82784742,
      "num_input_tokens_seen": 151635795,
      "step": 7018,
      "time_per_iteration": 2.7147743701934814
    },
    {
      "auxiliary_loss_clip": 0.01171196,
      "auxiliary_loss_mlp": 0.01023101,
      "balance_loss_clip": 1.04981017,
      "balance_loss_mlp": 1.01670504,
      "epoch": 0.8439848493957794,
      "flos": 23289937194240.0,
      "grad_norm": 18.53417404336236,
      "language_loss": 0.80503786,
      "learning_rate": 2.498089052674359e-07,
      "loss": 0.82698083,
      "num_input_tokens_seen": 151653770,
      "step": 7019,
      "time_per_iteration": 2.657071352005005
    },
    {
      "auxiliary_loss_clip": 0.01223862,
      "auxiliary_loss_mlp": 0.01027945,
      "balance_loss_clip": 1.04973078,
      "balance_loss_mlp": 1.02121568,
      "epoch": 0.8441050922864186,
      "flos": 19719339782400.0,
      "grad_norm": 14.068440739014978,
      "language_loss": 0.75289905,
      "learning_rate": 2.494320545543007e-07,
      "loss": 0.77541715,
      "num_input_tokens_seen": 151673340,
      "step": 7020,
      "time_per_iteration": 2.701582908630371
    },
    {
      "auxiliary_loss_clip": 0.01178517,
      "auxiliary_loss_mlp": 0.01025299,
      "balance_loss_clip": 1.05079579,
      "balance_loss_mlp": 1.01831317,
      "epoch": 0.8442253351770577,
      "flos": 21835268202240.0,
      "grad_norm": 1.7468694521278074,
      "language_loss": 0.66574496,
      "learning_rate": 2.490554694009308e-07,
      "loss": 0.68778312,
      "num_input_tokens_seen": 151694205,
      "step": 7021,
      "time_per_iteration": 4.492710828781128
    },
    {
      "auxiliary_loss_clip": 0.01228034,
      "auxiliary_loss_mlp": 0.01024374,
      "balance_loss_clip": 1.04728425,
      "balance_loss_mlp": 1.01744175,
      "epoch": 0.8443455780676967,
      "flos": 34346365447680.0,
      "grad_norm": 1.7101307838423652,
      "language_loss": 0.78118527,
      "learning_rate": 2.4867914986445426e-07,
      "loss": 0.80370933,
      "num_input_tokens_seen": 151716595,
      "step": 7022,
      "time_per_iteration": 2.7526743412017822
    },
    {
      "auxiliary_loss_clip": 0.01275589,
      "auxiliary_loss_mlp": 0.0102441,
      "balance_loss_clip": 1.04333282,
      "balance_loss_mlp": 1.01736736,
      "epoch": 0.8444658209583359,
      "flos": 48214599281280.0,
      "grad_norm": 2.1338110082829305,
      "language_loss": 0.71531069,
      "learning_rate": 2.483030960019581e-07,
      "loss": 0.73831069,
      "num_input_tokens_seen": 151740525,
      "step": 7023,
      "time_per_iteration": 2.9122376441955566
    },
    {
      "auxiliary_loss_clip": 0.01277208,
      "auxiliary_loss_mlp": 0.00999927,
      "balance_loss_clip": 1.00748706,
      "balance_loss_mlp": 0.99894983,
      "epoch": 0.8445860638489749,
      "flos": 68484773105280.0,
      "grad_norm": 0.7440198633075089,
      "language_loss": 0.55386925,
      "learning_rate": 2.479273078704891e-07,
      "loss": 0.57664061,
      "num_input_tokens_seen": 151793890,
      "step": 7024,
      "time_per_iteration": 3.1329386234283447
    },
    {
      "auxiliary_loss_clip": 0.01261062,
      "auxiliary_loss_mlp": 0.00999435,
      "balance_loss_clip": 1.00709724,
      "balance_loss_mlp": 0.99850476,
      "epoch": 0.844706306739614,
      "flos": 62833331882880.0,
      "grad_norm": 0.7803348961499635,
      "language_loss": 0.64706492,
      "learning_rate": 2.475517855270552e-07,
      "loss": 0.66966987,
      "num_input_tokens_seen": 151853970,
      "step": 7025,
      "time_per_iteration": 4.179176092147827
    },
    {
      "auxiliary_loss_clip": 0.01172379,
      "auxiliary_loss_mlp": 0.01026629,
      "balance_loss_clip": 1.04960275,
      "balance_loss_mlp": 1.02002144,
      "epoch": 0.8448265496302532,
      "flos": 14976114969600.0,
      "grad_norm": 2.824545013286073,
      "language_loss": 0.7315191,
      "learning_rate": 2.4717652902862143e-07,
      "loss": 0.75350916,
      "num_input_tokens_seen": 151872945,
      "step": 7026,
      "time_per_iteration": 2.6281204223632812
    },
    {
      "auxiliary_loss_clip": 0.01174679,
      "auxiliary_loss_mlp": 0.01023627,
      "balance_loss_clip": 1.04672837,
      "balance_loss_mlp": 1.01684117,
      "epoch": 0.8449467925208922,
      "flos": 23441265192960.0,
      "grad_norm": 2.137664204080232,
      "language_loss": 0.81399393,
      "learning_rate": 2.4680153843211495e-07,
      "loss": 0.83597696,
      "num_input_tokens_seen": 151892875,
      "step": 7027,
      "time_per_iteration": 2.698202610015869
    },
    {
      "auxiliary_loss_clip": 0.01269919,
      "auxiliary_loss_mlp": 0.0102751,
      "balance_loss_clip": 1.04958105,
      "balance_loss_mlp": 1.01988029,
      "epoch": 0.8450670354115313,
      "flos": 22748045639040.0,
      "grad_norm": 1.8060796812083695,
      "language_loss": 0.72571415,
      "learning_rate": 2.464268137944212e-07,
      "loss": 0.74868846,
      "num_input_tokens_seen": 151914170,
      "step": 7028,
      "time_per_iteration": 2.724576950073242
    },
    {
      "auxiliary_loss_clip": 0.01367867,
      "auxiliary_loss_mlp": 0.01027092,
      "balance_loss_clip": 1.04366565,
      "balance_loss_mlp": 1.01925993,
      "epoch": 0.8451872783021703,
      "flos": 29825571605760.0,
      "grad_norm": 3.827668425624646,
      "language_loss": 0.78368664,
      "learning_rate": 2.46052355172385e-07,
      "loss": 0.80763626,
      "num_input_tokens_seen": 151932210,
      "step": 7029,
      "time_per_iteration": 2.7969510555267334
    },
    {
      "auxiliary_loss_clip": 0.01175522,
      "auxiliary_loss_mlp": 0.01027194,
      "balance_loss_clip": 1.05125213,
      "balance_loss_mlp": 1.01979423,
      "epoch": 0.8453075211928095,
      "flos": 21870029589120.0,
      "grad_norm": 1.8959078076364035,
      "language_loss": 0.74766886,
      "learning_rate": 2.456781626228128e-07,
      "loss": 0.769696,
      "num_input_tokens_seen": 151951715,
      "step": 7030,
      "time_per_iteration": 3.5684847831726074
    },
    {
      "auxiliary_loss_clip": 0.01276692,
      "auxiliary_loss_mlp": 0.02506066,
      "balance_loss_clip": 1.00620365,
      "balance_loss_mlp": 0.99988502,
      "epoch": 0.8454277640834486,
      "flos": 58751869288320.0,
      "grad_norm": 0.9183987759127491,
      "language_loss": 0.66192454,
      "learning_rate": 2.453042362024675e-07,
      "loss": 0.69975209,
      "num_input_tokens_seen": 152004960,
      "step": 7031,
      "time_per_iteration": 3.355557680130005
    },
    {
      "auxiliary_loss_clip": 0.01173112,
      "auxiliary_loss_mlp": 0.01023256,
      "balance_loss_clip": 1.04912174,
      "balance_loss_mlp": 1.01672328,
      "epoch": 0.8455480069740876,
      "flos": 27090076469760.0,
      "grad_norm": 2.445286158309845,
      "language_loss": 0.73289609,
      "learning_rate": 2.449305759680751e-07,
      "loss": 0.75485975,
      "num_input_tokens_seen": 152026285,
      "step": 7032,
      "time_per_iteration": 2.675636053085327
    },
    {
      "auxiliary_loss_clip": 0.01314747,
      "auxiliary_loss_mlp": 0.01023099,
      "balance_loss_clip": 1.04722559,
      "balance_loss_mlp": 1.01620889,
      "epoch": 0.8456682498647268,
      "flos": 27198670262400.0,
      "grad_norm": 1.392519197878043,
      "language_loss": 0.75306726,
      "learning_rate": 2.445571819763188e-07,
      "loss": 0.77644575,
      "num_input_tokens_seen": 152048585,
      "step": 7033,
      "time_per_iteration": 2.772468328475952
    },
    {
      "auxiliary_loss_clip": 0.01174426,
      "auxiliary_loss_mlp": 0.01029498,
      "balance_loss_clip": 1.05044591,
      "balance_loss_mlp": 1.02253318,
      "epoch": 0.8457884927553658,
      "flos": 20631901737600.0,
      "grad_norm": 3.104431144221861,
      "language_loss": 0.58782512,
      "learning_rate": 2.4418405428384227e-07,
      "loss": 0.60986435,
      "num_input_tokens_seen": 152068795,
      "step": 7034,
      "time_per_iteration": 2.6435599327087402
    },
    {
      "auxiliary_loss_clip": 0.01171819,
      "auxiliary_loss_mlp": 0.02569311,
      "balance_loss_clip": 1.04910398,
      "balance_loss_mlp": 0.999897,
      "epoch": 0.8459087356460049,
      "flos": 15299023259520.0,
      "grad_norm": 3.391276581792001,
      "language_loss": 0.71910119,
      "learning_rate": 2.4381119294724864e-07,
      "loss": 0.75651252,
      "num_input_tokens_seen": 152086240,
      "step": 7035,
      "time_per_iteration": 2.5936970710754395
    },
    {
      "auxiliary_loss_clip": 0.01171537,
      "auxiliary_loss_mlp": 0.0102362,
      "balance_loss_clip": 1.04809999,
      "balance_loss_mlp": 1.01695275,
      "epoch": 0.846028978536644,
      "flos": 18843155326080.0,
      "grad_norm": 2.055132960719797,
      "language_loss": 0.54088676,
      "learning_rate": 2.434385980231004e-07,
      "loss": 0.56283832,
      "num_input_tokens_seen": 152105080,
      "step": 7036,
      "time_per_iteration": 2.534133195877075
    },
    {
      "auxiliary_loss_clip": 0.01221676,
      "auxiliary_loss_mlp": 0.01022284,
      "balance_loss_clip": 1.04724967,
      "balance_loss_mlp": 1.0150423,
      "epoch": 0.8461492214272831,
      "flos": 52661740285440.0,
      "grad_norm": 1.698273843779693,
      "language_loss": 0.65497833,
      "learning_rate": 2.4306626956792043e-07,
      "loss": 0.67741793,
      "num_input_tokens_seen": 152130025,
      "step": 7037,
      "time_per_iteration": 2.897120475769043
    },
    {
      "auxiliary_loss_clip": 0.01222296,
      "auxiliary_loss_mlp": 0.01028602,
      "balance_loss_clip": 1.04604316,
      "balance_loss_mlp": 1.02147079,
      "epoch": 0.8462694643179222,
      "flos": 18588405093120.0,
      "grad_norm": 1.7449002619439167,
      "language_loss": 0.75543702,
      "learning_rate": 2.4269420763819017e-07,
      "loss": 0.777946,
      "num_input_tokens_seen": 152148070,
      "step": 7038,
      "time_per_iteration": 2.6249570846557617
    },
    {
      "auxiliary_loss_clip": 0.01219837,
      "auxiliary_loss_mlp": 0.01026435,
      "balance_loss_clip": 1.04804802,
      "balance_loss_mlp": 1.02012599,
      "epoch": 0.8463897072085613,
      "flos": 24387080163840.0,
      "grad_norm": 2.799760069310872,
      "language_loss": 0.83413142,
      "learning_rate": 2.4232241229035223e-07,
      "loss": 0.85659415,
      "num_input_tokens_seen": 152165825,
      "step": 7039,
      "time_per_iteration": 2.6393847465515137
    },
    {
      "auxiliary_loss_clip": 0.01116591,
      "auxiliary_loss_mlp": 0.0100118,
      "balance_loss_clip": 1.00695372,
      "balance_loss_mlp": 1.00028598,
      "epoch": 0.8465099500992004,
      "flos": 68702140258560.0,
      "grad_norm": 0.750931789776879,
      "language_loss": 0.56718463,
      "learning_rate": 2.419508835808064e-07,
      "loss": 0.58836234,
      "num_input_tokens_seen": 152222380,
      "step": 7040,
      "time_per_iteration": 3.1498489379882812
    },
    {
      "auxiliary_loss_clip": 0.01268759,
      "auxiliary_loss_mlp": 0.01023466,
      "balance_loss_clip": 1.04596996,
      "balance_loss_mlp": 1.01601875,
      "epoch": 0.8466301929898394,
      "flos": 13735724561280.0,
      "grad_norm": 2.0203198500148503,
      "language_loss": 0.63451236,
      "learning_rate": 2.415796215659134e-07,
      "loss": 0.65743458,
      "num_input_tokens_seen": 152239085,
      "step": 7041,
      "time_per_iteration": 2.633406400680542
    },
    {
      "auxiliary_loss_clip": 0.01326152,
      "auxiliary_loss_mlp": 0.01022035,
      "balance_loss_clip": 1.04126918,
      "balance_loss_mlp": 1.01528454,
      "epoch": 0.8467504358804786,
      "flos": 19241260738560.0,
      "grad_norm": 2.0747288407989455,
      "language_loss": 0.77343285,
      "learning_rate": 2.412086263019939e-07,
      "loss": 0.79691476,
      "num_input_tokens_seen": 152257110,
      "step": 7042,
      "time_per_iteration": 2.705965995788574
    },
    {
      "auxiliary_loss_clip": 0.0117012,
      "auxiliary_loss_mlp": 0.01020981,
      "balance_loss_clip": 1.04966211,
      "balance_loss_mlp": 1.01446342,
      "epoch": 0.8468706787711177,
      "flos": 21324115710720.0,
      "grad_norm": 1.8731607749859291,
      "language_loss": 0.80041558,
      "learning_rate": 2.408378978453276e-07,
      "loss": 0.8223266,
      "num_input_tokens_seen": 152277230,
      "step": 7043,
      "time_per_iteration": 2.596576452255249
    },
    {
      "auxiliary_loss_clip": 0.01115725,
      "auxiliary_loss_mlp": 0.01000555,
      "balance_loss_clip": 1.00675607,
      "balance_loss_mlp": 0.99973232,
      "epoch": 0.8469909216617567,
      "flos": 64877439058560.0,
      "grad_norm": 0.8339788756901579,
      "language_loss": 0.63944185,
      "learning_rate": 2.404674362521533e-07,
      "loss": 0.6606046,
      "num_input_tokens_seen": 152335725,
      "step": 7044,
      "time_per_iteration": 3.0944736003875732
    },
    {
      "auxiliary_loss_clip": 0.01218391,
      "auxiliary_loss_mlp": 0.01024254,
      "balance_loss_clip": 1.04819989,
      "balance_loss_mlp": 1.01761067,
      "epoch": 0.8471111645523959,
      "flos": 19280583152640.0,
      "grad_norm": 2.367279043849664,
      "language_loss": 0.74462247,
      "learning_rate": 2.4009724157866997e-07,
      "loss": 0.76704895,
      "num_input_tokens_seen": 152352785,
      "step": 7045,
      "time_per_iteration": 2.592853546142578
    },
    {
      "auxiliary_loss_clip": 0.01171457,
      "auxiliary_loss_mlp": 0.01024122,
      "balance_loss_clip": 1.04945791,
      "balance_loss_mlp": 1.0179317,
      "epoch": 0.8472314074430349,
      "flos": 22015826893440.0,
      "grad_norm": 2.2482424412831774,
      "language_loss": 0.76720268,
      "learning_rate": 2.3972731388103564e-07,
      "loss": 0.78915846,
      "num_input_tokens_seen": 152371265,
      "step": 7046,
      "time_per_iteration": 2.655508518218994
    },
    {
      "auxiliary_loss_clip": 0.0131118,
      "auxiliary_loss_mlp": 0.0099883,
      "balance_loss_clip": 1.00640678,
      "balance_loss_mlp": 0.99801332,
      "epoch": 0.847351650333674,
      "flos": 57882580243200.0,
      "grad_norm": 0.7970537624742845,
      "language_loss": 0.62386847,
      "learning_rate": 2.393576532153687e-07,
      "loss": 0.64696854,
      "num_input_tokens_seen": 152435050,
      "step": 7047,
      "time_per_iteration": 4.450084209442139
    },
    {
      "auxiliary_loss_clip": 0.01112105,
      "auxiliary_loss_mlp": 0.01000426,
      "balance_loss_clip": 1.00739622,
      "balance_loss_mlp": 0.99962157,
      "epoch": 0.8474718932243132,
      "flos": 41284238313600.0,
      "grad_norm": 0.9288163913910275,
      "language_loss": 0.57793653,
      "learning_rate": 2.389882596377453e-07,
      "loss": 0.59906185,
      "num_input_tokens_seen": 152489315,
      "step": 7048,
      "time_per_iteration": 3.3292903900146484
    },
    {
      "auxiliary_loss_clip": 0.01172059,
      "auxiliary_loss_mlp": 0.01026188,
      "balance_loss_clip": 1.04694867,
      "balance_loss_mlp": 1.01919389,
      "epoch": 0.8475921361149522,
      "flos": 38180906974080.0,
      "grad_norm": 1.8566012143090733,
      "language_loss": 0.76266086,
      "learning_rate": 2.386191332042031e-07,
      "loss": 0.78464329,
      "num_input_tokens_seen": 152511210,
      "step": 7049,
      "time_per_iteration": 2.7319445610046387
    },
    {
      "auxiliary_loss_clip": 0.01176753,
      "auxiliary_loss_mlp": 0.01027021,
      "balance_loss_clip": 1.05053067,
      "balance_loss_mlp": 1.01968646,
      "epoch": 0.8477123790055913,
      "flos": 25375054723200.0,
      "grad_norm": 2.161636181502183,
      "language_loss": 0.73024702,
      "learning_rate": 2.3825027397073794e-07,
      "loss": 0.75228471,
      "num_input_tokens_seen": 152531685,
      "step": 7050,
      "time_per_iteration": 2.5889928340911865
    },
    {
      "auxiliary_loss_clip": 0.01220271,
      "auxiliary_loss_mlp": 0.01026916,
      "balance_loss_clip": 1.04938316,
      "balance_loss_mlp": 1.01969826,
      "epoch": 0.8478326218962304,
      "flos": 30225185389440.0,
      "grad_norm": 2.6826475722981287,
      "language_loss": 0.66757149,
      "learning_rate": 2.3788168199330515e-07,
      "loss": 0.69004339,
      "num_input_tokens_seen": 152553245,
      "step": 7051,
      "time_per_iteration": 2.6991865634918213
    },
    {
      "auxiliary_loss_clip": 0.01261552,
      "auxiliary_loss_mlp": 0.0102642,
      "balance_loss_clip": 1.04032493,
      "balance_loss_mlp": 1.01946449,
      "epoch": 0.8479528647868695,
      "flos": 38213800853760.0,
      "grad_norm": 1.7272147691371265,
      "language_loss": 0.72620732,
      "learning_rate": 2.3751335732782074e-07,
      "loss": 0.74908698,
      "num_input_tokens_seen": 152574505,
      "step": 7052,
      "time_per_iteration": 3.7221755981445312
    },
    {
      "auxiliary_loss_clip": 0.01222864,
      "auxiliary_loss_mlp": 0.01023189,
      "balance_loss_clip": 1.0497905,
      "balance_loss_mlp": 1.01628137,
      "epoch": 0.8480731076775085,
      "flos": 20957790856320.0,
      "grad_norm": 2.12829469342397,
      "language_loss": 0.79492849,
      "learning_rate": 2.371453000301582e-07,
      "loss": 0.81738901,
      "num_input_tokens_seen": 152593190,
      "step": 7053,
      "time_per_iteration": 2.6205191612243652
    },
    {
      "auxiliary_loss_clip": 0.01316296,
      "auxiliary_loss_mlp": 0.01025798,
      "balance_loss_clip": 1.04492259,
      "balance_loss_mlp": 1.0188539,
      "epoch": 0.8481933505681477,
      "flos": 32596510487040.0,
      "grad_norm": 1.8050444995430732,
      "language_loss": 0.74613315,
      "learning_rate": 2.3677751015615222e-07,
      "loss": 0.76955402,
      "num_input_tokens_seen": 152615265,
      "step": 7054,
      "time_per_iteration": 2.8097801208496094
    },
    {
      "auxiliary_loss_clip": 0.01269304,
      "auxiliary_loss_mlp": 0.01026359,
      "balance_loss_clip": 1.04421675,
      "balance_loss_mlp": 1.0193404,
      "epoch": 0.8483135934587868,
      "flos": 20741177888640.0,
      "grad_norm": 2.1291257512002546,
      "language_loss": 0.85623431,
      "learning_rate": 2.3640998776159593e-07,
      "loss": 0.87919092,
      "num_input_tokens_seen": 152632770,
      "step": 7055,
      "time_per_iteration": 3.8510067462921143
    },
    {
      "auxiliary_loss_clip": 0.01273455,
      "auxiliary_loss_mlp": 0.01026607,
      "balance_loss_clip": 1.04871285,
      "balance_loss_mlp": 1.02015793,
      "epoch": 0.8484338363494258,
      "flos": 21653057485440.0,
      "grad_norm": 1.8675466290461138,
      "language_loss": 0.81291497,
      "learning_rate": 2.3604273290224253e-07,
      "loss": 0.83591557,
      "num_input_tokens_seen": 152653485,
      "step": 7056,
      "time_per_iteration": 2.6814444065093994
    },
    {
      "auxiliary_loss_clip": 0.01272608,
      "auxiliary_loss_mlp": 0.01029617,
      "balance_loss_clip": 1.04846072,
      "balance_loss_mlp": 1.02255416,
      "epoch": 0.848554079240065,
      "flos": 15013964926080.0,
      "grad_norm": 2.2880913266947966,
      "language_loss": 0.7457875,
      "learning_rate": 2.356757456338039e-07,
      "loss": 0.7688098,
      "num_input_tokens_seen": 152670970,
      "step": 7057,
      "time_per_iteration": 2.670318365097046
    },
    {
      "auxiliary_loss_clip": 0.01165112,
      "auxiliary_loss_mlp": 0.01007557,
      "balance_loss_clip": 1.00929511,
      "balance_loss_mlp": 1.00669324,
      "epoch": 0.848674322130704,
      "flos": 68060453742720.0,
      "grad_norm": 0.7667875660343402,
      "language_loss": 0.58988398,
      "learning_rate": 2.3530902601195147e-07,
      "loss": 0.61161065,
      "num_input_tokens_seen": 152739460,
      "step": 7058,
      "time_per_iteration": 3.360110282897949
    },
    {
      "auxiliary_loss_clip": 0.01221975,
      "auxiliary_loss_mlp": 0.01028292,
      "balance_loss_clip": 1.04710054,
      "balance_loss_mlp": 1.02098179,
      "epoch": 0.8487945650213431,
      "flos": 18475788977280.0,
      "grad_norm": 2.1863254123718496,
      "language_loss": 0.7913053,
      "learning_rate": 2.34942574092317e-07,
      "loss": 0.81380802,
      "num_input_tokens_seen": 152754710,
      "step": 7059,
      "time_per_iteration": 2.6469030380249023
    },
    {
      "auxiliary_loss_clip": 0.01227078,
      "auxiliary_loss_mlp": 0.01028633,
      "balance_loss_clip": 1.04926658,
      "balance_loss_mlp": 1.02211785,
      "epoch": 0.8489148079119821,
      "flos": 23473189405440.0,
      "grad_norm": 1.9431977892643022,
      "language_loss": 0.76787102,
      "learning_rate": 2.3457638993049045e-07,
      "loss": 0.79042816,
      "num_input_tokens_seen": 152772700,
      "step": 7060,
      "time_per_iteration": 2.6472175121307373
    },
    {
      "auxiliary_loss_clip": 0.01417451,
      "auxiliary_loss_mlp": 0.0102656,
      "balance_loss_clip": 1.04466748,
      "balance_loss_mlp": 1.01840878,
      "epoch": 0.8490350508026213,
      "flos": 19937604775680.0,
      "grad_norm": 2.906907614936066,
      "language_loss": 0.63944197,
      "learning_rate": 2.3421047358202252e-07,
      "loss": 0.66388214,
      "num_input_tokens_seen": 152791550,
      "step": 7061,
      "time_per_iteration": 2.8119475841522217
    },
    {
      "auxiliary_loss_clip": 0.01225113,
      "auxiliary_loss_mlp": 0.01024499,
      "balance_loss_clip": 1.04865813,
      "balance_loss_mlp": 1.01719451,
      "epoch": 0.8491552936932604,
      "flos": 24279958828800.0,
      "grad_norm": 3.2270693414447145,
      "language_loss": 0.83541703,
      "learning_rate": 2.3384482510242144e-07,
      "loss": 0.8579132,
      "num_input_tokens_seen": 152809410,
      "step": 7062,
      "time_per_iteration": 2.634420394897461
    },
    {
      "auxiliary_loss_clip": 0.01176026,
      "auxiliary_loss_mlp": 0.01021986,
      "balance_loss_clip": 1.04853547,
      "balance_loss_mlp": 1.01537621,
      "epoch": 0.8492755365838994,
      "flos": 22522526098560.0,
      "grad_norm": 2.930366293754969,
      "language_loss": 0.77364743,
      "learning_rate": 2.3347944454715575e-07,
      "loss": 0.79562759,
      "num_input_tokens_seen": 152825800,
      "step": 7063,
      "time_per_iteration": 2.6298670768737793
    },
    {
      "auxiliary_loss_clip": 0.01178344,
      "auxiliary_loss_mlp": 0.01025021,
      "balance_loss_clip": 1.0513469,
      "balance_loss_mlp": 1.01742744,
      "epoch": 0.8493957794745386,
      "flos": 26980441182720.0,
      "grad_norm": 2.1532929876027302,
      "language_loss": 0.67453206,
      "learning_rate": 2.331143319716542e-07,
      "loss": 0.69656569,
      "num_input_tokens_seen": 152845330,
      "step": 7064,
      "time_per_iteration": 2.6348681449890137
    },
    {
      "auxiliary_loss_clip": 0.01227566,
      "auxiliary_loss_mlp": 0.01025154,
      "balance_loss_clip": 1.04706931,
      "balance_loss_mlp": 1.01811433,
      "epoch": 0.8495160223651776,
      "flos": 29861985018240.0,
      "grad_norm": 1.8834869232316611,
      "language_loss": 0.6584909,
      "learning_rate": 2.3274948743130363e-07,
      "loss": 0.68101811,
      "num_input_tokens_seen": 152865165,
      "step": 7065,
      "time_per_iteration": 2.765509605407715
    },
    {
      "auxiliary_loss_clip": 0.01173588,
      "auxiliary_loss_mlp": 0.01026771,
      "balance_loss_clip": 1.04769576,
      "balance_loss_mlp": 1.01982069,
      "epoch": 0.8496362652558167,
      "flos": 23075443128960.0,
      "grad_norm": 1.774954211679828,
      "language_loss": 0.79605013,
      "learning_rate": 2.3238491098145085e-07,
      "loss": 0.81805372,
      "num_input_tokens_seen": 152884695,
      "step": 7066,
      "time_per_iteration": 2.665830612182617
    },
    {
      "auxiliary_loss_clip": 0.01222708,
      "auxiliary_loss_mlp": 0.01031656,
      "balance_loss_clip": 1.04812944,
      "balance_loss_mlp": 1.02402091,
      "epoch": 0.8497565081464559,
      "flos": 14609107756800.0,
      "grad_norm": 2.142410292985417,
      "language_loss": 0.73051018,
      "learning_rate": 2.3202060267740141e-07,
      "loss": 0.75305378,
      "num_input_tokens_seen": 152902220,
      "step": 7067,
      "time_per_iteration": 2.5768885612487793
    },
    {
      "auxiliary_loss_clip": 0.01367016,
      "auxiliary_loss_mlp": 0.01020603,
      "balance_loss_clip": 1.03896475,
      "balance_loss_mlp": 1.01373005,
      "epoch": 0.8498767510370949,
      "flos": 21136446126720.0,
      "grad_norm": 2.5047945058907577,
      "language_loss": 0.76697576,
      "learning_rate": 2.3165656257442044e-07,
      "loss": 0.79085189,
      "num_input_tokens_seen": 152920740,
      "step": 7068,
      "time_per_iteration": 2.765084743499756
    },
    {
      "auxiliary_loss_clip": 0.01220013,
      "auxiliary_loss_mlp": 0.01027177,
      "balance_loss_clip": 1.04862571,
      "balance_loss_mlp": 1.02090406,
      "epoch": 0.849996993927734,
      "flos": 23654538195840.0,
      "grad_norm": 2.1646985244528896,
      "language_loss": 0.90028191,
      "learning_rate": 2.31292790727734e-07,
      "loss": 0.92275381,
      "num_input_tokens_seen": 152938305,
      "step": 7069,
      "time_per_iteration": 2.682459831237793
    },
    {
      "auxiliary_loss_clip": 0.01169947,
      "auxiliary_loss_mlp": 0.01024775,
      "balance_loss_clip": 1.04758954,
      "balance_loss_mlp": 1.01810873,
      "epoch": 0.8501172368183731,
      "flos": 20558069331840.0,
      "grad_norm": 2.3764304718083324,
      "language_loss": 0.80227774,
      "learning_rate": 2.3092928719252392e-07,
      "loss": 0.82422495,
      "num_input_tokens_seen": 152956705,
      "step": 7070,
      "time_per_iteration": 2.6079750061035156
    },
    {
      "auxiliary_loss_clip": 0.01221534,
      "auxiliary_loss_mlp": 0.01024012,
      "balance_loss_clip": 1.04561257,
      "balance_loss_mlp": 1.01758969,
      "epoch": 0.8502374797090122,
      "flos": 22272624201600.0,
      "grad_norm": 2.130889831247209,
      "language_loss": 0.78026599,
      "learning_rate": 2.3056605202393475e-07,
      "loss": 0.80272144,
      "num_input_tokens_seen": 152974265,
      "step": 7071,
      "time_per_iteration": 2.647853136062622
    },
    {
      "auxiliary_loss_clip": 0.01217427,
      "auxiliary_loss_mlp": 0.02568741,
      "balance_loss_clip": 1.04304266,
      "balance_loss_mlp": 0.99988377,
      "epoch": 0.8503577225996513,
      "flos": 23659817495040.0,
      "grad_norm": 2.1271434964637654,
      "language_loss": 0.67112231,
      "learning_rate": 2.3020308527706888e-07,
      "loss": 0.70898396,
      "num_input_tokens_seen": 152993680,
      "step": 7072,
      "time_per_iteration": 2.8079164028167725
    },
    {
      "auxiliary_loss_clip": 0.01276332,
      "auxiliary_loss_mlp": 0.01020551,
      "balance_loss_clip": 1.04366601,
      "balance_loss_mlp": 1.01404214,
      "epoch": 0.8504779654902904,
      "flos": 26758513002240.0,
      "grad_norm": 1.7842819830960281,
      "language_loss": 0.89419377,
      "learning_rate": 2.2984038700698715e-07,
      "loss": 0.91716266,
      "num_input_tokens_seen": 153012990,
      "step": 7073,
      "time_per_iteration": 4.5713207721710205
    },
    {
      "auxiliary_loss_clip": 0.01220351,
      "auxiliary_loss_mlp": 0.01027733,
      "balance_loss_clip": 1.04939842,
      "balance_loss_mlp": 1.02101552,
      "epoch": 0.8505982083809295,
      "flos": 26468247196800.0,
      "grad_norm": 4.556003432840408,
      "language_loss": 0.78823805,
      "learning_rate": 2.2947795726871222e-07,
      "loss": 0.81071889,
      "num_input_tokens_seen": 153034015,
      "step": 7074,
      "time_per_iteration": 2.73752760887146
    },
    {
      "auxiliary_loss_clip": 0.012284,
      "auxiliary_loss_mlp": 0.02566997,
      "balance_loss_clip": 1.05421436,
      "balance_loss_mlp": 0.99989885,
      "epoch": 0.8507184512715685,
      "flos": 20303390926080.0,
      "grad_norm": 2.07628640277676,
      "language_loss": 0.85918778,
      "learning_rate": 2.2911579611722253e-07,
      "loss": 0.89714175,
      "num_input_tokens_seen": 153053160,
      "step": 7075,
      "time_per_iteration": 2.776139736175537
    },
    {
      "auxiliary_loss_clip": 0.01169582,
      "auxiliary_loss_mlp": 0.01027076,
      "balance_loss_clip": 1.04483366,
      "balance_loss_mlp": 1.02040327,
      "epoch": 0.8508386941622077,
      "flos": 19025186474880.0,
      "grad_norm": 2.5221136370891144,
      "language_loss": 0.87282604,
      "learning_rate": 2.2875390360745905e-07,
      "loss": 0.89479268,
      "num_input_tokens_seen": 153072565,
      "step": 7076,
      "time_per_iteration": 2.654996156692505
    },
    {
      "auxiliary_loss_clip": 0.01329641,
      "auxiliary_loss_mlp": 0.01023537,
      "balance_loss_clip": 1.04597449,
      "balance_loss_mlp": 1.01598227,
      "epoch": 0.8509589370528468,
      "flos": 16433405654400.0,
      "grad_norm": 1.5735106493761657,
      "language_loss": 0.77684915,
      "learning_rate": 2.2839227979432008e-07,
      "loss": 0.80038089,
      "num_input_tokens_seen": 153090215,
      "step": 7077,
      "time_per_iteration": 2.7534525394439697
    },
    {
      "auxiliary_loss_clip": 0.01271128,
      "auxiliary_loss_mlp": 0.01025246,
      "balance_loss_clip": 1.0446465,
      "balance_loss_mlp": 1.01807249,
      "epoch": 0.8510791799434858,
      "flos": 18259714713600.0,
      "grad_norm": 1.8835502844344914,
      "language_loss": 0.85236758,
      "learning_rate": 2.2803092473266373e-07,
      "loss": 0.87533128,
      "num_input_tokens_seen": 153107740,
      "step": 7078,
      "time_per_iteration": 3.625230312347412
    },
    {
      "auxiliary_loss_clip": 0.01176277,
      "auxiliary_loss_mlp": 0.01033802,
      "balance_loss_clip": 1.05009484,
      "balance_loss_mlp": 1.02709377,
      "epoch": 0.851199422834125,
      "flos": 23441372933760.0,
      "grad_norm": 3.448770308203283,
      "language_loss": 0.8675977,
      "learning_rate": 2.2766983847730724e-07,
      "loss": 0.88969851,
      "num_input_tokens_seen": 153127410,
      "step": 7079,
      "time_per_iteration": 2.7156238555908203
    },
    {
      "auxiliary_loss_clip": 0.01330869,
      "auxiliary_loss_mlp": 0.01028409,
      "balance_loss_clip": 1.0445981,
      "balance_loss_mlp": 1.02095222,
      "epoch": 0.851319665724764,
      "flos": 16289404030080.0,
      "grad_norm": 2.0787063728805646,
      "language_loss": 0.66863507,
      "learning_rate": 2.2730902108302663e-07,
      "loss": 0.69222784,
      "num_input_tokens_seen": 153144325,
      "step": 7080,
      "time_per_iteration": 3.633113384246826
    },
    {
      "auxiliary_loss_clip": 0.01270224,
      "auxiliary_loss_mlp": 0.01026835,
      "balance_loss_clip": 1.04260278,
      "balance_loss_mlp": 1.01966178,
      "epoch": 0.8514399086154031,
      "flos": 18989347680000.0,
      "grad_norm": 1.7257088339299664,
      "language_loss": 0.68729913,
      "learning_rate": 2.269484726045583e-07,
      "loss": 0.71026981,
      "num_input_tokens_seen": 153163240,
      "step": 7081,
      "time_per_iteration": 2.6334805488586426
    },
    {
      "auxiliary_loss_clip": 0.01325988,
      "auxiliary_loss_mlp": 0.01024911,
      "balance_loss_clip": 1.04660213,
      "balance_loss_mlp": 1.01849198,
      "epoch": 0.8515601515060423,
      "flos": 24571194301440.0,
      "grad_norm": 2.2319178782734954,
      "language_loss": 0.79141092,
      "learning_rate": 2.2658819309659672e-07,
      "loss": 0.81491989,
      "num_input_tokens_seen": 153183440,
      "step": 7082,
      "time_per_iteration": 2.9081461429595947
    },
    {
      "auxiliary_loss_clip": 0.01269729,
      "auxiliary_loss_mlp": 0.01025344,
      "balance_loss_clip": 1.0485419,
      "balance_loss_mlp": 1.01893926,
      "epoch": 0.8516803943966813,
      "flos": 19529443555200.0,
      "grad_norm": 2.437377520858372,
      "language_loss": 0.84981281,
      "learning_rate": 2.2622818261379706e-07,
      "loss": 0.87276357,
      "num_input_tokens_seen": 153200460,
      "step": 7083,
      "time_per_iteration": 2.757622003555298
    },
    {
      "auxiliary_loss_clip": 0.01273975,
      "auxiliary_loss_mlp": 0.01026798,
      "balance_loss_clip": 1.04497063,
      "balance_loss_mlp": 1.01950216,
      "epoch": 0.8518006372873204,
      "flos": 20265792364800.0,
      "grad_norm": 1.844362731171038,
      "language_loss": 0.7478922,
      "learning_rate": 2.2586844121077142e-07,
      "loss": 0.77089989,
      "num_input_tokens_seen": 153218970,
      "step": 7084,
      "time_per_iteration": 2.666950225830078
    },
    {
      "auxiliary_loss_clip": 0.01276354,
      "auxiliary_loss_mlp": 0.0102919,
      "balance_loss_clip": 1.04291058,
      "balance_loss_mlp": 1.02214789,
      "epoch": 0.8519208801779595,
      "flos": 24133227770880.0,
      "grad_norm": 1.956154870066799,
      "language_loss": 0.71998638,
      "learning_rate": 2.2550896894209215e-07,
      "loss": 0.74304187,
      "num_input_tokens_seen": 153238485,
      "step": 7085,
      "time_per_iteration": 2.8709681034088135
    },
    {
      "auxiliary_loss_clip": 0.01324199,
      "auxiliary_loss_mlp": 0.00998193,
      "balance_loss_clip": 1.00737154,
      "balance_loss_mlp": 0.99731666,
      "epoch": 0.8520411230685986,
      "flos": 63035223252480.0,
      "grad_norm": 0.678144030794841,
      "language_loss": 0.56640661,
      "learning_rate": 2.2514976586229184e-07,
      "loss": 0.5896306,
      "num_input_tokens_seen": 153306430,
      "step": 7086,
      "time_per_iteration": 3.408799409866333
    },
    {
      "auxiliary_loss_clip": 0.01115479,
      "auxiliary_loss_mlp": 0.00999131,
      "balance_loss_clip": 1.00790191,
      "balance_loss_mlp": 0.99827832,
      "epoch": 0.8521613659592376,
      "flos": 65836865283840.0,
      "grad_norm": 0.7537808304098034,
      "language_loss": 0.54601419,
      "learning_rate": 2.247908320258609e-07,
      "loss": 0.56716025,
      "num_input_tokens_seen": 153366520,
      "step": 7087,
      "time_per_iteration": 3.1665263175964355
    },
    {
      "auxiliary_loss_clip": 0.01360968,
      "auxiliary_loss_mlp": 0.01025687,
      "balance_loss_clip": 1.04422212,
      "balance_loss_mlp": 1.01775408,
      "epoch": 0.8522816088498768,
      "flos": 23112323418240.0,
      "grad_norm": 2.251356618527483,
      "language_loss": 0.80019855,
      "learning_rate": 2.2443216748724914e-07,
      "loss": 0.82406515,
      "num_input_tokens_seen": 153387230,
      "step": 7088,
      "time_per_iteration": 2.7077419757843018
    },
    {
      "auxiliary_loss_clip": 0.01226542,
      "auxiliary_loss_mlp": 0.02565769,
      "balance_loss_clip": 1.0500865,
      "balance_loss_mlp": 0.99988735,
      "epoch": 0.8524018517405159,
      "flos": 31758140073600.0,
      "grad_norm": 1.8211191575125913,
      "language_loss": 0.74291831,
      "learning_rate": 2.2407377230086588e-07,
      "loss": 0.78084141,
      "num_input_tokens_seen": 153409585,
      "step": 7089,
      "time_per_iteration": 2.7235043048858643
    },
    {
      "auxiliary_loss_clip": 0.01325378,
      "auxiliary_loss_mlp": 0.01028383,
      "balance_loss_clip": 1.04947329,
      "balance_loss_mlp": 1.02129626,
      "epoch": 0.8525220946311549,
      "flos": 18690318956160.0,
      "grad_norm": 2.110068038332757,
      "language_loss": 0.83272111,
      "learning_rate": 2.23715646521079e-07,
      "loss": 0.85625875,
      "num_input_tokens_seen": 153427105,
      "step": 7090,
      "time_per_iteration": 2.720362424850464
    },
    {
      "auxiliary_loss_clip": 0.01227497,
      "auxiliary_loss_mlp": 0.02570626,
      "balance_loss_clip": 1.04836702,
      "balance_loss_mlp": 0.999897,
      "epoch": 0.852642337521794,
      "flos": 21793216354560.0,
      "grad_norm": 1.7384142871118957,
      "language_loss": 0.84074992,
      "learning_rate": 2.2335779020221724e-07,
      "loss": 0.87873113,
      "num_input_tokens_seen": 153443725,
      "step": 7091,
      "time_per_iteration": 2.692399501800537
    },
    {
      "auxiliary_loss_clip": 0.01128022,
      "auxiliary_loss_mlp": 0.00999899,
      "balance_loss_clip": 1.02070093,
      "balance_loss_mlp": 0.99896282,
      "epoch": 0.8527625804124331,
      "flos": 69040132260480.0,
      "grad_norm": 0.8006565790539628,
      "language_loss": 0.56422448,
      "learning_rate": 2.2300020339856497e-07,
      "loss": 0.5855037,
      "num_input_tokens_seen": 153506410,
      "step": 7092,
      "time_per_iteration": 3.2376606464385986
    },
    {
      "auxiliary_loss_clip": 0.01267247,
      "auxiliary_loss_mlp": 0.01031076,
      "balance_loss_clip": 1.04560483,
      "balance_loss_mlp": 1.02430773,
      "epoch": 0.8528828233030722,
      "flos": 26979399688320.0,
      "grad_norm": 2.4590777249369786,
      "language_loss": 0.77742898,
      "learning_rate": 2.2264288616436966e-07,
      "loss": 0.80041218,
      "num_input_tokens_seen": 153526665,
      "step": 7093,
      "time_per_iteration": 2.6055119037628174
    },
    {
      "auxiliary_loss_clip": 0.01265987,
      "auxiliary_loss_mlp": 0.0102582,
      "balance_loss_clip": 1.04556632,
      "balance_loss_mlp": 1.01880157,
      "epoch": 0.8530030661937112,
      "flos": 17487598936320.0,
      "grad_norm": 1.9896289482356657,
      "language_loss": 0.72248614,
      "learning_rate": 2.222858385538351e-07,
      "loss": 0.74540424,
      "num_input_tokens_seen": 153543465,
      "step": 7094,
      "time_per_iteration": 2.579873561859131
    },
    {
      "auxiliary_loss_clip": 0.0122246,
      "auxiliary_loss_mlp": 0.01021778,
      "balance_loss_clip": 1.04744673,
      "balance_loss_mlp": 1.0152868,
      "epoch": 0.8531233090843504,
      "flos": 22160798184960.0,
      "grad_norm": 1.8164378744164682,
      "language_loss": 0.68116409,
      "learning_rate": 2.2192906062112527e-07,
      "loss": 0.70360649,
      "num_input_tokens_seen": 153563340,
      "step": 7095,
      "time_per_iteration": 2.541853904724121
    },
    {
      "auxiliary_loss_clip": 0.01171234,
      "auxiliary_loss_mlp": 0.01023104,
      "balance_loss_clip": 1.04728663,
      "balance_loss_mlp": 1.01616335,
      "epoch": 0.8532435519749895,
      "flos": 37635388145280.0,
      "grad_norm": 2.2832168618450015,
      "language_loss": 0.70663983,
      "learning_rate": 2.2157255242036377e-07,
      "loss": 0.72858322,
      "num_input_tokens_seen": 153587005,
      "step": 7096,
      "time_per_iteration": 2.6452081203460693
    },
    {
      "auxiliary_loss_clip": 0.01314037,
      "auxiliary_loss_mlp": 0.01029064,
      "balance_loss_clip": 1.04405653,
      "balance_loss_mlp": 1.02219772,
      "epoch": 0.8533637948656285,
      "flos": 21398163598080.0,
      "grad_norm": 1.62029700284662,
      "language_loss": 0.74289089,
      "learning_rate": 2.2121631400563135e-07,
      "loss": 0.7663219,
      "num_input_tokens_seen": 153606835,
      "step": 7097,
      "time_per_iteration": 2.6063220500946045
    },
    {
      "auxiliary_loss_clip": 0.01111299,
      "auxiliary_loss_mlp": 0.01008965,
      "balance_loss_clip": 1.00839186,
      "balance_loss_mlp": 1.00808299,
      "epoch": 0.8534840377562677,
      "flos": 53345122490880.0,
      "grad_norm": 0.7592032689062704,
      "language_loss": 0.52954614,
      "learning_rate": 2.208603454309701e-07,
      "loss": 0.55074877,
      "num_input_tokens_seen": 153664925,
      "step": 7098,
      "time_per_iteration": 3.1698896884918213
    },
    {
      "auxiliary_loss_clip": 0.01366638,
      "auxiliary_loss_mlp": 0.01032939,
      "balance_loss_clip": 1.04185593,
      "balance_loss_mlp": 1.02522647,
      "epoch": 0.8536042806469067,
      "flos": 20814148368000.0,
      "grad_norm": 2.1585341802272033,
      "language_loss": 0.70945251,
      "learning_rate": 2.2050464675037994e-07,
      "loss": 0.73344827,
      "num_input_tokens_seen": 153683550,
      "step": 7099,
      "time_per_iteration": 3.6095168590545654
    },
    {
      "auxiliary_loss_clip": 0.01267983,
      "auxiliary_loss_mlp": 0.01027794,
      "balance_loss_clip": 1.04589295,
      "balance_loss_mlp": 1.02064109,
      "epoch": 0.8537245235375458,
      "flos": 24681368292480.0,
      "grad_norm": 7.181630395265267,
      "language_loss": 0.73796093,
      "learning_rate": 2.2014921801782016e-07,
      "loss": 0.76091874,
      "num_input_tokens_seen": 153703040,
      "step": 7100,
      "time_per_iteration": 3.6590042114257812
    },
    {
      "auxiliary_loss_clip": 0.01271915,
      "auxiliary_loss_mlp": 0.01030073,
      "balance_loss_clip": 1.04237545,
      "balance_loss_mlp": 1.02336776,
      "epoch": 0.853844766428185,
      "flos": 24384817607040.0,
      "grad_norm": 2.629718294824743,
      "language_loss": 0.73770517,
      "learning_rate": 2.1979405928720872e-07,
      "loss": 0.76072514,
      "num_input_tokens_seen": 153722695,
      "step": 7101,
      "time_per_iteration": 2.7108728885650635
    },
    {
      "auxiliary_loss_clip": 0.01275008,
      "auxiliary_loss_mlp": 0.01025824,
      "balance_loss_clip": 1.04502988,
      "balance_loss_mlp": 1.01967001,
      "epoch": 0.853965009318824,
      "flos": 20955707867520.0,
      "grad_norm": 1.6096564855310138,
      "language_loss": 0.79291463,
      "learning_rate": 2.1943917061242257e-07,
      "loss": 0.81592292,
      "num_input_tokens_seen": 153742550,
      "step": 7102,
      "time_per_iteration": 2.7051103115081787
    },
    {
      "auxiliary_loss_clip": 0.01230972,
      "auxiliary_loss_mlp": 0.02570151,
      "balance_loss_clip": 1.05009687,
      "balance_loss_mlp": 0.99987674,
      "epoch": 0.8540852522094631,
      "flos": 24201816791040.0,
      "grad_norm": 1.6003671746342971,
      "language_loss": 0.66649556,
      "learning_rate": 2.1908455204729903e-07,
      "loss": 0.70450675,
      "num_input_tokens_seen": 153761700,
      "step": 7103,
      "time_per_iteration": 3.577935218811035
    },
    {
      "auxiliary_loss_clip": 0.01274819,
      "auxiliary_loss_mlp": 0.01024092,
      "balance_loss_clip": 1.04379809,
      "balance_loss_mlp": 1.01718986,
      "epoch": 0.8542054951001022,
      "flos": 25082921410560.0,
      "grad_norm": 2.5481643936437055,
      "language_loss": 0.7832607,
      "learning_rate": 2.1873020364563265e-07,
      "loss": 0.80624974,
      "num_input_tokens_seen": 153780765,
      "step": 7104,
      "time_per_iteration": 2.6990814208984375
    },
    {
      "auxiliary_loss_clip": 0.01218328,
      "auxiliary_loss_mlp": 0.01023212,
      "balance_loss_clip": 1.04697657,
      "balance_loss_mlp": 1.01599348,
      "epoch": 0.8543257379907413,
      "flos": 24316551809280.0,
      "grad_norm": 2.2696408756559534,
      "language_loss": 0.76105672,
      "learning_rate": 2.183761254611789e-07,
      "loss": 0.78347212,
      "num_input_tokens_seen": 153801090,
      "step": 7105,
      "time_per_iteration": 2.6332359313964844
    },
    {
      "auxiliary_loss_clip": 0.01222038,
      "auxiliary_loss_mlp": 0.0102005,
      "balance_loss_clip": 1.04953182,
      "balance_loss_mlp": 1.01351118,
      "epoch": 0.8544459808813804,
      "flos": 55286630467200.0,
      "grad_norm": 2.508116020139061,
      "language_loss": 0.70596993,
      "learning_rate": 2.1802231754764987e-07,
      "loss": 0.72839081,
      "num_input_tokens_seen": 153826530,
      "step": 7106,
      "time_per_iteration": 3.790107488632202
    },
    {
      "auxiliary_loss_clip": 0.01272113,
      "auxiliary_loss_mlp": 0.01027747,
      "balance_loss_clip": 1.04407167,
      "balance_loss_mlp": 1.02053523,
      "epoch": 0.8545662237720195,
      "flos": 25776248705280.0,
      "grad_norm": 1.9080476286306662,
      "language_loss": 0.76438308,
      "learning_rate": 2.17668779958718e-07,
      "loss": 0.78738171,
      "num_input_tokens_seen": 153849110,
      "step": 7107,
      "time_per_iteration": 2.7463247776031494
    },
    {
      "auxiliary_loss_clip": 0.01174486,
      "auxiliary_loss_mlp": 0.01028877,
      "balance_loss_clip": 1.05043483,
      "balance_loss_mlp": 1.02215421,
      "epoch": 0.8546864666626586,
      "flos": 11108320427520.0,
      "grad_norm": 2.498296814413374,
      "language_loss": 0.80578262,
      "learning_rate": 2.1731551274801553e-07,
      "loss": 0.82781625,
      "num_input_tokens_seen": 153865550,
      "step": 7108,
      "time_per_iteration": 2.535413980484009
    },
    {
      "auxiliary_loss_clip": 0.01279127,
      "auxiliary_loss_mlp": 0.01023892,
      "balance_loss_clip": 1.04876101,
      "balance_loss_mlp": 1.01667142,
      "epoch": 0.8548067095532976,
      "flos": 25520169669120.0,
      "grad_norm": 2.221834581778872,
      "language_loss": 0.61405659,
      "learning_rate": 2.169625159691324e-07,
      "loss": 0.63708681,
      "num_input_tokens_seen": 153885425,
      "step": 7109,
      "time_per_iteration": 2.7056283950805664
    },
    {
      "auxiliary_loss_clip": 0.01375373,
      "auxiliary_loss_mlp": 0.01024731,
      "balance_loss_clip": 1.04155838,
      "balance_loss_mlp": 1.01754856,
      "epoch": 0.8549269524439368,
      "flos": 24717853532160.0,
      "grad_norm": 2.9742204762396907,
      "language_loss": 0.73994905,
      "learning_rate": 2.1660978967561784e-07,
      "loss": 0.76395011,
      "num_input_tokens_seen": 153904760,
      "step": 7110,
      "time_per_iteration": 2.735150098800659
    },
    {
      "auxiliary_loss_clip": 0.01171144,
      "auxiliary_loss_mlp": 0.01021726,
      "balance_loss_clip": 1.0474844,
      "balance_loss_mlp": 1.01521134,
      "epoch": 0.8550471953345758,
      "flos": 19825599191040.0,
      "grad_norm": 7.315408951371546,
      "language_loss": 0.79462832,
      "learning_rate": 2.1625733392098035e-07,
      "loss": 0.81655705,
      "num_input_tokens_seen": 153920370,
      "step": 7111,
      "time_per_iteration": 2.5988078117370605
    },
    {
      "auxiliary_loss_clip": 0.01170734,
      "auxiliary_loss_mlp": 0.01024176,
      "balance_loss_clip": 1.04688728,
      "balance_loss_mlp": 1.01764631,
      "epoch": 0.8551674382252149,
      "flos": 22820441500800.0,
      "grad_norm": 1.8935783141159253,
      "language_loss": 0.79744446,
      "learning_rate": 2.159051487586867e-07,
      "loss": 0.81939352,
      "num_input_tokens_seen": 153940500,
      "step": 7112,
      "time_per_iteration": 2.611186981201172
    },
    {
      "auxiliary_loss_clip": 0.01278157,
      "auxiliary_loss_mlp": 0.01029216,
      "balance_loss_clip": 1.05016279,
      "balance_loss_mlp": 1.02127099,
      "epoch": 0.8552876811158541,
      "flos": 20631255292800.0,
      "grad_norm": 2.740933469933974,
      "language_loss": 0.72479439,
      "learning_rate": 2.155532342421642e-07,
      "loss": 0.74786806,
      "num_input_tokens_seen": 153958500,
      "step": 7113,
      "time_per_iteration": 2.7635550498962402
    },
    {
      "auxiliary_loss_clip": 0.01226263,
      "auxiliary_loss_mlp": 0.01025863,
      "balance_loss_clip": 1.0483501,
      "balance_loss_mlp": 1.01814711,
      "epoch": 0.8554079240064931,
      "flos": 23112359331840.0,
      "grad_norm": 1.938509382520293,
      "language_loss": 0.78413165,
      "learning_rate": 2.1520159042479636e-07,
      "loss": 0.8066529,
      "num_input_tokens_seen": 153976790,
      "step": 7114,
      "time_per_iteration": 2.6376612186431885
    },
    {
      "auxiliary_loss_clip": 0.01222437,
      "auxiliary_loss_mlp": 0.01028718,
      "balance_loss_clip": 1.05000591,
      "balance_loss_mlp": 1.02207804,
      "epoch": 0.8555281668971322,
      "flos": 22128047959680.0,
      "grad_norm": 1.966945365426684,
      "language_loss": 0.70918751,
      "learning_rate": 2.148502173599287e-07,
      "loss": 0.73169905,
      "num_input_tokens_seen": 153994930,
      "step": 7115,
      "time_per_iteration": 2.6434996128082275
    },
    {
      "auxiliary_loss_clip": 0.0126777,
      "auxiliary_loss_mlp": 0.01028296,
      "balance_loss_clip": 1.04520893,
      "balance_loss_mlp": 1.02128959,
      "epoch": 0.8556484097877713,
      "flos": 31139040234240.0,
      "grad_norm": 1.7609861839565348,
      "language_loss": 0.65673572,
      "learning_rate": 2.1449911510086372e-07,
      "loss": 0.67969632,
      "num_input_tokens_seen": 154014400,
      "step": 7116,
      "time_per_iteration": 2.719958543777466
    },
    {
      "auxiliary_loss_clip": 0.01218006,
      "auxiliary_loss_mlp": 0.01022923,
      "balance_loss_clip": 1.04645872,
      "balance_loss_mlp": 1.01607478,
      "epoch": 0.8557686526784104,
      "flos": 24316551809280.0,
      "grad_norm": 1.8806755534480286,
      "language_loss": 0.77009046,
      "learning_rate": 2.141482837008628e-07,
      "loss": 0.79249978,
      "num_input_tokens_seen": 154034940,
      "step": 7117,
      "time_per_iteration": 2.6958961486816406
    },
    {
      "auxiliary_loss_clip": 0.01221317,
      "auxiliary_loss_mlp": 0.01027873,
      "balance_loss_clip": 1.0472734,
      "balance_loss_mlp": 1.02117622,
      "epoch": 0.8558888955690495,
      "flos": 17712723427200.0,
      "grad_norm": 1.8851207139393977,
      "language_loss": 0.72285086,
      "learning_rate": 2.1379772321314826e-07,
      "loss": 0.74534279,
      "num_input_tokens_seen": 154052985,
      "step": 7118,
      "time_per_iteration": 2.6762945652008057
    },
    {
      "auxiliary_loss_clip": 0.01418733,
      "auxiliary_loss_mlp": 0.01033674,
      "balance_loss_clip": 1.04500961,
      "balance_loss_mlp": 1.02631938,
      "epoch": 0.8560091384596886,
      "flos": 19171702051200.0,
      "grad_norm": 2.3460211481387843,
      "language_loss": 0.81594896,
      "learning_rate": 2.1344743369089802e-07,
      "loss": 0.840473,
      "num_input_tokens_seen": 154068765,
      "step": 7119,
      "time_per_iteration": 2.768770694732666
    },
    {
      "auxiliary_loss_clip": 0.01280523,
      "auxiliary_loss_mlp": 0.01026575,
      "balance_loss_clip": 1.05113399,
      "balance_loss_mlp": 1.01962185,
      "epoch": 0.8561293813503277,
      "flos": 23914855036800.0,
      "grad_norm": 1.7342826281058894,
      "language_loss": 0.81992066,
      "learning_rate": 2.130974151872522e-07,
      "loss": 0.84299171,
      "num_input_tokens_seen": 154089100,
      "step": 7120,
      "time_per_iteration": 2.880168914794922
    },
    {
      "auxiliary_loss_clip": 0.01225705,
      "auxiliary_loss_mlp": 0.0102808,
      "balance_loss_clip": 1.04883504,
      "balance_loss_mlp": 1.02091503,
      "epoch": 0.8562496242409667,
      "flos": 22529206028160.0,
      "grad_norm": 1.7736337578173655,
      "language_loss": 0.78594774,
      "learning_rate": 2.1274766775530773e-07,
      "loss": 0.80848563,
      "num_input_tokens_seen": 154108965,
      "step": 7121,
      "time_per_iteration": 2.7085988521575928
    },
    {
      "auxiliary_loss_clip": 0.01175751,
      "auxiliary_loss_mlp": 0.01024362,
      "balance_loss_clip": 1.0492053,
      "balance_loss_mlp": 1.01754642,
      "epoch": 0.8563698671316058,
      "flos": 14712745472640.0,
      "grad_norm": 2.4428321711095213,
      "language_loss": 0.79554528,
      "learning_rate": 2.1239819144812077e-07,
      "loss": 0.81754643,
      "num_input_tokens_seen": 154123425,
      "step": 7122,
      "time_per_iteration": 2.5918619632720947
    },
    {
      "auxiliary_loss_clip": 0.0131568,
      "auxiliary_loss_mlp": 0.01024127,
      "balance_loss_clip": 1.04080033,
      "balance_loss_mlp": 1.01737726,
      "epoch": 0.856490110022245,
      "flos": 39167768211840.0,
      "grad_norm": 1.7257777286178515,
      "language_loss": 0.69934464,
      "learning_rate": 2.1204898631870716e-07,
      "loss": 0.72274268,
      "num_input_tokens_seen": 154148315,
      "step": 7123,
      "time_per_iteration": 2.8727822303771973
    },
    {
      "auxiliary_loss_clip": 0.01273926,
      "auxiliary_loss_mlp": 0.01024701,
      "balance_loss_clip": 1.0503068,
      "balance_loss_mlp": 1.01765883,
      "epoch": 0.856610352912884,
      "flos": 29059345658880.0,
      "grad_norm": 2.072365661642435,
      "language_loss": 0.75976539,
      "learning_rate": 2.1170005242004006e-07,
      "loss": 0.78275168,
      "num_input_tokens_seen": 154169665,
      "step": 7124,
      "time_per_iteration": 2.7887942790985107
    },
    {
      "auxiliary_loss_clip": 0.0117471,
      "auxiliary_loss_mlp": 0.01026717,
      "balance_loss_clip": 1.04445899,
      "balance_loss_mlp": 1.01990402,
      "epoch": 0.8567305958035231,
      "flos": 23878333883520.0,
      "grad_norm": 1.902117454151841,
      "language_loss": 0.78216249,
      "learning_rate": 2.1135138980505384e-07,
      "loss": 0.80417675,
      "num_input_tokens_seen": 154190335,
      "step": 7125,
      "time_per_iteration": 3.7117791175842285
    },
    {
      "auxiliary_loss_clip": 0.01265502,
      "auxiliary_loss_mlp": 0.0102715,
      "balance_loss_clip": 1.04596281,
      "balance_loss_mlp": 1.02027726,
      "epoch": 0.8568508386941622,
      "flos": 22200120599040.0,
      "grad_norm": 1.8403957912342845,
      "language_loss": 0.72230768,
      "learning_rate": 2.110029985266395e-07,
      "loss": 0.74523413,
      "num_input_tokens_seen": 154210040,
      "step": 7126,
      "time_per_iteration": 3.56243896484375
    },
    {
      "auxiliary_loss_clip": 0.01174328,
      "auxiliary_loss_mlp": 0.01021635,
      "balance_loss_clip": 1.04402626,
      "balance_loss_mlp": 1.01463199,
      "epoch": 0.8569710815848013,
      "flos": 17307507121920.0,
      "grad_norm": 1.8565904898130845,
      "language_loss": 0.73831636,
      "learning_rate": 2.1065487863764787e-07,
      "loss": 0.76027596,
      "num_input_tokens_seen": 154228385,
      "step": 7127,
      "time_per_iteration": 2.6811881065368652
    },
    {
      "auxiliary_loss_clip": 0.01368185,
      "auxiliary_loss_mlp": 0.01022317,
      "balance_loss_clip": 1.03834677,
      "balance_loss_mlp": 1.01578736,
      "epoch": 0.8570913244754403,
      "flos": 23732285184000.0,
      "grad_norm": 1.7387713271208687,
      "language_loss": 0.85568786,
      "learning_rate": 2.1030703019088846e-07,
      "loss": 0.8795929,
      "num_input_tokens_seen": 154249015,
      "step": 7128,
      "time_per_iteration": 2.739590883255005
    },
    {
      "auxiliary_loss_clip": 0.01216465,
      "auxiliary_loss_mlp": 0.01021239,
      "balance_loss_clip": 1.04762053,
      "balance_loss_mlp": 1.01435757,
      "epoch": 0.8572115673660795,
      "flos": 20048748433920.0,
      "grad_norm": 1.8682849118554685,
      "language_loss": 0.70723236,
      "learning_rate": 2.099594532391291e-07,
      "loss": 0.72960937,
      "num_input_tokens_seen": 154267700,
      "step": 7129,
      "time_per_iteration": 3.5113184452056885
    },
    {
      "auxiliary_loss_clip": 0.01218016,
      "auxiliary_loss_mlp": 0.01022476,
      "balance_loss_clip": 1.04601026,
      "balance_loss_mlp": 1.01580954,
      "epoch": 0.8573318102567186,
      "flos": 27160389342720.0,
      "grad_norm": 3.224048948361951,
      "language_loss": 0.78986323,
      "learning_rate": 2.0961214783509806e-07,
      "loss": 0.81226814,
      "num_input_tokens_seen": 154290580,
      "step": 7130,
      "time_per_iteration": 2.732994556427002
    },
    {
      "auxiliary_loss_clip": 0.01274344,
      "auxiliary_loss_mlp": 0.01022301,
      "balance_loss_clip": 1.04367852,
      "balance_loss_mlp": 1.01556253,
      "epoch": 0.8574520531473576,
      "flos": 24936585402240.0,
      "grad_norm": 1.986541758041221,
      "language_loss": 0.74865294,
      "learning_rate": 2.0926511403148051e-07,
      "loss": 0.77161944,
      "num_input_tokens_seen": 154309545,
      "step": 7131,
      "time_per_iteration": 2.688246011734009
    },
    {
      "auxiliary_loss_clip": 0.0122678,
      "auxiliary_loss_mlp": 0.01025389,
      "balance_loss_clip": 1.04618382,
      "balance_loss_mlp": 1.01877618,
      "epoch": 0.8575722960379968,
      "flos": 18771154513920.0,
      "grad_norm": 1.8506812563614037,
      "language_loss": 0.75713319,
      "learning_rate": 2.0891835188092143e-07,
      "loss": 0.77965486,
      "num_input_tokens_seen": 154326545,
      "step": 7132,
      "time_per_iteration": 3.5964648723602295
    },
    {
      "auxiliary_loss_clip": 0.01226118,
      "auxiliary_loss_mlp": 0.01028751,
      "balance_loss_clip": 1.04479671,
      "balance_loss_mlp": 1.02225733,
      "epoch": 0.8576925389286358,
      "flos": 22200300167040.0,
      "grad_norm": 2.0775148778607417,
      "language_loss": 0.81275278,
      "learning_rate": 2.0857186143602434e-07,
      "loss": 0.83530152,
      "num_input_tokens_seen": 154345190,
      "step": 7133,
      "time_per_iteration": 2.6437976360321045
    },
    {
      "auxiliary_loss_clip": 0.01316738,
      "auxiliary_loss_mlp": 0.01033798,
      "balance_loss_clip": 1.04208422,
      "balance_loss_mlp": 1.02656245,
      "epoch": 0.8578127818192749,
      "flos": 22894345733760.0,
      "grad_norm": 2.3167525996261604,
      "language_loss": 0.67753381,
      "learning_rate": 2.0822564274935094e-07,
      "loss": 0.7010392,
      "num_input_tokens_seen": 154364615,
      "step": 7134,
      "time_per_iteration": 2.727989673614502
    },
    {
      "auxiliary_loss_clip": 0.01278556,
      "auxiliary_loss_mlp": 0.01024127,
      "balance_loss_clip": 1.05098605,
      "balance_loss_mlp": 1.0169301,
      "epoch": 0.8579330247099141,
      "flos": 34824839541120.0,
      "grad_norm": 2.142203714824796,
      "language_loss": 0.66968918,
      "learning_rate": 2.078796958734239e-07,
      "loss": 0.692716,
      "num_input_tokens_seen": 154387335,
      "step": 7135,
      "time_per_iteration": 2.779693841934204
    },
    {
      "auxiliary_loss_clip": 0.01223395,
      "auxiliary_loss_mlp": 0.01026954,
      "balance_loss_clip": 1.04946029,
      "balance_loss_mlp": 1.01970565,
      "epoch": 0.8580532676005531,
      "flos": 19755681367680.0,
      "grad_norm": 2.010545049830162,
      "language_loss": 0.75378633,
      "learning_rate": 2.0753402086072124e-07,
      "loss": 0.77628988,
      "num_input_tokens_seen": 154405965,
      "step": 7136,
      "time_per_iteration": 2.5976436138153076
    },
    {
      "auxiliary_loss_clip": 0.01378242,
      "auxiliary_loss_mlp": 0.01027982,
      "balance_loss_clip": 1.04230499,
      "balance_loss_mlp": 1.02031112,
      "epoch": 0.8581735104911922,
      "flos": 22739318634240.0,
      "grad_norm": 2.221994954516584,
      "language_loss": 0.75383317,
      "learning_rate": 2.071886177636828e-07,
      "loss": 0.77789539,
      "num_input_tokens_seen": 154422750,
      "step": 7137,
      "time_per_iteration": 2.8362371921539307
    },
    {
      "auxiliary_loss_clip": 0.01217448,
      "auxiliary_loss_mlp": 0.01027185,
      "balance_loss_clip": 1.04762173,
      "balance_loss_mlp": 1.02041137,
      "epoch": 0.8582937533818313,
      "flos": 23149131880320.0,
      "grad_norm": 1.7212728703755524,
      "language_loss": 0.83031315,
      "learning_rate": 2.0684348663470575e-07,
      "loss": 0.85275948,
      "num_input_tokens_seen": 154442930,
      "step": 7138,
      "time_per_iteration": 2.7576563358306885
    },
    {
      "auxiliary_loss_clip": 0.01272702,
      "auxiliary_loss_mlp": 0.01022325,
      "balance_loss_clip": 1.04354405,
      "balance_loss_mlp": 1.0151453,
      "epoch": 0.8584139962724704,
      "flos": 19498668577920.0,
      "grad_norm": 1.6676308915267604,
      "language_loss": 0.61683059,
      "learning_rate": 2.0649862752614555e-07,
      "loss": 0.63978088,
      "num_input_tokens_seen": 154461640,
      "step": 7139,
      "time_per_iteration": 2.690241575241089
    },
    {
      "auxiliary_loss_clip": 0.01170531,
      "auxiliary_loss_mlp": 0.01000377,
      "balance_loss_clip": 1.00782657,
      "balance_loss_mlp": 0.99946499,
      "epoch": 0.8585342391631094,
      "flos": 71276577788160.0,
      "grad_norm": 0.7510147933497497,
      "language_loss": 0.57017314,
      "learning_rate": 2.0615404049031838e-07,
      "loss": 0.59188217,
      "num_input_tokens_seen": 154518610,
      "step": 7140,
      "time_per_iteration": 3.2307791709899902
    },
    {
      "auxiliary_loss_clip": 0.01225238,
      "auxiliary_loss_mlp": 0.01030612,
      "balance_loss_clip": 1.05054939,
      "balance_loss_mlp": 1.02304828,
      "epoch": 0.8586544820537486,
      "flos": 10815432929280.0,
      "grad_norm": 2.4997275143379416,
      "language_loss": 0.77643663,
      "learning_rate": 2.0580972557949616e-07,
      "loss": 0.79899514,
      "num_input_tokens_seen": 154533700,
      "step": 7141,
      "time_per_iteration": 2.6286935806274414
    },
    {
      "auxiliary_loss_clip": 0.01120043,
      "auxiliary_loss_mlp": 0.01001317,
      "balance_loss_clip": 1.00654542,
      "balance_loss_mlp": 1.00039887,
      "epoch": 0.8587747249443877,
      "flos": 64811184422400.0,
      "grad_norm": 0.8467397844800908,
      "language_loss": 0.54274869,
      "learning_rate": 2.054656828459125e-07,
      "loss": 0.56396234,
      "num_input_tokens_seen": 154597810,
      "step": 7142,
      "time_per_iteration": 3.2142879962921143
    },
    {
      "auxiliary_loss_clip": 0.01369353,
      "auxiliary_loss_mlp": 0.01027813,
      "balance_loss_clip": 1.04331398,
      "balance_loss_mlp": 1.02043724,
      "epoch": 0.8588949678350267,
      "flos": 26834607964800.0,
      "grad_norm": 1.817438343621723,
      "language_loss": 0.7738778,
      "learning_rate": 2.051219123417578e-07,
      "loss": 0.79784942,
      "num_input_tokens_seen": 154617870,
      "step": 7143,
      "time_per_iteration": 2.8245208263397217
    },
    {
      "auxiliary_loss_clip": 0.01175458,
      "auxiliary_loss_mlp": 0.01024075,
      "balance_loss_clip": 1.04980838,
      "balance_loss_mlp": 1.01673794,
      "epoch": 0.8590152107256659,
      "flos": 26104256726400.0,
      "grad_norm": 2.4940712547793567,
      "language_loss": 0.59963185,
      "learning_rate": 2.0477841411918196e-07,
      "loss": 0.62162715,
      "num_input_tokens_seen": 154637395,
      "step": 7144,
      "time_per_iteration": 2.6424264907836914
    },
    {
      "auxiliary_loss_clip": 0.01215501,
      "auxiliary_loss_mlp": 0.01021967,
      "balance_loss_clip": 1.04522336,
      "balance_loss_mlp": 1.01578355,
      "epoch": 0.859135453616305,
      "flos": 26140885620480.0,
      "grad_norm": 12.609717282113953,
      "language_loss": 0.74803883,
      "learning_rate": 2.0443518823029326e-07,
      "loss": 0.77041352,
      "num_input_tokens_seen": 154657935,
      "step": 7145,
      "time_per_iteration": 2.6270806789398193
    },
    {
      "auxiliary_loss_clip": 0.01312353,
      "auxiliary_loss_mlp": 0.01027337,
      "balance_loss_clip": 1.04254556,
      "balance_loss_mlp": 1.01975822,
      "epoch": 0.859255696506944,
      "flos": 12969319046400.0,
      "grad_norm": 2.106494821809929,
      "language_loss": 0.7683751,
      "learning_rate": 2.0409223472715854e-07,
      "loss": 0.79177201,
      "num_input_tokens_seen": 154675080,
      "step": 7146,
      "time_per_iteration": 2.719205617904663
    },
    {
      "auxiliary_loss_clip": 0.01317403,
      "auxiliary_loss_mlp": 0.02562863,
      "balance_loss_clip": 1.04520059,
      "balance_loss_mlp": 0.99991566,
      "epoch": 0.8593759393975832,
      "flos": 18475753063680.0,
      "grad_norm": 2.1770550375841027,
      "language_loss": 0.74596179,
      "learning_rate": 2.0374955366180434e-07,
      "loss": 0.78476441,
      "num_input_tokens_seen": 154692720,
      "step": 7147,
      "time_per_iteration": 2.687831401824951
    },
    {
      "auxiliary_loss_clip": 0.01324202,
      "auxiliary_loss_mlp": 0.01023538,
      "balance_loss_clip": 1.04213977,
      "balance_loss_mlp": 1.01644492,
      "epoch": 0.8594961822882222,
      "flos": 22200156512640.0,
      "grad_norm": 1.6781721400680045,
      "language_loss": 0.72391748,
      "learning_rate": 2.034071450862147e-07,
      "loss": 0.7473948,
      "num_input_tokens_seen": 154710190,
      "step": 7148,
      "time_per_iteration": 2.7988133430480957
    },
    {
      "auxiliary_loss_clip": 0.0127879,
      "auxiliary_loss_mlp": 0.01027618,
      "balance_loss_clip": 1.04500294,
      "balance_loss_mlp": 1.0204531,
      "epoch": 0.8596164251788613,
      "flos": 23294749616640.0,
      "grad_norm": 1.5609390756973065,
      "language_loss": 0.76770788,
      "learning_rate": 2.030650090523327e-07,
      "loss": 0.79077196,
      "num_input_tokens_seen": 154729380,
      "step": 7149,
      "time_per_iteration": 2.6710429191589355
    },
    {
      "auxiliary_loss_clip": 0.01318523,
      "auxiliary_loss_mlp": 0.01024972,
      "balance_loss_clip": 1.04309607,
      "balance_loss_mlp": 1.01777792,
      "epoch": 0.8597366680695004,
      "flos": 31649905416960.0,
      "grad_norm": 1.7799823762569957,
      "language_loss": 0.5949502,
      "learning_rate": 2.0272314561205995e-07,
      "loss": 0.61838508,
      "num_input_tokens_seen": 154749775,
      "step": 7150,
      "time_per_iteration": 2.7734408378601074
    },
    {
      "auxiliary_loss_clip": 0.01314238,
      "auxiliary_loss_mlp": 0.01020463,
      "balance_loss_clip": 1.03963768,
      "balance_loss_mlp": 1.01406455,
      "epoch": 0.8598569109601395,
      "flos": 21287738211840.0,
      "grad_norm": 1.7853209046283556,
      "language_loss": 0.72763878,
      "learning_rate": 2.023815548172567e-07,
      "loss": 0.75098574,
      "num_input_tokens_seen": 154769845,
      "step": 7151,
      "time_per_iteration": 4.582414150238037
    },
    {
      "auxiliary_loss_clip": 0.01223691,
      "auxiliary_loss_mlp": 0.01029489,
      "balance_loss_clip": 1.04580688,
      "balance_loss_mlp": 1.02219391,
      "epoch": 0.8599771538507786,
      "flos": 25447809720960.0,
      "grad_norm": 1.8592779810667703,
      "language_loss": 0.66011995,
      "learning_rate": 2.0204023671974267e-07,
      "loss": 0.68265176,
      "num_input_tokens_seen": 154789230,
      "step": 7152,
      "time_per_iteration": 2.6357274055480957
    },
    {
      "auxiliary_loss_clip": 0.01217811,
      "auxiliary_loss_mlp": 0.01025857,
      "balance_loss_clip": 1.04571092,
      "balance_loss_mlp": 1.01904166,
      "epoch": 0.8600973967414177,
      "flos": 16723958768640.0,
      "grad_norm": 1.9828782095509374,
      "language_loss": 0.81324464,
      "learning_rate": 2.0169919137129532e-07,
      "loss": 0.83568132,
      "num_input_tokens_seen": 154807670,
      "step": 7153,
      "time_per_iteration": 2.62125301361084
    },
    {
      "auxiliary_loss_clip": 0.01225072,
      "auxiliary_loss_mlp": 0.01028,
      "balance_loss_clip": 1.05093288,
      "balance_loss_mlp": 1.01980472,
      "epoch": 0.8602176396320568,
      "flos": 25227928615680.0,
      "grad_norm": 2.1638525007392615,
      "language_loss": 0.70278633,
      "learning_rate": 2.013584188236508e-07,
      "loss": 0.725317,
      "num_input_tokens_seen": 154825575,
      "step": 7154,
      "time_per_iteration": 3.570202589035034
    },
    {
      "auxiliary_loss_clip": 0.0117574,
      "auxiliary_loss_mlp": 0.01025318,
      "balance_loss_clip": 1.05030239,
      "balance_loss_mlp": 1.01819491,
      "epoch": 0.8603378825226958,
      "flos": 20412236113920.0,
      "grad_norm": 2.380458323043156,
      "language_loss": 0.79587972,
      "learning_rate": 2.0101791912850396e-07,
      "loss": 0.81789029,
      "num_input_tokens_seen": 154845115,
      "step": 7155,
      "time_per_iteration": 2.5747623443603516
    },
    {
      "auxiliary_loss_clip": 0.01275334,
      "auxiliary_loss_mlp": 0.01021921,
      "balance_loss_clip": 1.04954362,
      "balance_loss_mlp": 1.01441383,
      "epoch": 0.8604581254133349,
      "flos": 34930201109760.0,
      "grad_norm": 1.8811023527865982,
      "language_loss": 0.6364454,
      "learning_rate": 2.006776923375082e-07,
      "loss": 0.65941799,
      "num_input_tokens_seen": 154866770,
      "step": 7156,
      "time_per_iteration": 2.825889825820923
    },
    {
      "auxiliary_loss_clip": 0.01172243,
      "auxiliary_loss_mlp": 0.01026545,
      "balance_loss_clip": 1.04820037,
      "balance_loss_mlp": 1.01987791,
      "epoch": 0.860578368303974,
      "flos": 22596538072320.0,
      "grad_norm": 1.689905483320047,
      "language_loss": 0.71493137,
      "learning_rate": 2.003377385022764e-07,
      "loss": 0.73691928,
      "num_input_tokens_seen": 154885595,
      "step": 7157,
      "time_per_iteration": 2.5989699363708496
    },
    {
      "auxiliary_loss_clip": 0.01271257,
      "auxiliary_loss_mlp": 0.01024822,
      "balance_loss_clip": 1.04536366,
      "balance_loss_mlp": 1.01761866,
      "epoch": 0.8606986111946131,
      "flos": 21324331192320.0,
      "grad_norm": 2.4940453684385586,
      "language_loss": 0.76953232,
      "learning_rate": 1.9999805767437826e-07,
      "loss": 0.7924931,
      "num_input_tokens_seen": 154904485,
      "step": 7158,
      "time_per_iteration": 3.518247604370117
    },
    {
      "auxiliary_loss_clip": 0.01268341,
      "auxiliary_loss_mlp": 0.01027152,
      "balance_loss_clip": 1.04402184,
      "balance_loss_mlp": 1.02036285,
      "epoch": 0.8608188540852522,
      "flos": 28877206769280.0,
      "grad_norm": 3.798415079817361,
      "language_loss": 0.71510756,
      "learning_rate": 1.9965864990534386e-07,
      "loss": 0.7380625,
      "num_input_tokens_seen": 154925010,
      "step": 7159,
      "time_per_iteration": 2.6871182918548584
    },
    {
      "auxiliary_loss_clip": 0.01316322,
      "auxiliary_loss_mlp": 0.01024547,
      "balance_loss_clip": 1.03944683,
      "balance_loss_mlp": 1.01774335,
      "epoch": 0.8609390969758913,
      "flos": 29716187713920.0,
      "grad_norm": 1.9732326574087675,
      "language_loss": 0.77501702,
      "learning_rate": 1.9931951524666092e-07,
      "loss": 0.79842567,
      "num_input_tokens_seen": 154946100,
      "step": 7160,
      "time_per_iteration": 2.7822988033294678
    },
    {
      "auxiliary_loss_clip": 0.0122429,
      "auxiliary_loss_mlp": 0.02565347,
      "balance_loss_clip": 1.04638839,
      "balance_loss_mlp": 0.99990845,
      "epoch": 0.8610593398665304,
      "flos": 21249349551360.0,
      "grad_norm": 1.873198388277629,
      "language_loss": 0.81100535,
      "learning_rate": 1.9898065374977534e-07,
      "loss": 0.84890175,
      "num_input_tokens_seen": 154966305,
      "step": 7161,
      "time_per_iteration": 2.6065545082092285
    },
    {
      "auxiliary_loss_clip": 0.01218596,
      "auxiliary_loss_mlp": 0.01018461,
      "balance_loss_clip": 1.04380298,
      "balance_loss_mlp": 1.0124197,
      "epoch": 0.8611795827571694,
      "flos": 14830102183680.0,
      "grad_norm": 1.9499910005323788,
      "language_loss": 0.73438263,
      "learning_rate": 1.9864206546609342e-07,
      "loss": 0.75675321,
      "num_input_tokens_seen": 154985145,
      "step": 7162,
      "time_per_iteration": 2.6885149478912354
    },
    {
      "auxiliary_loss_clip": 0.01172475,
      "auxiliary_loss_mlp": 0.01030108,
      "balance_loss_clip": 1.04755378,
      "balance_loss_mlp": 1.02314639,
      "epoch": 0.8612998256478086,
      "flos": 24243258107520.0,
      "grad_norm": 1.823016193457825,
      "language_loss": 0.84370482,
      "learning_rate": 1.983037504469771e-07,
      "loss": 0.86573064,
      "num_input_tokens_seen": 155003855,
      "step": 7163,
      "time_per_iteration": 2.621830463409424
    },
    {
      "auxiliary_loss_clip": 0.01220495,
      "auxiliary_loss_mlp": 0.01029001,
      "balance_loss_clip": 1.04763126,
      "balance_loss_mlp": 1.02220035,
      "epoch": 0.8614200685384477,
      "flos": 21252653602560.0,
      "grad_norm": 2.0059187960322653,
      "language_loss": 0.66435552,
      "learning_rate": 1.9796570874374984e-07,
      "loss": 0.68685049,
      "num_input_tokens_seen": 155023960,
      "step": 7164,
      "time_per_iteration": 2.6388633251190186
    },
    {
      "auxiliary_loss_clip": 0.01272043,
      "auxiliary_loss_mlp": 0.01033093,
      "balance_loss_clip": 1.04528952,
      "balance_loss_mlp": 1.02614284,
      "epoch": 0.8615403114290867,
      "flos": 20007738080640.0,
      "grad_norm": 3.0333973578680387,
      "language_loss": 0.77476573,
      "learning_rate": 1.976279404076917e-07,
      "loss": 0.79781711,
      "num_input_tokens_seen": 155043360,
      "step": 7165,
      "time_per_iteration": 2.699483633041382
    },
    {
      "auxiliary_loss_clip": 0.01319238,
      "auxiliary_loss_mlp": 0.01024244,
      "balance_loss_clip": 1.04700291,
      "balance_loss_mlp": 1.01704061,
      "epoch": 0.8616605543197259,
      "flos": 29789373674880.0,
      "grad_norm": 1.9335854476139598,
      "language_loss": 0.76206052,
      "learning_rate": 1.9729044549004193e-07,
      "loss": 0.78549534,
      "num_input_tokens_seen": 155064745,
      "step": 7166,
      "time_per_iteration": 2.7774856090545654
    },
    {
      "auxiliary_loss_clip": 0.01218707,
      "auxiliary_loss_mlp": 0.01026516,
      "balance_loss_clip": 1.04781675,
      "balance_loss_mlp": 1.02029943,
      "epoch": 0.8617807972103649,
      "flos": 28911609020160.0,
      "grad_norm": 1.9064939964129115,
      "language_loss": 0.70284086,
      "learning_rate": 1.9695322404199822e-07,
      "loss": 0.7252931,
      "num_input_tokens_seen": 155086790,
      "step": 7167,
      "time_per_iteration": 2.705867290496826
    },
    {
      "auxiliary_loss_clip": 0.01280381,
      "auxiliary_loss_mlp": 0.01028167,
      "balance_loss_clip": 1.05013788,
      "balance_loss_mlp": 1.02117562,
      "epoch": 0.861901040101004,
      "flos": 27673804391040.0,
      "grad_norm": 1.86758850406926,
      "language_loss": 0.82095194,
      "learning_rate": 1.9661627611471654e-07,
      "loss": 0.84403747,
      "num_input_tokens_seen": 155106585,
      "step": 7168,
      "time_per_iteration": 2.7076170444488525
    },
    {
      "auxiliary_loss_clip": 0.01175737,
      "auxiliary_loss_mlp": 0.01022975,
      "balance_loss_clip": 1.04525995,
      "balance_loss_mlp": 1.01571798,
      "epoch": 0.8620212829916432,
      "flos": 49748056755840.0,
      "grad_norm": 4.493968314864206,
      "language_loss": 0.7001828,
      "learning_rate": 1.9627960175931246e-07,
      "loss": 0.72216988,
      "num_input_tokens_seen": 155131285,
      "step": 7169,
      "time_per_iteration": 3.0105855464935303
    },
    {
      "auxiliary_loss_clip": 0.0122417,
      "auxiliary_loss_mlp": 0.01023601,
      "balance_loss_clip": 1.05042434,
      "balance_loss_mlp": 1.01701164,
      "epoch": 0.8621415258822822,
      "flos": 21138672769920.0,
      "grad_norm": 1.6810518969704322,
      "language_loss": 0.74404496,
      "learning_rate": 1.9594320102685847e-07,
      "loss": 0.76652265,
      "num_input_tokens_seen": 155150555,
      "step": 7170,
      "time_per_iteration": 2.6171040534973145
    },
    {
      "auxiliary_loss_clip": 0.01263967,
      "auxiliary_loss_mlp": 0.02563841,
      "balance_loss_clip": 1.0431236,
      "balance_loss_mlp": 0.99991208,
      "epoch": 0.8622617687729213,
      "flos": 21689039934720.0,
      "grad_norm": 1.940721954219524,
      "language_loss": 0.64087105,
      "learning_rate": 1.956070739683864e-07,
      "loss": 0.67914915,
      "num_input_tokens_seen": 155169890,
      "step": 7171,
      "time_per_iteration": 2.786644697189331
    },
    {
      "auxiliary_loss_clip": 0.01311572,
      "auxiliary_loss_mlp": 0.01022814,
      "balance_loss_clip": 1.0412513,
      "balance_loss_mlp": 1.01634133,
      "epoch": 0.8623820116635604,
      "flos": 26250592734720.0,
      "grad_norm": 2.070604679665568,
      "language_loss": 0.74486107,
      "learning_rate": 1.9527122063488678e-07,
      "loss": 0.76820493,
      "num_input_tokens_seen": 155191005,
      "step": 7172,
      "time_per_iteration": 2.713794469833374
    },
    {
      "auxiliary_loss_clip": 0.0127222,
      "auxiliary_loss_mlp": 0.01022285,
      "balance_loss_clip": 1.04336524,
      "balance_loss_mlp": 1.01552296,
      "epoch": 0.8625022545541995,
      "flos": 19647554451840.0,
      "grad_norm": 2.1153041797573433,
      "language_loss": 0.80351096,
      "learning_rate": 1.9493564107730755e-07,
      "loss": 0.82645601,
      "num_input_tokens_seen": 155211005,
      "step": 7173,
      "time_per_iteration": 2.7079355716705322
    },
    {
      "auxiliary_loss_clip": 0.01267572,
      "auxiliary_loss_mlp": 0.01028757,
      "balance_loss_clip": 1.04168701,
      "balance_loss_mlp": 1.02225375,
      "epoch": 0.8626224974448385,
      "flos": 21908382336000.0,
      "grad_norm": 1.8772516577209177,
      "language_loss": 0.61001527,
      "learning_rate": 1.9460033534655684e-07,
      "loss": 0.63297856,
      "num_input_tokens_seen": 155230365,
      "step": 7174,
      "time_per_iteration": 2.701894760131836
    },
    {
      "auxiliary_loss_clip": 0.01269862,
      "auxiliary_loss_mlp": 0.0102344,
      "balance_loss_clip": 1.04194093,
      "balance_loss_mlp": 1.01667833,
      "epoch": 0.8627427403354777,
      "flos": 23331198942720.0,
      "grad_norm": 1.5455928229661164,
      "language_loss": 0.8427462,
      "learning_rate": 1.9426530349349978e-07,
      "loss": 0.86567926,
      "num_input_tokens_seen": 155250815,
      "step": 7175,
      "time_per_iteration": 2.6765291690826416
    },
    {
      "auxiliary_loss_clip": 0.01219967,
      "auxiliary_loss_mlp": 0.02562666,
      "balance_loss_clip": 1.04526794,
      "balance_loss_mlp": 0.99990308,
      "epoch": 0.8628629832261168,
      "flos": 16362877299840.0,
      "grad_norm": 2.7346719414381475,
      "language_loss": 0.64729232,
      "learning_rate": 1.9393054556896038e-07,
      "loss": 0.68511868,
      "num_input_tokens_seen": 155268515,
      "step": 7176,
      "time_per_iteration": 2.6293444633483887
    },
    {
      "auxiliary_loss_clip": 0.0122153,
      "auxiliary_loss_mlp": 0.0102619,
      "balance_loss_clip": 1.04386234,
      "balance_loss_mlp": 1.01901031,
      "epoch": 0.8629832261167558,
      "flos": 28103941756800.0,
      "grad_norm": 2.266342483083716,
      "language_loss": 0.69496799,
      "learning_rate": 1.9359606162372133e-07,
      "loss": 0.71744514,
      "num_input_tokens_seen": 155290120,
      "step": 7177,
      "time_per_iteration": 3.6562130451202393
    },
    {
      "auxiliary_loss_clip": 0.01172727,
      "auxiliary_loss_mlp": 0.01025429,
      "balance_loss_clip": 1.04924583,
      "balance_loss_mlp": 1.01897395,
      "epoch": 0.863103469007395,
      "flos": 20230061310720.0,
      "grad_norm": 2.1916132636969206,
      "language_loss": 0.70769846,
      "learning_rate": 1.9326185170852293e-07,
      "loss": 0.72968,
      "num_input_tokens_seen": 155309085,
      "step": 7178,
      "time_per_iteration": 3.5124011039733887
    },
    {
      "auxiliary_loss_clip": 0.0122411,
      "auxiliary_loss_mlp": 0.01030928,
      "balance_loss_clip": 1.04837394,
      "balance_loss_mlp": 1.02386165,
      "epoch": 0.863223711898034,
      "flos": 24498547044480.0,
      "grad_norm": 4.511605974562868,
      "language_loss": 0.72151148,
      "learning_rate": 1.9292791587406598e-07,
      "loss": 0.74406189,
      "num_input_tokens_seen": 155327945,
      "step": 7179,
      "time_per_iteration": 2.672240734100342
    },
    {
      "auxiliary_loss_clip": 0.01219477,
      "auxiliary_loss_mlp": 0.02565467,
      "balance_loss_clip": 1.04409528,
      "balance_loss_mlp": 0.99988484,
      "epoch": 0.8633439547886731,
      "flos": 17675376261120.0,
      "grad_norm": 2.190167298456796,
      "language_loss": 0.86937094,
      "learning_rate": 1.9259425417100661e-07,
      "loss": 0.90722036,
      "num_input_tokens_seen": 155344060,
      "step": 7180,
      "time_per_iteration": 3.555724859237671
    },
    {
      "auxiliary_loss_clip": 0.01419702,
      "auxiliary_loss_mlp": 0.01024305,
      "balance_loss_clip": 1.03488529,
      "balance_loss_mlp": 1.01730728,
      "epoch": 0.8634641976793123,
      "flos": 12895055677440.0,
      "grad_norm": 2.0837146848930796,
      "language_loss": 0.74909675,
      "learning_rate": 1.9226086664996234e-07,
      "loss": 0.7735368,
      "num_input_tokens_seen": 155362305,
      "step": 7181,
      "time_per_iteration": 2.805751323699951
    },
    {
      "auxiliary_loss_clip": 0.01173216,
      "auxiliary_loss_mlp": 0.01026851,
      "balance_loss_clip": 1.04702652,
      "balance_loss_mlp": 1.02032185,
      "epoch": 0.8635844405699513,
      "flos": 23878980328320.0,
      "grad_norm": 1.903245459564033,
      "language_loss": 0.74139762,
      "learning_rate": 1.9192775336150712e-07,
      "loss": 0.76339829,
      "num_input_tokens_seen": 155382605,
      "step": 7182,
      "time_per_iteration": 2.63535213470459
    },
    {
      "auxiliary_loss_clip": 0.01111134,
      "auxiliary_loss_mlp": 0.00998291,
      "balance_loss_clip": 1.00692785,
      "balance_loss_mlp": 0.99735481,
      "epoch": 0.8637046834605904,
      "flos": 60453387521280.0,
      "grad_norm": 0.7721358135522656,
      "language_loss": 0.56216431,
      "learning_rate": 1.915949143561739e-07,
      "loss": 0.58325857,
      "num_input_tokens_seen": 155437280,
      "step": 7183,
      "time_per_iteration": 3.1802616119384766
    },
    {
      "auxiliary_loss_clip": 0.01221216,
      "auxiliary_loss_mlp": 0.01030688,
      "balance_loss_clip": 1.04861164,
      "balance_loss_mlp": 1.0233326,
      "epoch": 0.8638249263512295,
      "flos": 20558751690240.0,
      "grad_norm": 2.2501137489047305,
      "language_loss": 0.77818286,
      "learning_rate": 1.9126234968445498e-07,
      "loss": 0.80070192,
      "num_input_tokens_seen": 155456970,
      "step": 7184,
      "time_per_iteration": 3.5913138389587402
    },
    {
      "auxiliary_loss_clip": 0.01172264,
      "auxiliary_loss_mlp": 0.01024194,
      "balance_loss_clip": 1.04801881,
      "balance_loss_mlp": 1.0176909,
      "epoch": 0.8639451692418686,
      "flos": 26615768353920.0,
      "grad_norm": 1.5458200634469823,
      "language_loss": 0.67901123,
      "learning_rate": 1.9093005939679884e-07,
      "loss": 0.70097589,
      "num_input_tokens_seen": 155478925,
      "step": 7185,
      "time_per_iteration": 2.62943172454834
    },
    {
      "auxiliary_loss_clip": 0.01222398,
      "auxiliary_loss_mlp": 0.01028258,
      "balance_loss_clip": 1.04922962,
      "balance_loss_mlp": 1.02133822,
      "epoch": 0.8640654121325076,
      "flos": 15122450977920.0,
      "grad_norm": 1.9496167194418843,
      "language_loss": 0.76618242,
      "learning_rate": 1.9059804354361452e-07,
      "loss": 0.78868896,
      "num_input_tokens_seen": 155496700,
      "step": 7186,
      "time_per_iteration": 2.6187753677368164
    },
    {
      "auxiliary_loss_clip": 0.01270868,
      "auxiliary_loss_mlp": 0.01027608,
      "balance_loss_clip": 1.04282463,
      "balance_loss_mlp": 1.02055693,
      "epoch": 0.8641856550231467,
      "flos": 31869068250240.0,
      "grad_norm": 1.9112062474111353,
      "language_loss": 0.70284939,
      "learning_rate": 1.902663021752684e-07,
      "loss": 0.72583413,
      "num_input_tokens_seen": 155518130,
      "step": 7187,
      "time_per_iteration": 2.7001821994781494
    },
    {
      "auxiliary_loss_clip": 0.0117584,
      "auxiliary_loss_mlp": 0.01033326,
      "balance_loss_clip": 1.05003166,
      "balance_loss_mlp": 1.026335,
      "epoch": 0.8643058979137859,
      "flos": 14976545932800.0,
      "grad_norm": 2.2798443996432542,
      "language_loss": 0.82632798,
      "learning_rate": 1.8993483534208556e-07,
      "loss": 0.84841967,
      "num_input_tokens_seen": 155537040,
      "step": 7188,
      "time_per_iteration": 2.588066339492798
    },
    {
      "auxiliary_loss_clip": 0.01279041,
      "auxiliary_loss_mlp": 0.01027492,
      "balance_loss_clip": 1.05087686,
      "balance_loss_mlp": 1.01985025,
      "epoch": 0.8644261408044249,
      "flos": 13115726881920.0,
      "grad_norm": 2.735224237645859,
      "language_loss": 0.74824405,
      "learning_rate": 1.8960364309434884e-07,
      "loss": 0.77130938,
      "num_input_tokens_seen": 155554535,
      "step": 7189,
      "time_per_iteration": 2.6173079013824463
    },
    {
      "auxiliary_loss_clip": 0.01415811,
      "auxiliary_loss_mlp": 0.02563791,
      "balance_loss_clip": 1.04164183,
      "balance_loss_mlp": 0.99990344,
      "epoch": 0.864546383695064,
      "flos": 20850920916480.0,
      "grad_norm": 1.865162389482846,
      "language_loss": 0.78161174,
      "learning_rate": 1.8927272548229967e-07,
      "loss": 0.82140774,
      "num_input_tokens_seen": 155574225,
      "step": 7190,
      "time_per_iteration": 2.769493818283081
    },
    {
      "auxiliary_loss_clip": 0.01375087,
      "auxiliary_loss_mlp": 0.01028239,
      "balance_loss_clip": 1.04566646,
      "balance_loss_mlp": 1.02174544,
      "epoch": 0.8646666265857031,
      "flos": 21324582587520.0,
      "grad_norm": 1.4982892323362118,
      "language_loss": 0.83271432,
      "learning_rate": 1.8894208255613876e-07,
      "loss": 0.85674757,
      "num_input_tokens_seen": 155593540,
      "step": 7191,
      "time_per_iteration": 2.742063522338867
    },
    {
      "auxiliary_loss_clip": 0.01171869,
      "auxiliary_loss_mlp": 0.01021309,
      "balance_loss_clip": 1.04792178,
      "balance_loss_mlp": 1.01476431,
      "epoch": 0.8647868694763422,
      "flos": 19750833031680.0,
      "grad_norm": 2.023824958781724,
      "language_loss": 0.77432513,
      "learning_rate": 1.8861171436602397e-07,
      "loss": 0.79625684,
      "num_input_tokens_seen": 155610655,
      "step": 7192,
      "time_per_iteration": 2.5917980670928955
    },
    {
      "auxiliary_loss_clip": 0.01226487,
      "auxiliary_loss_mlp": 0.01022253,
      "balance_loss_clip": 1.05078876,
      "balance_loss_mlp": 1.01552117,
      "epoch": 0.8649071123669813,
      "flos": 26176760328960.0,
      "grad_norm": 2.522767661263201,
      "language_loss": 0.80559909,
      "learning_rate": 1.882816209620719e-07,
      "loss": 0.8280865,
      "num_input_tokens_seen": 155627365,
      "step": 7193,
      "time_per_iteration": 2.6608574390411377
    },
    {
      "auxiliary_loss_clip": 0.01283288,
      "auxiliary_loss_mlp": 0.01027732,
      "balance_loss_clip": 1.05179238,
      "balance_loss_mlp": 1.02072823,
      "epoch": 0.8650273552576204,
      "flos": 20302888135680.0,
      "grad_norm": 1.7797905438977808,
      "language_loss": 0.77399361,
      "learning_rate": 1.8795180239435738e-07,
      "loss": 0.79710376,
      "num_input_tokens_seen": 155646220,
      "step": 7194,
      "time_per_iteration": 2.7024707794189453
    },
    {
      "auxiliary_loss_clip": 0.01277665,
      "auxiliary_loss_mlp": 0.01029222,
      "balance_loss_clip": 1.04729128,
      "balance_loss_mlp": 1.02207303,
      "epoch": 0.8651475981482595,
      "flos": 23951088881280.0,
      "grad_norm": 6.13625742153834,
      "language_loss": 0.76148522,
      "learning_rate": 1.8762225871291348e-07,
      "loss": 0.78455406,
      "num_input_tokens_seen": 155662095,
      "step": 7195,
      "time_per_iteration": 2.73245906829834
    },
    {
      "auxiliary_loss_clip": 0.01172342,
      "auxiliary_loss_mlp": 0.02565554,
      "balance_loss_clip": 1.0481689,
      "balance_loss_mlp": 0.99990481,
      "epoch": 0.8652678410388985,
      "flos": 21684622561920.0,
      "grad_norm": 1.5584366898567708,
      "language_loss": 0.8090359,
      "learning_rate": 1.8729298996773201e-07,
      "loss": 0.84641492,
      "num_input_tokens_seen": 155680845,
      "step": 7196,
      "time_per_iteration": 2.5950980186462402
    },
    {
      "auxiliary_loss_clip": 0.01108346,
      "auxiliary_loss_mlp": 0.01001595,
      "balance_loss_clip": 1.00723612,
      "balance_loss_mlp": 1.00067735,
      "epoch": 0.8653880839295377,
      "flos": 65224660855680.0,
      "grad_norm": 0.8299916907826251,
      "language_loss": 0.60887146,
      "learning_rate": 1.8696399620876301e-07,
      "loss": 0.62997091,
      "num_input_tokens_seen": 155737875,
      "step": 7197,
      "time_per_iteration": 3.1500356197357178
    },
    {
      "auxiliary_loss_clip": 0.01323008,
      "auxiliary_loss_mlp": 0.01022269,
      "balance_loss_clip": 1.04020691,
      "balance_loss_mlp": 1.01492524,
      "epoch": 0.8655083268201768,
      "flos": 17749172753280.0,
      "grad_norm": 2.3432726519294182,
      "language_loss": 0.79402125,
      "learning_rate": 1.866352774859141e-07,
      "loss": 0.81747401,
      "num_input_tokens_seen": 155753100,
      "step": 7198,
      "time_per_iteration": 2.6741766929626465
    },
    {
      "auxiliary_loss_clip": 0.01325055,
      "auxiliary_loss_mlp": 0.01024062,
      "balance_loss_clip": 1.0429014,
      "balance_loss_mlp": 1.01794362,
      "epoch": 0.8656285697108158,
      "flos": 20703974376960.0,
      "grad_norm": 2.3994209309085113,
      "language_loss": 0.69334626,
      "learning_rate": 1.8630683384905188e-07,
      "loss": 0.71683741,
      "num_input_tokens_seen": 155772430,
      "step": 7199,
      "time_per_iteration": 2.794703483581543
    },
    {
      "auxiliary_loss_clip": 0.01173807,
      "auxiliary_loss_mlp": 0.02567874,
      "balance_loss_clip": 1.05063272,
      "balance_loss_mlp": 0.99991518,
      "epoch": 0.865748812601455,
      "flos": 18653833716480.0,
      "grad_norm": 2.000595194377792,
      "language_loss": 0.8895632,
      "learning_rate": 1.8597866534800045e-07,
      "loss": 0.92698002,
      "num_input_tokens_seen": 155787545,
      "step": 7200,
      "time_per_iteration": 2.589456558227539
    },
    {
      "auxiliary_loss_clip": 0.01224341,
      "auxiliary_loss_mlp": 0.02565935,
      "balance_loss_clip": 1.04640305,
      "balance_loss_mlp": 0.99989861,
      "epoch": 0.865869055492094,
      "flos": 70652554807680.0,
      "grad_norm": 2.705367504852749,
      "language_loss": 0.74611151,
      "learning_rate": 1.8565077203254398e-07,
      "loss": 0.78401434,
      "num_input_tokens_seen": 155813005,
      "step": 7201,
      "time_per_iteration": 2.9976017475128174
    },
    {
      "auxiliary_loss_clip": 0.01322553,
      "auxiliary_loss_mlp": 0.01025118,
      "balance_loss_clip": 1.04968739,
      "balance_loss_mlp": 1.01809907,
      "epoch": 0.8659892983827331,
      "flos": 17383961220480.0,
      "grad_norm": 4.341621125366666,
      "language_loss": 0.73248386,
      "learning_rate": 1.8532315395242203e-07,
      "loss": 0.75596058,
      "num_input_tokens_seen": 155829455,
      "step": 7202,
      "time_per_iteration": 2.645419120788574
    },
    {
      "auxiliary_loss_clip": 0.01320166,
      "auxiliary_loss_mlp": 0.01023061,
      "balance_loss_clip": 1.04390478,
      "balance_loss_mlp": 1.01632297,
      "epoch": 0.8661095412733723,
      "flos": 17895221452800.0,
      "grad_norm": 1.9634900977293166,
      "language_loss": 0.72196138,
      "learning_rate": 1.849958111573353e-07,
      "loss": 0.74539369,
      "num_input_tokens_seen": 155848060,
      "step": 7203,
      "time_per_iteration": 3.556546449661255
    },
    {
      "auxiliary_loss_clip": 0.0116917,
      "auxiliary_loss_mlp": 0.01026236,
      "balance_loss_clip": 1.04736066,
      "balance_loss_mlp": 1.01945853,
      "epoch": 0.8662297841640113,
      "flos": 18224163227520.0,
      "grad_norm": 1.7020681872076473,
      "language_loss": 0.64407933,
      "learning_rate": 1.8466874369694074e-07,
      "loss": 0.66603339,
      "num_input_tokens_seen": 155865755,
      "step": 7204,
      "time_per_iteration": 3.5028061866760254
    },
    {
      "auxiliary_loss_clip": 0.01322904,
      "auxiliary_loss_mlp": 0.01019947,
      "balance_loss_clip": 1.04037023,
      "balance_loss_mlp": 1.01336658,
      "epoch": 0.8663500270546504,
      "flos": 16362159027840.0,
      "grad_norm": 3.027565860232214,
      "language_loss": 0.70131773,
      "learning_rate": 1.843419516208542e-07,
      "loss": 0.72474629,
      "num_input_tokens_seen": 155882680,
      "step": 7205,
      "time_per_iteration": 2.660959005355835
    },
    {
      "auxiliary_loss_clip": 0.01225574,
      "auxiliary_loss_mlp": 0.01025592,
      "balance_loss_clip": 1.04999161,
      "balance_loss_mlp": 1.01817703,
      "epoch": 0.8664702699452895,
      "flos": 17894431353600.0,
      "grad_norm": 2.1104661656395822,
      "language_loss": 0.79567289,
      "learning_rate": 1.8401543497865047e-07,
      "loss": 0.81818461,
      "num_input_tokens_seen": 155900680,
      "step": 7206,
      "time_per_iteration": 3.514117956161499
    },
    {
      "auxiliary_loss_clip": 0.01224197,
      "auxiliary_loss_mlp": 0.02567145,
      "balance_loss_clip": 1.04792309,
      "balance_loss_mlp": 0.99989009,
      "epoch": 0.8665905128359286,
      "flos": 30736373794560.0,
      "grad_norm": 2.0033810856031584,
      "language_loss": 0.6404177,
      "learning_rate": 1.836891938198608e-07,
      "loss": 0.67833108,
      "num_input_tokens_seen": 155921105,
      "step": 7207,
      "time_per_iteration": 2.7039260864257812
    },
    {
      "auxiliary_loss_clip": 0.01272939,
      "auxiliary_loss_mlp": 0.01024881,
      "balance_loss_clip": 1.04829717,
      "balance_loss_mlp": 1.01772237,
      "epoch": 0.8667107557265676,
      "flos": 18656419495680.0,
      "grad_norm": 2.4550616137577537,
      "language_loss": 0.71529543,
      "learning_rate": 1.8336322819397677e-07,
      "loss": 0.73827362,
      "num_input_tokens_seen": 155938640,
      "step": 7208,
      "time_per_iteration": 2.613107204437256
    },
    {
      "auxiliary_loss_clip": 0.0132791,
      "auxiliary_loss_mlp": 0.01025086,
      "balance_loss_clip": 1.04338717,
      "balance_loss_mlp": 1.0181663,
      "epoch": 0.8668309986172068,
      "flos": 20083725302400.0,
      "grad_norm": 1.9375337519604254,
      "language_loss": 0.62529677,
      "learning_rate": 1.8303753815044654e-07,
      "loss": 0.64882672,
      "num_input_tokens_seen": 155957945,
      "step": 7209,
      "time_per_iteration": 2.7591116428375244
    },
    {
      "auxiliary_loss_clip": 0.01284231,
      "auxiliary_loss_mlp": 0.01026988,
      "balance_loss_clip": 1.04775751,
      "balance_loss_mlp": 1.01945734,
      "epoch": 0.8669512415078459,
      "flos": 21615099788160.0,
      "grad_norm": 2.298840181286915,
      "language_loss": 0.71009421,
      "learning_rate": 1.827121237386773e-07,
      "loss": 0.73320639,
      "num_input_tokens_seen": 155975390,
      "step": 7210,
      "time_per_iteration": 3.569711208343506
    },
    {
      "auxiliary_loss_clip": 0.01274785,
      "auxiliary_loss_mlp": 0.01025159,
      "balance_loss_clip": 1.04673839,
      "balance_loss_mlp": 1.01801276,
      "epoch": 0.8670714843984849,
      "flos": 17703601372800.0,
      "grad_norm": 2.8430004156555984,
      "language_loss": 0.7517451,
      "learning_rate": 1.8238698500803374e-07,
      "loss": 0.77474451,
      "num_input_tokens_seen": 155988155,
      "step": 7211,
      "time_per_iteration": 2.697288990020752
    },
    {
      "auxiliary_loss_clip": 0.01117747,
      "auxiliary_loss_mlp": 0.01001042,
      "balance_loss_clip": 1.00650358,
      "balance_loss_mlp": 1.00016022,
      "epoch": 0.8671917272891241,
      "flos": 60705483125760.0,
      "grad_norm": 0.7156548181386322,
      "language_loss": 0.5622319,
      "learning_rate": 1.820621220078391e-07,
      "loss": 0.5834198,
      "num_input_tokens_seen": 156052065,
      "step": 7212,
      "time_per_iteration": 3.2645740509033203
    },
    {
      "auxiliary_loss_clip": 0.01173797,
      "auxiliary_loss_mlp": 0.01025815,
      "balance_loss_clip": 1.04918134,
      "balance_loss_mlp": 1.01874638,
      "epoch": 0.8673119701797631,
      "flos": 20451881750400.0,
      "grad_norm": 2.46679630088311,
      "language_loss": 0.67644751,
      "learning_rate": 1.8173753478737553e-07,
      "loss": 0.69844359,
      "num_input_tokens_seen": 156072500,
      "step": 7213,
      "time_per_iteration": 2.5928521156311035
    },
    {
      "auxiliary_loss_clip": 0.01174349,
      "auxiliary_loss_mlp": 0.01026124,
      "balance_loss_clip": 1.04954004,
      "balance_loss_mlp": 1.01837277,
      "epoch": 0.8674322130704022,
      "flos": 19647410797440.0,
      "grad_norm": 2.225992147525365,
      "language_loss": 0.80364728,
      "learning_rate": 1.8141322339588205e-07,
      "loss": 0.825652,
      "num_input_tokens_seen": 156089840,
      "step": 7214,
      "time_per_iteration": 2.591423273086548
    },
    {
      "auxiliary_loss_clip": 0.01171253,
      "auxiliary_loss_mlp": 0.01026957,
      "balance_loss_clip": 1.04864216,
      "balance_loss_mlp": 1.01993561,
      "epoch": 0.8675524559610414,
      "flos": 26025001367040.0,
      "grad_norm": 2.415504958278649,
      "language_loss": 0.6998229,
      "learning_rate": 1.810891878825569e-07,
      "loss": 0.72180504,
      "num_input_tokens_seen": 156109815,
      "step": 7215,
      "time_per_iteration": 2.586688280105591
    },
    {
      "auxiliary_loss_clip": 0.01271542,
      "auxiliary_loss_mlp": 0.01023802,
      "balance_loss_clip": 1.04529583,
      "balance_loss_mlp": 1.01711416,
      "epoch": 0.8676726988516804,
      "flos": 15049444584960.0,
      "grad_norm": 2.1522811927799554,
      "language_loss": 0.72049844,
      "learning_rate": 1.8076542829655561e-07,
      "loss": 0.74345183,
      "num_input_tokens_seen": 156128620,
      "step": 7216,
      "time_per_iteration": 2.7121453285217285
    },
    {
      "auxiliary_loss_clip": 0.01278604,
      "auxiliary_loss_mlp": 0.01023015,
      "balance_loss_clip": 1.04985595,
      "balance_loss_mlp": 1.01585674,
      "epoch": 0.8677929417423195,
      "flos": 16288111140480.0,
      "grad_norm": 2.4808623224022193,
      "language_loss": 0.79406095,
      "learning_rate": 1.8044194468699203e-07,
      "loss": 0.81707716,
      "num_input_tokens_seen": 156145930,
      "step": 7217,
      "time_per_iteration": 2.6733429431915283
    },
    {
      "auxiliary_loss_clip": 0.01269554,
      "auxiliary_loss_mlp": 0.01023816,
      "balance_loss_clip": 1.04838121,
      "balance_loss_mlp": 1.01697302,
      "epoch": 0.8679131846329585,
      "flos": 18844160906880.0,
      "grad_norm": 2.104588401022987,
      "language_loss": 0.75677884,
      "learning_rate": 1.8011873710293912e-07,
      "loss": 0.77971256,
      "num_input_tokens_seen": 156164435,
      "step": 7218,
      "time_per_iteration": 2.7172179222106934
    },
    {
      "auxiliary_loss_clip": 0.01221892,
      "auxiliary_loss_mlp": 0.01024368,
      "balance_loss_clip": 1.0482409,
      "balance_loss_mlp": 1.01773357,
      "epoch": 0.8680334275235977,
      "flos": 33620718890880.0,
      "grad_norm": 2.0423517172518775,
      "language_loss": 0.69153291,
      "learning_rate": 1.7979580559342677e-07,
      "loss": 0.71399552,
      "num_input_tokens_seen": 156185165,
      "step": 7219,
      "time_per_iteration": 2.68587327003479
    },
    {
      "auxiliary_loss_clip": 0.01271319,
      "auxiliary_loss_mlp": 0.01026742,
      "balance_loss_clip": 1.04628408,
      "balance_loss_mlp": 1.0195179,
      "epoch": 0.8681536704142367,
      "flos": 24681152810880.0,
      "grad_norm": 1.766303338033194,
      "language_loss": 0.66647184,
      "learning_rate": 1.7947315020744358e-07,
      "loss": 0.68945247,
      "num_input_tokens_seen": 156206260,
      "step": 7220,
      "time_per_iteration": 2.7502055168151855
    },
    {
      "auxiliary_loss_clip": 0.01269476,
      "auxiliary_loss_mlp": 0.01028739,
      "balance_loss_clip": 1.04480457,
      "balance_loss_mlp": 1.02260303,
      "epoch": 0.8682739133048758,
      "flos": 20011042131840.0,
      "grad_norm": 3.325664489634154,
      "language_loss": 0.80545592,
      "learning_rate": 1.7915077099393594e-07,
      "loss": 0.82843804,
      "num_input_tokens_seen": 156222860,
      "step": 7221,
      "time_per_iteration": 2.6302578449249268
    },
    {
      "auxiliary_loss_clip": 0.01226202,
      "auxiliary_loss_mlp": 0.01021135,
      "balance_loss_clip": 1.04746306,
      "balance_loss_mlp": 1.01402986,
      "epoch": 0.868394156195515,
      "flos": 16654759217280.0,
      "grad_norm": 2.830891944777469,
      "language_loss": 0.73185372,
      "learning_rate": 1.788286680018083e-07,
      "loss": 0.75432706,
      "num_input_tokens_seen": 156241570,
      "step": 7222,
      "time_per_iteration": 2.618861675262451
    },
    {
      "auxiliary_loss_clip": 0.01272075,
      "auxiliary_loss_mlp": 0.01026779,
      "balance_loss_clip": 1.0455339,
      "balance_loss_mlp": 1.02000237,
      "epoch": 0.868514399086154,
      "flos": 28001381448960.0,
      "grad_norm": 3.755842915359066,
      "language_loss": 0.72285926,
      "learning_rate": 1.7850684127992443e-07,
      "loss": 0.74584782,
      "num_input_tokens_seen": 156261315,
      "step": 7223,
      "time_per_iteration": 2.76210880279541
    },
    {
      "auxiliary_loss_clip": 0.01322946,
      "auxiliary_loss_mlp": 0.01027469,
      "balance_loss_clip": 1.04738569,
      "balance_loss_mlp": 1.02040625,
      "epoch": 0.8686346419767931,
      "flos": 20084587228800.0,
      "grad_norm": 2.354893433993935,
      "language_loss": 0.7037437,
      "learning_rate": 1.7818529087710378e-07,
      "loss": 0.72724789,
      "num_input_tokens_seen": 156281670,
      "step": 7224,
      "time_per_iteration": 2.697479009628296
    },
    {
      "auxiliary_loss_clip": 0.01220591,
      "auxiliary_loss_mlp": 0.02565952,
      "balance_loss_clip": 1.04672146,
      "balance_loss_mlp": 0.99990398,
      "epoch": 0.8687548848674322,
      "flos": 18223516782720.0,
      "grad_norm": 2.1012541489526946,
      "language_loss": 0.84467167,
      "learning_rate": 1.7786401684212637e-07,
      "loss": 0.88253713,
      "num_input_tokens_seen": 156300500,
      "step": 7225,
      "time_per_iteration": 2.697335958480835
    },
    {
      "auxiliary_loss_clip": 0.0111628,
      "auxiliary_loss_mlp": 0.01001239,
      "balance_loss_clip": 1.00643122,
      "balance_loss_mlp": 1.00037479,
      "epoch": 0.8688751277580713,
      "flos": 70457885049600.0,
      "grad_norm": 0.7296660817934916,
      "language_loss": 0.55883682,
      "learning_rate": 1.7754301922372883e-07,
      "loss": 0.58001208,
      "num_input_tokens_seen": 156350145,
      "step": 7226,
      "time_per_iteration": 3.109631299972534
    },
    {
      "auxiliary_loss_clip": 0.01430915,
      "auxiliary_loss_mlp": 0.01026452,
      "balance_loss_clip": 1.04252911,
      "balance_loss_mlp": 1.01884913,
      "epoch": 0.8689953706487104,
      "flos": 26906788344960.0,
      "grad_norm": 2.857138800371409,
      "language_loss": 0.81374323,
      "learning_rate": 1.7722229807060617e-07,
      "loss": 0.83831692,
      "num_input_tokens_seen": 156368725,
      "step": 7227,
      "time_per_iteration": 2.8185858726501465
    },
    {
      "auxiliary_loss_clip": 0.01314012,
      "auxiliary_loss_mlp": 0.0102637,
      "balance_loss_clip": 1.03965974,
      "balance_loss_mlp": 1.01972961,
      "epoch": 0.8691156135393495,
      "flos": 34637385438720.0,
      "grad_norm": 2.3841697679748437,
      "language_loss": 0.82301927,
      "learning_rate": 1.7690185343141172e-07,
      "loss": 0.84642309,
      "num_input_tokens_seen": 156388640,
      "step": 7228,
      "time_per_iteration": 2.777899980545044
    },
    {
      "auxiliary_loss_clip": 0.01272073,
      "auxiliary_loss_mlp": 0.01022536,
      "balance_loss_clip": 1.04518461,
      "balance_loss_mlp": 1.01624775,
      "epoch": 0.8692358564299886,
      "flos": 18989814556800.0,
      "grad_norm": 1.9567845518691074,
      "language_loss": 0.69787502,
      "learning_rate": 1.7658168535475615e-07,
      "loss": 0.72082114,
      "num_input_tokens_seen": 156406425,
      "step": 7229,
      "time_per_iteration": 3.5387065410614014
    },
    {
      "auxiliary_loss_clip": 0.0127759,
      "auxiliary_loss_mlp": 0.01025039,
      "balance_loss_clip": 1.04890585,
      "balance_loss_mlp": 1.01808,
      "epoch": 0.8693560993206276,
      "flos": 30370839039360.0,
      "grad_norm": 1.5872083566348008,
      "language_loss": 0.64167351,
      "learning_rate": 1.7626179388920948e-07,
      "loss": 0.66469979,
      "num_input_tokens_seen": 156427705,
      "step": 7230,
      "time_per_iteration": 3.6919443607330322
    },
    {
      "auxiliary_loss_clip": 0.01271483,
      "auxiliary_loss_mlp": 0.02564464,
      "balance_loss_clip": 1.04727244,
      "balance_loss_mlp": 0.99991107,
      "epoch": 0.8694763422112668,
      "flos": 27200430028800.0,
      "grad_norm": 1.5744389768928813,
      "language_loss": 0.80669606,
      "learning_rate": 1.7594217908329866e-07,
      "loss": 0.84505558,
      "num_input_tokens_seen": 156449890,
      "step": 7231,
      "time_per_iteration": 2.690922260284424
    },
    {
      "auxiliary_loss_clip": 0.0126605,
      "auxiliary_loss_mlp": 0.01021751,
      "balance_loss_clip": 1.04622042,
      "balance_loss_mlp": 1.01503921,
      "epoch": 0.8695965851019059,
      "flos": 26139161767680.0,
      "grad_norm": 2.2527013836820804,
      "language_loss": 0.74184877,
      "learning_rate": 1.7562284098550895e-07,
      "loss": 0.76472676,
      "num_input_tokens_seen": 156469600,
      "step": 7232,
      "time_per_iteration": 3.666689872741699
    },
    {
      "auxiliary_loss_clip": 0.01161939,
      "auxiliary_loss_mlp": 0.01007506,
      "balance_loss_clip": 1.00853276,
      "balance_loss_mlp": 1.00662982,
      "epoch": 0.8697168279925449,
      "flos": 67332616456320.0,
      "grad_norm": 0.8307521206894776,
      "language_loss": 0.62137765,
      "learning_rate": 1.753037796442838e-07,
      "loss": 0.64307213,
      "num_input_tokens_seen": 156529040,
      "step": 7233,
      "time_per_iteration": 3.322922945022583
    },
    {
      "auxiliary_loss_clip": 0.01174568,
      "auxiliary_loss_mlp": 0.01026696,
      "balance_loss_clip": 1.048455,
      "balance_loss_mlp": 1.01903713,
      "epoch": 0.8698370708831841,
      "flos": 19718693337600.0,
      "grad_norm": 2.2346206949545393,
      "language_loss": 0.74853158,
      "learning_rate": 1.74984995108024e-07,
      "loss": 0.77054423,
      "num_input_tokens_seen": 156546970,
      "step": 7234,
      "time_per_iteration": 2.600470781326294
    },
    {
      "auxiliary_loss_clip": 0.0122288,
      "auxiliary_loss_mlp": 0.01025425,
      "balance_loss_clip": 1.0467602,
      "balance_loss_mlp": 1.01835299,
      "epoch": 0.8699573137738231,
      "flos": 12859971068160.0,
      "grad_norm": 4.470354721182958,
      "language_loss": 0.83636534,
      "learning_rate": 1.7466648742508981e-07,
      "loss": 0.85884833,
      "num_input_tokens_seen": 156563155,
      "step": 7235,
      "time_per_iteration": 2.6726980209350586
    },
    {
      "auxiliary_loss_clip": 0.01267972,
      "auxiliary_loss_mlp": 0.01026425,
      "balance_loss_clip": 1.04780841,
      "balance_loss_mlp": 1.01994872,
      "epoch": 0.8700775566644622,
      "flos": 17420733768960.0,
      "grad_norm": 1.844518998965995,
      "language_loss": 0.84796143,
      "learning_rate": 1.7434825664379837e-07,
      "loss": 0.87090534,
      "num_input_tokens_seen": 156581660,
      "step": 7236,
      "time_per_iteration": 3.5627617835998535
    },
    {
      "auxiliary_loss_clip": 0.01220219,
      "auxiliary_loss_mlp": 0.01025963,
      "balance_loss_clip": 1.04654479,
      "balance_loss_mlp": 1.0188638,
      "epoch": 0.8701977995551013,
      "flos": 13735221770880.0,
      "grad_norm": 2.7528237699909837,
      "language_loss": 0.85977274,
      "learning_rate": 1.740303028124246e-07,
      "loss": 0.88223457,
      "num_input_tokens_seen": 156597720,
      "step": 7237,
      "time_per_iteration": 2.554696559906006
    },
    {
      "auxiliary_loss_clip": 0.01422226,
      "auxiliary_loss_mlp": 0.01027109,
      "balance_loss_clip": 1.04100645,
      "balance_loss_mlp": 1.01995635,
      "epoch": 0.8703180424457404,
      "flos": 30555707362560.0,
      "grad_norm": 1.907717532918966,
      "language_loss": 0.75833392,
      "learning_rate": 1.7371262597920212e-07,
      "loss": 0.78282726,
      "num_input_tokens_seen": 156619780,
      "step": 7238,
      "time_per_iteration": 2.8358161449432373
    },
    {
      "auxiliary_loss_clip": 0.01366424,
      "auxiliary_loss_mlp": 0.01027645,
      "balance_loss_clip": 1.04654646,
      "balance_loss_mlp": 1.02074289,
      "epoch": 0.8704382853363795,
      "flos": 19608986223360.0,
      "grad_norm": 1.6763738937739623,
      "language_loss": 0.76380134,
      "learning_rate": 1.7339522619232195e-07,
      "loss": 0.78774202,
      "num_input_tokens_seen": 156638160,
      "step": 7239,
      "time_per_iteration": 2.718743324279785
    },
    {
      "auxiliary_loss_clip": 0.01279483,
      "auxiliary_loss_mlp": 0.01024748,
      "balance_loss_clip": 1.04472804,
      "balance_loss_mlp": 1.01727974,
      "epoch": 0.8705585282270186,
      "flos": 26613900846720.0,
      "grad_norm": 2.95064586235358,
      "language_loss": 0.75605273,
      "learning_rate": 1.730781034999338e-07,
      "loss": 0.77909505,
      "num_input_tokens_seen": 156659740,
      "step": 7240,
      "time_per_iteration": 2.682053804397583
    },
    {
      "auxiliary_loss_clip": 0.01172107,
      "auxiliary_loss_mlp": 0.01024519,
      "balance_loss_clip": 1.05070114,
      "balance_loss_mlp": 1.01748276,
      "epoch": 0.8706787711176577,
      "flos": 34090465979520.0,
      "grad_norm": 1.8093878797064162,
      "language_loss": 0.73487794,
      "learning_rate": 1.7276125795014497e-07,
      "loss": 0.75684422,
      "num_input_tokens_seen": 156678190,
      "step": 7241,
      "time_per_iteration": 2.6612532138824463
    },
    {
      "auxiliary_loss_clip": 0.01275451,
      "auxiliary_loss_mlp": 0.01026607,
      "balance_loss_clip": 1.04297733,
      "balance_loss_mlp": 1.0189476,
      "epoch": 0.8707990140082967,
      "flos": 14611513968000.0,
      "grad_norm": 3.284991836287034,
      "language_loss": 0.67517859,
      "learning_rate": 1.7244468959102054e-07,
      "loss": 0.69819915,
      "num_input_tokens_seen": 156695245,
      "step": 7242,
      "time_per_iteration": 2.6647346019744873
    },
    {
      "auxiliary_loss_clip": 0.01220585,
      "auxiliary_loss_mlp": 0.01028481,
      "balance_loss_clip": 1.04878211,
      "balance_loss_mlp": 1.02161813,
      "epoch": 0.8709192568989359,
      "flos": 20084156265600.0,
      "grad_norm": 7.582020195771247,
      "language_loss": 0.85003275,
      "learning_rate": 1.7212839847058348e-07,
      "loss": 0.87252343,
      "num_input_tokens_seen": 156710375,
      "step": 7243,
      "time_per_iteration": 2.600057363510132
    },
    {
      "auxiliary_loss_clip": 0.01328111,
      "auxiliary_loss_mlp": 0.0102046,
      "balance_loss_clip": 1.04162705,
      "balance_loss_mlp": 1.01402545,
      "epoch": 0.871039499789575,
      "flos": 16727083251840.0,
      "grad_norm": 1.7994326438096948,
      "language_loss": 0.74149191,
      "learning_rate": 1.718123846368147e-07,
      "loss": 0.76497757,
      "num_input_tokens_seen": 156729420,
      "step": 7244,
      "time_per_iteration": 2.7846667766571045
    },
    {
      "auxiliary_loss_clip": 0.01269765,
      "auxiliary_loss_mlp": 0.02563209,
      "balance_loss_clip": 1.04663563,
      "balance_loss_mlp": 0.99987864,
      "epoch": 0.871159742680214,
      "flos": 21068790860160.0,
      "grad_norm": 1.9345704187677966,
      "language_loss": 0.71840501,
      "learning_rate": 1.714966481376543e-07,
      "loss": 0.75673473,
      "num_input_tokens_seen": 156746100,
      "step": 7245,
      "time_per_iteration": 2.6568706035614014
    },
    {
      "auxiliary_loss_clip": 0.01220444,
      "auxiliary_loss_mlp": 0.01024039,
      "balance_loss_clip": 1.0465827,
      "balance_loss_mlp": 1.01723182,
      "epoch": 0.8712799855708532,
      "flos": 28256526731520.0,
      "grad_norm": 1.9553927113951548,
      "language_loss": 0.83097583,
      "learning_rate": 1.7118118902099797e-07,
      "loss": 0.85342056,
      "num_input_tokens_seen": 156764185,
      "step": 7246,
      "time_per_iteration": 2.6947813034057617
    },
    {
      "auxiliary_loss_clip": 0.01220809,
      "auxiliary_loss_mlp": 0.01027999,
      "balance_loss_clip": 1.04616344,
      "balance_loss_mlp": 1.02155542,
      "epoch": 0.8714002284614922,
      "flos": 22236677665920.0,
      "grad_norm": 2.060335975416442,
      "language_loss": 0.80804443,
      "learning_rate": 1.7086600733470146e-07,
      "loss": 0.83053255,
      "num_input_tokens_seen": 156784855,
      "step": 7247,
      "time_per_iteration": 2.6640751361846924
    },
    {
      "auxiliary_loss_clip": 0.01215002,
      "auxiliary_loss_mlp": 0.01026253,
      "balance_loss_clip": 1.04468143,
      "balance_loss_mlp": 1.01910377,
      "epoch": 0.8715204713521313,
      "flos": 21431919404160.0,
      "grad_norm": 1.7925051296795547,
      "language_loss": 0.77320999,
      "learning_rate": 1.7055110312657738e-07,
      "loss": 0.79562253,
      "num_input_tokens_seen": 156804350,
      "step": 7248,
      "time_per_iteration": 2.615983724594116
    },
    {
      "auxiliary_loss_clip": 0.01268852,
      "auxiliary_loss_mlp": 0.01028709,
      "balance_loss_clip": 1.04446793,
      "balance_loss_mlp": 1.0211091,
      "epoch": 0.8716407142427703,
      "flos": 23440439180160.0,
      "grad_norm": 2.0088631598205082,
      "language_loss": 0.74551201,
      "learning_rate": 1.702364764443962e-07,
      "loss": 0.76848763,
      "num_input_tokens_seen": 156823425,
      "step": 7249,
      "time_per_iteration": 2.6625828742980957
    },
    {
      "auxiliary_loss_clip": 0.01417326,
      "auxiliary_loss_mlp": 0.01027873,
      "balance_loss_clip": 1.03928089,
      "balance_loss_mlp": 1.02108455,
      "epoch": 0.8717609571334095,
      "flos": 27958683156480.0,
      "grad_norm": 1.8395805246892423,
      "language_loss": 0.72538871,
      "learning_rate": 1.6992212733588685e-07,
      "loss": 0.74984074,
      "num_input_tokens_seen": 156843090,
      "step": 7250,
      "time_per_iteration": 2.8678717613220215
    },
    {
      "auxiliary_loss_clip": 0.01270488,
      "auxiliary_loss_mlp": 0.01023054,
      "balance_loss_clip": 1.0447228,
      "balance_loss_mlp": 1.01644039,
      "epoch": 0.8718812000240486,
      "flos": 25479482538240.0,
      "grad_norm": 1.8938928233976158,
      "language_loss": 0.75003284,
      "learning_rate": 1.6960805584873538e-07,
      "loss": 0.77296823,
      "num_input_tokens_seen": 156861090,
      "step": 7251,
      "time_per_iteration": 2.756009340286255
    },
    {
      "auxiliary_loss_clip": 0.01369032,
      "auxiliary_loss_mlp": 0.01023578,
      "balance_loss_clip": 1.0425061,
      "balance_loss_mlp": 1.01686668,
      "epoch": 0.8720014429146876,
      "flos": 23403056100480.0,
      "grad_norm": 3.8429575705826595,
      "language_loss": 0.78364843,
      "learning_rate": 1.6929426203058684e-07,
      "loss": 0.80757457,
      "num_input_tokens_seen": 156881515,
      "step": 7252,
      "time_per_iteration": 2.7248144149780273
    },
    {
      "auxiliary_loss_clip": 0.01176766,
      "auxiliary_loss_mlp": 0.02571456,
      "balance_loss_clip": 1.04851806,
      "balance_loss_mlp": 0.999915,
      "epoch": 0.8721216858053268,
      "flos": 24352821567360.0,
      "grad_norm": 2.4386539138653895,
      "language_loss": 0.8014226,
      "learning_rate": 1.689807459290431e-07,
      "loss": 0.8389048,
      "num_input_tokens_seen": 156900170,
      "step": 7253,
      "time_per_iteration": 2.6701109409332275
    },
    {
      "auxiliary_loss_clip": 0.01273262,
      "auxiliary_loss_mlp": 0.01024103,
      "balance_loss_clip": 1.04769421,
      "balance_loss_mlp": 1.01726305,
      "epoch": 0.8722419286959658,
      "flos": 33869687034240.0,
      "grad_norm": 1.9298300059642002,
      "language_loss": 0.70986384,
      "learning_rate": 1.6866750759166437e-07,
      "loss": 0.73283744,
      "num_input_tokens_seen": 156920150,
      "step": 7254,
      "time_per_iteration": 2.7512967586517334
    },
    {
      "auxiliary_loss_clip": 0.0131686,
      "auxiliary_loss_mlp": 0.0102531,
      "balance_loss_clip": 1.03945029,
      "balance_loss_mlp": 1.018116,
      "epoch": 0.8723621715866049,
      "flos": 18369385914240.0,
      "grad_norm": 2.750076820587762,
      "language_loss": 0.77331686,
      "learning_rate": 1.6835454706596865e-07,
      "loss": 0.79673856,
      "num_input_tokens_seen": 156937980,
      "step": 7255,
      "time_per_iteration": 2.6963298320770264
    },
    {
      "auxiliary_loss_clip": 0.01174591,
      "auxiliary_loss_mlp": 0.01031102,
      "balance_loss_clip": 1.0504607,
      "balance_loss_mlp": 1.02398872,
      "epoch": 0.8724824144772441,
      "flos": 22013348855040.0,
      "grad_norm": 2.416825719507602,
      "language_loss": 0.73906958,
      "learning_rate": 1.680418643994317e-07,
      "loss": 0.76112658,
      "num_input_tokens_seen": 156956550,
      "step": 7256,
      "time_per_iteration": 4.452859878540039
    },
    {
      "auxiliary_loss_clip": 0.01060727,
      "auxiliary_loss_mlp": 0.00999338,
      "balance_loss_clip": 1.00648403,
      "balance_loss_mlp": 0.99850982,
      "epoch": 0.8726026573678831,
      "flos": 66698720213760.0,
      "grad_norm": 0.8828002334620544,
      "language_loss": 0.64503884,
      "learning_rate": 1.6772945963948738e-07,
      "loss": 0.66563952,
      "num_input_tokens_seen": 157014715,
      "step": 7257,
      "time_per_iteration": 3.2035293579101562
    },
    {
      "auxiliary_loss_clip": 0.01270502,
      "auxiliary_loss_mlp": 0.01023286,
      "balance_loss_clip": 1.04788411,
      "balance_loss_mlp": 1.01599002,
      "epoch": 0.8727229002585222,
      "flos": 13370908078080.0,
      "grad_norm": 2.1885039435933886,
      "language_loss": 0.77242756,
      "learning_rate": 1.6741733283352733e-07,
      "loss": 0.79536545,
      "num_input_tokens_seen": 157032320,
      "step": 7258,
      "time_per_iteration": 3.5899226665496826
    },
    {
      "auxiliary_loss_clip": 0.01270205,
      "auxiliary_loss_mlp": 0.01024762,
      "balance_loss_clip": 1.04418921,
      "balance_loss_mlp": 1.01760912,
      "epoch": 0.8728431431491613,
      "flos": 21796987282560.0,
      "grad_norm": 1.7900337572931107,
      "language_loss": 0.83853173,
      "learning_rate": 1.6710548402890102e-07,
      "loss": 0.86148143,
      "num_input_tokens_seen": 157052845,
      "step": 7259,
      "time_per_iteration": 2.739159107208252
    },
    {
      "auxiliary_loss_clip": 0.01176097,
      "auxiliary_loss_mlp": 0.01022599,
      "balance_loss_clip": 1.04804611,
      "balance_loss_mlp": 1.01551461,
      "epoch": 0.8729633860398004,
      "flos": 36173823742080.0,
      "grad_norm": 2.223322150686324,
      "language_loss": 0.66612101,
      "learning_rate": 1.6679391327291527e-07,
      "loss": 0.68810797,
      "num_input_tokens_seen": 157074050,
      "step": 7260,
      "time_per_iteration": 2.678316116333008
    },
    {
      "auxiliary_loss_clip": 0.0127332,
      "auxiliary_loss_mlp": 0.01025121,
      "balance_loss_clip": 1.04380715,
      "balance_loss_mlp": 1.01894307,
      "epoch": 0.8730836289304394,
      "flos": 16359680989440.0,
      "grad_norm": 2.4735930975529414,
      "language_loss": 0.68422478,
      "learning_rate": 1.6648262061283492e-07,
      "loss": 0.70720917,
      "num_input_tokens_seen": 157089350,
      "step": 7261,
      "time_per_iteration": 2.6932668685913086
    },
    {
      "auxiliary_loss_clip": 0.01319276,
      "auxiliary_loss_mlp": 0.01025382,
      "balance_loss_clip": 1.04093575,
      "balance_loss_mlp": 1.018399,
      "epoch": 0.8732038718210786,
      "flos": 21215126868480.0,
      "grad_norm": 1.9078543599500044,
      "language_loss": 0.73559558,
      "learning_rate": 1.6617160609588353e-07,
      "loss": 0.75904226,
      "num_input_tokens_seen": 157108525,
      "step": 7262,
      "time_per_iteration": 3.6529572010040283
    },
    {
      "auxiliary_loss_clip": 0.01171702,
      "auxiliary_loss_mlp": 0.0102869,
      "balance_loss_clip": 1.04551888,
      "balance_loss_mlp": 1.02229691,
      "epoch": 0.8733241147117177,
      "flos": 16610696208000.0,
      "grad_norm": 2.1562650723897008,
      "language_loss": 0.7191788,
      "learning_rate": 1.6586086976924163e-07,
      "loss": 0.74118268,
      "num_input_tokens_seen": 157124025,
      "step": 7263,
      "time_per_iteration": 2.6637539863586426
    },
    {
      "auxiliary_loss_clip": 0.01222308,
      "auxiliary_loss_mlp": 0.01027989,
      "balance_loss_clip": 1.04593682,
      "balance_loss_mlp": 1.02113748,
      "epoch": 0.8734443576023567,
      "flos": 20193935207040.0,
      "grad_norm": 1.8215966790633111,
      "language_loss": 0.7845965,
      "learning_rate": 1.6555041168004747e-07,
      "loss": 0.80709946,
      "num_input_tokens_seen": 157143345,
      "step": 7264,
      "time_per_iteration": 2.5863094329833984
    },
    {
      "auxiliary_loss_clip": 0.01267126,
      "auxiliary_loss_mlp": 0.01024525,
      "balance_loss_clip": 1.04456401,
      "balance_loss_mlp": 1.01807904,
      "epoch": 0.8735646004929959,
      "flos": 18041162411520.0,
      "grad_norm": 1.9797178599614467,
      "language_loss": 0.69295871,
      "learning_rate": 1.6524023187539715e-07,
      "loss": 0.71587515,
      "num_input_tokens_seen": 157161630,
      "step": 7265,
      "time_per_iteration": 2.67750883102417
    },
    {
      "auxiliary_loss_clip": 0.01273386,
      "auxiliary_loss_mlp": 0.01027046,
      "balance_loss_clip": 1.04630828,
      "balance_loss_mlp": 1.01991165,
      "epoch": 0.873684843383635,
      "flos": 20262344659200.0,
      "grad_norm": 1.7738626872033276,
      "language_loss": 0.74653912,
      "learning_rate": 1.649303304023446e-07,
      "loss": 0.76954353,
      "num_input_tokens_seen": 157181385,
      "step": 7266,
      "time_per_iteration": 2.6677963733673096
    },
    {
      "auxiliary_loss_clip": 0.01312236,
      "auxiliary_loss_mlp": 0.0102937,
      "balance_loss_clip": 1.04609776,
      "balance_loss_mlp": 1.02278078,
      "epoch": 0.873805086274274,
      "flos": 16947287579520.0,
      "grad_norm": 1.8545058978768867,
      "language_loss": 0.7872479,
      "learning_rate": 1.6462070730790246e-07,
      "loss": 0.81066394,
      "num_input_tokens_seen": 157200545,
      "step": 7267,
      "time_per_iteration": 2.6584227085113525
    },
    {
      "auxiliary_loss_clip": 0.01266409,
      "auxiliary_loss_mlp": 0.01023359,
      "balance_loss_clip": 1.04095614,
      "balance_loss_mlp": 1.01631987,
      "epoch": 0.8739253291649132,
      "flos": 18041270152320.0,
      "grad_norm": 3.0174759446470403,
      "language_loss": 0.7881223,
      "learning_rate": 1.6431136263903912e-07,
      "loss": 0.81101996,
      "num_input_tokens_seen": 157219545,
      "step": 7268,
      "time_per_iteration": 2.600555658340454
    },
    {
      "auxiliary_loss_clip": 0.01224922,
      "auxiliary_loss_mlp": 0.0256625,
      "balance_loss_clip": 1.04473019,
      "balance_loss_mlp": 0.99993038,
      "epoch": 0.8740455720555522,
      "flos": 21325085377920.0,
      "grad_norm": 1.942263662819865,
      "language_loss": 0.73599565,
      "learning_rate": 1.6400229644268282e-07,
      "loss": 0.7739073,
      "num_input_tokens_seen": 157237900,
      "step": 7269,
      "time_per_iteration": 2.6496424674987793
    },
    {
      "auxiliary_loss_clip": 0.01322431,
      "auxiliary_loss_mlp": 0.01027986,
      "balance_loss_clip": 1.05026722,
      "balance_loss_mlp": 1.02079141,
      "epoch": 0.8741658149461913,
      "flos": 15158684822400.0,
      "grad_norm": 2.0584340386345983,
      "language_loss": 0.81325531,
      "learning_rate": 1.6369350876571852e-07,
      "loss": 0.83675945,
      "num_input_tokens_seen": 157256055,
      "step": 7270,
      "time_per_iteration": 2.67555570602417
    },
    {
      "auxiliary_loss_clip": 0.01369423,
      "auxiliary_loss_mlp": 0.01026595,
      "balance_loss_clip": 1.0412581,
      "balance_loss_mlp": 1.02013397,
      "epoch": 0.8742860578368304,
      "flos": 23039855729280.0,
      "grad_norm": 4.557980655882612,
      "language_loss": 0.8191613,
      "learning_rate": 1.6338499965498874e-07,
      "loss": 0.84312147,
      "num_input_tokens_seen": 157274785,
      "step": 7271,
      "time_per_iteration": 2.7363264560699463
    },
    {
      "auxiliary_loss_clip": 0.0132556,
      "auxiliary_loss_mlp": 0.01024451,
      "balance_loss_clip": 1.04538369,
      "balance_loss_mlp": 1.01753128,
      "epoch": 0.8744063007274695,
      "flos": 28145347159680.0,
      "grad_norm": 1.4632771711812924,
      "language_loss": 0.77353054,
      "learning_rate": 1.630767691572943e-07,
      "loss": 0.79703069,
      "num_input_tokens_seen": 157294805,
      "step": 7272,
      "time_per_iteration": 2.80120587348938
    },
    {
      "auxiliary_loss_clip": 0.01171166,
      "auxiliary_loss_mlp": 0.01001065,
      "balance_loss_clip": 1.00675893,
      "balance_loss_mlp": 1.00014675,
      "epoch": 0.8745265436181086,
      "flos": 64034076654720.0,
      "grad_norm": 0.8339325195841324,
      "language_loss": 0.5341382,
      "learning_rate": 1.6276881731939306e-07,
      "loss": 0.55586052,
      "num_input_tokens_seen": 157356695,
      "step": 7273,
      "time_per_iteration": 3.2606029510498047
    },
    {
      "auxiliary_loss_clip": 0.01218223,
      "auxiliary_loss_mlp": 0.01025843,
      "balance_loss_clip": 1.04792285,
      "balance_loss_mlp": 1.01883054,
      "epoch": 0.8746467865087477,
      "flos": 28658618553600.0,
      "grad_norm": 1.8224837514830876,
      "language_loss": 0.7533682,
      "learning_rate": 1.6246114418800193e-07,
      "loss": 0.77580881,
      "num_input_tokens_seen": 157376975,
      "step": 7274,
      "time_per_iteration": 2.6761975288391113
    },
    {
      "auxiliary_loss_clip": 0.01221452,
      "auxiliary_loss_mlp": 0.01026175,
      "balance_loss_clip": 1.04632854,
      "balance_loss_mlp": 1.01874197,
      "epoch": 0.8747670293993868,
      "flos": 23985850268160.0,
      "grad_norm": 1.9455384201456611,
      "language_loss": 0.76626617,
      "learning_rate": 1.6215374980979423e-07,
      "loss": 0.78874248,
      "num_input_tokens_seen": 157397385,
      "step": 7275,
      "time_per_iteration": 2.6318647861480713
    },
    {
      "auxiliary_loss_clip": 0.01218759,
      "auxiliary_loss_mlp": 0.01027233,
      "balance_loss_clip": 1.0484128,
      "balance_loss_mlp": 1.02052116,
      "epoch": 0.8748872722900258,
      "flos": 45221624478720.0,
      "grad_norm": 1.9276114521344512,
      "language_loss": 0.68974984,
      "learning_rate": 1.6184663423140133e-07,
      "loss": 0.71220976,
      "num_input_tokens_seen": 157417685,
      "step": 7276,
      "time_per_iteration": 2.848367929458618
    },
    {
      "auxiliary_loss_clip": 0.01374198,
      "auxiliary_loss_mlp": 0.01027523,
      "balance_loss_clip": 1.04404128,
      "balance_loss_mlp": 1.02061749,
      "epoch": 0.875007515180665,
      "flos": 19754280737280.0,
      "grad_norm": 1.848551771749569,
      "language_loss": 0.63806665,
      "learning_rate": 1.615397974994126e-07,
      "loss": 0.66208386,
      "num_input_tokens_seen": 157435490,
      "step": 7277,
      "time_per_iteration": 2.7555747032165527
    },
    {
      "auxiliary_loss_clip": 0.01169141,
      "auxiliary_loss_mlp": 0.01024989,
      "balance_loss_clip": 1.04734969,
      "balance_loss_mlp": 1.01846516,
      "epoch": 0.875127758071304,
      "flos": 22710734386560.0,
      "grad_norm": 1.6068096034591661,
      "language_loss": 0.80852932,
      "learning_rate": 1.6123323966037438e-07,
      "loss": 0.83047062,
      "num_input_tokens_seen": 157454010,
      "step": 7278,
      "time_per_iteration": 2.6317808628082275
    },
    {
      "auxiliary_loss_clip": 0.01174966,
      "auxiliary_loss_mlp": 0.01027775,
      "balance_loss_clip": 1.05132401,
      "balance_loss_mlp": 1.02097452,
      "epoch": 0.8752480009619431,
      "flos": 23403846199680.0,
      "grad_norm": 3.3867422334282935,
      "language_loss": 0.78866166,
      "learning_rate": 1.6092696076079216e-07,
      "loss": 0.81068909,
      "num_input_tokens_seen": 157472385,
      "step": 7279,
      "time_per_iteration": 2.5973289012908936
    },
    {
      "auxiliary_loss_clip": 0.01313119,
      "auxiliary_loss_mlp": 0.01022257,
      "balance_loss_clip": 1.04460979,
      "balance_loss_mlp": 1.01555157,
      "epoch": 0.8753682438525822,
      "flos": 26213101914240.0,
      "grad_norm": 1.7481174452637855,
      "language_loss": 0.73850811,
      "learning_rate": 1.6062096084712785e-07,
      "loss": 0.76186192,
      "num_input_tokens_seen": 157493735,
      "step": 7280,
      "time_per_iteration": 2.6817333698272705
    },
    {
      "auxiliary_loss_clip": 0.01268967,
      "auxiliary_loss_mlp": 0.02567918,
      "balance_loss_clip": 1.04264426,
      "balance_loss_mlp": 0.99990058,
      "epoch": 0.8754884867432213,
      "flos": 23326745656320.0,
      "grad_norm": 7.115094618798673,
      "language_loss": 0.70583993,
      "learning_rate": 1.6031523996580098e-07,
      "loss": 0.74420881,
      "num_input_tokens_seen": 157511295,
      "step": 7281,
      "time_per_iteration": 2.688387870788574
    },
    {
      "auxiliary_loss_clip": 0.01329205,
      "auxiliary_loss_mlp": 0.01023538,
      "balance_loss_clip": 1.04601002,
      "balance_loss_mlp": 1.0159204,
      "epoch": 0.8756087296338604,
      "flos": 12495226412160.0,
      "grad_norm": 2.2849786327536568,
      "language_loss": 0.66366887,
      "learning_rate": 1.6000979816318981e-07,
      "loss": 0.68719625,
      "num_input_tokens_seen": 157529760,
      "step": 7282,
      "time_per_iteration": 4.469428062438965
    },
    {
      "auxiliary_loss_clip": 0.01218822,
      "auxiliary_loss_mlp": 0.01024163,
      "balance_loss_clip": 1.04780579,
      "balance_loss_mlp": 1.01679552,
      "epoch": 0.8757289725244994,
      "flos": 18952898353920.0,
      "grad_norm": 3.1100181118712733,
      "language_loss": 0.75120068,
      "learning_rate": 1.5970463548562886e-07,
      "loss": 0.7736305,
      "num_input_tokens_seen": 157548915,
      "step": 7283,
      "time_per_iteration": 2.605156183242798
    },
    {
      "auxiliary_loss_clip": 0.01267633,
      "auxiliary_loss_mlp": 0.01020066,
      "balance_loss_clip": 1.04705477,
      "balance_loss_mlp": 1.01335168,
      "epoch": 0.8758492154151386,
      "flos": 25265958140160.0,
      "grad_norm": 1.7796112309965653,
      "language_loss": 0.71193957,
      "learning_rate": 1.5939975197941192e-07,
      "loss": 0.73481655,
      "num_input_tokens_seen": 157570570,
      "step": 7284,
      "time_per_iteration": 2.7326533794403076
    },
    {
      "auxiliary_loss_clip": 0.01169671,
      "auxiliary_loss_mlp": 0.01003299,
      "balance_loss_clip": 1.00656867,
      "balance_loss_mlp": 1.00248861,
      "epoch": 0.8759694583057777,
      "flos": 65571664193280.0,
      "grad_norm": 0.8133505225788741,
      "language_loss": 0.53286684,
      "learning_rate": 1.5909514769078892e-07,
      "loss": 0.55459654,
      "num_input_tokens_seen": 157635675,
      "step": 7285,
      "time_per_iteration": 4.215396404266357
    },
    {
      "auxiliary_loss_clip": 0.01321804,
      "auxiliary_loss_mlp": 0.01026084,
      "balance_loss_clip": 1.05024314,
      "balance_loss_mlp": 1.01937866,
      "epoch": 0.8760897011964167,
      "flos": 25446193608960.0,
      "grad_norm": 2.006905225073446,
      "language_loss": 0.77939236,
      "learning_rate": 1.5879082266596867e-07,
      "loss": 0.80287123,
      "num_input_tokens_seen": 157657015,
      "step": 7286,
      "time_per_iteration": 2.726262092590332
    },
    {
      "auxiliary_loss_clip": 0.01267166,
      "auxiliary_loss_mlp": 0.01025139,
      "balance_loss_clip": 1.04186678,
      "balance_loss_mlp": 1.01859713,
      "epoch": 0.8762099440870559,
      "flos": 28984830894720.0,
      "grad_norm": 1.8141115377924155,
      "language_loss": 0.71769762,
      "learning_rate": 1.5848677695111645e-07,
      "loss": 0.74062061,
      "num_input_tokens_seen": 157678615,
      "step": 7287,
      "time_per_iteration": 2.756207227706909
    },
    {
      "auxiliary_loss_clip": 0.01330101,
      "auxiliary_loss_mlp": 0.01024369,
      "balance_loss_clip": 1.04833555,
      "balance_loss_mlp": 1.01685548,
      "epoch": 0.8763301869776949,
      "flos": 21609461352960.0,
      "grad_norm": 2.413527128464261,
      "language_loss": 0.69531393,
      "learning_rate": 1.5818301059235562e-07,
      "loss": 0.71885866,
      "num_input_tokens_seen": 157693790,
      "step": 7288,
      "time_per_iteration": 3.6210267543792725
    },
    {
      "auxiliary_loss_clip": 0.01273891,
      "auxiliary_loss_mlp": 0.01024387,
      "balance_loss_clip": 1.04801297,
      "balance_loss_mlp": 1.01759839,
      "epoch": 0.876450429868334,
      "flos": 24644416176000.0,
      "grad_norm": 2.023849907593808,
      "language_loss": 0.81233525,
      "learning_rate": 1.578795236357684e-07,
      "loss": 0.83531803,
      "num_input_tokens_seen": 157715255,
      "step": 7289,
      "time_per_iteration": 2.7239997386932373
    },
    {
      "auxiliary_loss_clip": 0.01271581,
      "auxiliary_loss_mlp": 0.01026746,
      "balance_loss_clip": 1.04691005,
      "balance_loss_mlp": 1.01974201,
      "epoch": 0.8765706727589732,
      "flos": 20260046188800.0,
      "grad_norm": 2.0968628138203234,
      "language_loss": 0.85690475,
      "learning_rate": 1.5757631612739218e-07,
      "loss": 0.879888,
      "num_input_tokens_seen": 157728800,
      "step": 7290,
      "time_per_iteration": 2.617967128753662
    },
    {
      "auxiliary_loss_clip": 0.0106069,
      "auxiliary_loss_mlp": 0.01000348,
      "balance_loss_clip": 1.00632918,
      "balance_loss_mlp": 0.99949598,
      "epoch": 0.8766909156496122,
      "flos": 71371165276800.0,
      "grad_norm": 0.7783074441011061,
      "language_loss": 0.61382759,
      "learning_rate": 1.572733881132242e-07,
      "loss": 0.63443798,
      "num_input_tokens_seen": 157789445,
      "step": 7291,
      "time_per_iteration": 3.2222490310668945
    },
    {
      "auxiliary_loss_clip": 0.01216268,
      "auxiliary_loss_mlp": 0.01008522,
      "balance_loss_clip": 1.01112175,
      "balance_loss_mlp": 1.00766373,
      "epoch": 0.8768111585402513,
      "flos": 69523490603520.0,
      "grad_norm": 0.7827386440960511,
      "language_loss": 0.58476287,
      "learning_rate": 1.5697073963921814e-07,
      "loss": 0.60701078,
      "num_input_tokens_seen": 157848685,
      "step": 7292,
      "time_per_iteration": 3.1866157054901123
    },
    {
      "auxiliary_loss_clip": 0.01224936,
      "auxiliary_loss_mlp": 0.01028037,
      "balance_loss_clip": 1.0495975,
      "balance_loss_mlp": 1.02033305,
      "epoch": 0.8769314014308904,
      "flos": 18838558385280.0,
      "grad_norm": 2.450665244400367,
      "language_loss": 0.84910476,
      "learning_rate": 1.566683707512857e-07,
      "loss": 0.87163448,
      "num_input_tokens_seen": 157866360,
      "step": 7293,
      "time_per_iteration": 2.6184444427490234
    },
    {
      "auxiliary_loss_clip": 0.01269196,
      "auxiliary_loss_mlp": 0.01030354,
      "balance_loss_clip": 1.04463887,
      "balance_loss_mlp": 1.02333879,
      "epoch": 0.8770516443215295,
      "flos": 14976402278400.0,
      "grad_norm": 2.4457691024074744,
      "language_loss": 0.79830199,
      "learning_rate": 1.5636628149529553e-07,
      "loss": 0.82129753,
      "num_input_tokens_seen": 157884150,
      "step": 7294,
      "time_per_iteration": 2.6317007541656494
    },
    {
      "auxiliary_loss_clip": 0.01268308,
      "auxiliary_loss_mlp": 0.01027283,
      "balance_loss_clip": 1.04313982,
      "balance_loss_mlp": 1.02057147,
      "epoch": 0.8771718872121685,
      "flos": 31649654021760.0,
      "grad_norm": 2.2082863910193744,
      "language_loss": 0.79929906,
      "learning_rate": 1.560644719170743e-07,
      "loss": 0.82225502,
      "num_input_tokens_seen": 157905020,
      "step": 7295,
      "time_per_iteration": 2.726923704147339
    },
    {
      "auxiliary_loss_clip": 0.01321973,
      "auxiliary_loss_mlp": 0.01024219,
      "balance_loss_clip": 1.04176927,
      "balance_loss_mlp": 1.01692641,
      "epoch": 0.8772921301028077,
      "flos": 36095466222720.0,
      "grad_norm": 1.9077670558609257,
      "language_loss": 0.72302043,
      "learning_rate": 1.5576294206240692e-07,
      "loss": 0.74648237,
      "num_input_tokens_seen": 157924545,
      "step": 7296,
      "time_per_iteration": 2.856168508529663
    },
    {
      "auxiliary_loss_clip": 0.0127575,
      "auxiliary_loss_mlp": 0.0102355,
      "balance_loss_clip": 1.04772949,
      "balance_loss_mlp": 1.01693642,
      "epoch": 0.8774123729934468,
      "flos": 57116961849600.0,
      "grad_norm": 1.8683403877625862,
      "language_loss": 0.67743069,
      "learning_rate": 1.5546169197703507e-07,
      "loss": 0.70042366,
      "num_input_tokens_seen": 157950820,
      "step": 7297,
      "time_per_iteration": 2.9696688652038574
    },
    {
      "auxiliary_loss_clip": 0.01276347,
      "auxiliary_loss_mlp": 0.01028492,
      "balance_loss_clip": 1.04461884,
      "balance_loss_mlp": 1.02187634,
      "epoch": 0.8775326158840858,
      "flos": 23914495900800.0,
      "grad_norm": 2.4038268669269818,
      "language_loss": 0.77404141,
      "learning_rate": 1.5516072170665774e-07,
      "loss": 0.79708982,
      "num_input_tokens_seen": 157968790,
      "step": 7298,
      "time_per_iteration": 2.7235536575317383
    },
    {
      "auxiliary_loss_clip": 0.01224766,
      "auxiliary_loss_mlp": 0.01025008,
      "balance_loss_clip": 1.04871821,
      "balance_loss_mlp": 1.01829374,
      "epoch": 0.877652858774725,
      "flos": 17123285243520.0,
      "grad_norm": 1.9333151642873636,
      "language_loss": 0.87116337,
      "learning_rate": 1.5486003129693214e-07,
      "loss": 0.89366102,
      "num_input_tokens_seen": 157986155,
      "step": 7299,
      "time_per_iteration": 2.5912039279937744
    },
    {
      "auxiliary_loss_clip": 0.01223457,
      "auxiliary_loss_mlp": 0.010222,
      "balance_loss_clip": 1.0472039,
      "balance_loss_mlp": 1.0151639,
      "epoch": 0.877773101665364,
      "flos": 16508961912960.0,
      "grad_norm": 2.243533712588285,
      "language_loss": 0.7809909,
      "learning_rate": 1.545596207934725e-07,
      "loss": 0.80344748,
      "num_input_tokens_seen": 158004640,
      "step": 7300,
      "time_per_iteration": 2.6126456260681152
    },
    {
      "auxiliary_loss_clip": 0.01270363,
      "auxiliary_loss_mlp": 0.01025253,
      "balance_loss_clip": 1.04544926,
      "balance_loss_mlp": 1.01905704,
      "epoch": 0.8778933445560031,
      "flos": 22053209973120.0,
      "grad_norm": 2.0104849594721306,
      "language_loss": 0.77675992,
      "learning_rate": 1.5425949024185147e-07,
      "loss": 0.79971606,
      "num_input_tokens_seen": 158024665,
      "step": 7301,
      "time_per_iteration": 2.7162909507751465
    },
    {
      "auxiliary_loss_clip": 0.01272232,
      "auxiliary_loss_mlp": 0.01024449,
      "balance_loss_clip": 1.04381895,
      "balance_loss_mlp": 1.01789522,
      "epoch": 0.8780135874466423,
      "flos": 22564757514240.0,
      "grad_norm": 1.8750784925296298,
      "language_loss": 0.67638093,
      "learning_rate": 1.5395963968759818e-07,
      "loss": 0.69934773,
      "num_input_tokens_seen": 158044940,
      "step": 7302,
      "time_per_iteration": 2.6975655555725098
    },
    {
      "auxiliary_loss_clip": 0.01266572,
      "auxiliary_loss_mlp": 0.0102531,
      "balance_loss_clip": 1.04156983,
      "balance_loss_mlp": 1.01907516,
      "epoch": 0.8781338303372813,
      "flos": 61531999073280.0,
      "grad_norm": 1.4673151012914587,
      "language_loss": 0.64401066,
      "learning_rate": 1.536600691761998e-07,
      "loss": 0.66692936,
      "num_input_tokens_seen": 158070770,
      "step": 7303,
      "time_per_iteration": 3.0682790279388428
    },
    {
      "auxiliary_loss_clip": 0.01322387,
      "auxiliary_loss_mlp": 0.01023033,
      "balance_loss_clip": 1.0462842,
      "balance_loss_mlp": 1.0163691,
      "epoch": 0.8782540732279204,
      "flos": 22674751937280.0,
      "grad_norm": 2.7171434043088065,
      "language_loss": 0.7171151,
      "learning_rate": 1.5336077875310084e-07,
      "loss": 0.74056929,
      "num_input_tokens_seen": 158089995,
      "step": 7304,
      "time_per_iteration": 2.7175538539886475
    },
    {
      "auxiliary_loss_clip": 0.01371832,
      "auxiliary_loss_mlp": 0.01026543,
      "balance_loss_clip": 1.04283142,
      "balance_loss_mlp": 1.019912,
      "epoch": 0.8783743161185595,
      "flos": 16070348937600.0,
      "grad_norm": 3.059271401542427,
      "language_loss": 0.74304777,
      "learning_rate": 1.5306176846370321e-07,
      "loss": 0.76703155,
      "num_input_tokens_seen": 158108140,
      "step": 7305,
      "time_per_iteration": 2.7502636909484863
    },
    {
      "auxiliary_loss_clip": 0.01277572,
      "auxiliary_loss_mlp": 0.01026497,
      "balance_loss_clip": 1.04445934,
      "balance_loss_mlp": 1.01932108,
      "epoch": 0.8784945590091986,
      "flos": 26067879227520.0,
      "grad_norm": 2.236616860863976,
      "language_loss": 0.7415821,
      "learning_rate": 1.5276303835336712e-07,
      "loss": 0.76462281,
      "num_input_tokens_seen": 158128680,
      "step": 7306,
      "time_per_iteration": 2.6747007369995117
    },
    {
      "auxiliary_loss_clip": 0.01116264,
      "auxiliary_loss_mlp": 0.0100159,
      "balance_loss_clip": 1.00652575,
      "balance_loss_mlp": 1.00076103,
      "epoch": 0.8786148018998376,
      "flos": 62720643939840.0,
      "grad_norm": 0.757177431661679,
      "language_loss": 0.5345735,
      "learning_rate": 1.524645884674094e-07,
      "loss": 0.55575204,
      "num_input_tokens_seen": 158185610,
      "step": 7307,
      "time_per_iteration": 3.170609474182129
    },
    {
      "auxiliary_loss_clip": 0.01174657,
      "auxiliary_loss_mlp": 0.02571,
      "balance_loss_clip": 1.04856467,
      "balance_loss_mlp": 0.99989963,
      "epoch": 0.8787350447904768,
      "flos": 21652734263040.0,
      "grad_norm": 2.416323688407495,
      "language_loss": 0.79243779,
      "learning_rate": 1.521664188511047e-07,
      "loss": 0.8298943,
      "num_input_tokens_seen": 158205635,
      "step": 7308,
      "time_per_iteration": 4.4936747550964355
    },
    {
      "auxiliary_loss_clip": 0.01279163,
      "auxiliary_loss_mlp": 0.02563836,
      "balance_loss_clip": 1.05235004,
      "balance_loss_mlp": 0.99990362,
      "epoch": 0.8788552876811159,
      "flos": 25478476957440.0,
      "grad_norm": 2.235931855373344,
      "language_loss": 0.80042064,
      "learning_rate": 1.518685295496851e-07,
      "loss": 0.83885062,
      "num_input_tokens_seen": 158223495,
      "step": 7309,
      "time_per_iteration": 2.7380194664001465
    },
    {
      "auxiliary_loss_clip": 0.0122243,
      "auxiliary_loss_mlp": 0.01023663,
      "balance_loss_clip": 1.04631281,
      "balance_loss_mlp": 1.01711893,
      "epoch": 0.8789755305717549,
      "flos": 22310222762880.0,
      "grad_norm": 1.9847423514176032,
      "language_loss": 0.85453618,
      "learning_rate": 1.5157092060833975e-07,
      "loss": 0.87699711,
      "num_input_tokens_seen": 158243145,
      "step": 7310,
      "time_per_iteration": 2.672056198120117
    },
    {
      "auxiliary_loss_clip": 0.01172144,
      "auxiliary_loss_mlp": 0.01026223,
      "balance_loss_clip": 1.04413879,
      "balance_loss_mlp": 1.01924038,
      "epoch": 0.879095773462394,
      "flos": 29310971408640.0,
      "grad_norm": 1.8656145751471023,
      "language_loss": 0.6616869,
      "learning_rate": 1.5127359207221658e-07,
      "loss": 0.68367058,
      "num_input_tokens_seen": 158262625,
      "step": 7311,
      "time_per_iteration": 3.6691126823425293
    },
    {
      "auxiliary_loss_clip": 0.01417892,
      "auxiliary_loss_mlp": 0.01025496,
      "balance_loss_clip": 1.03726077,
      "balance_loss_mlp": 1.01802731,
      "epoch": 0.8792160163530331,
      "flos": 16690023394560.0,
      "grad_norm": 2.284784536246755,
      "language_loss": 0.73737812,
      "learning_rate": 1.5097654398641923e-07,
      "loss": 0.76181197,
      "num_input_tokens_seen": 158280530,
      "step": 7312,
      "time_per_iteration": 2.7443835735321045
    },
    {
      "auxiliary_loss_clip": 0.01126546,
      "auxiliary_loss_mlp": 0.01023941,
      "balance_loss_clip": 1.0489192,
      "balance_loss_mlp": 1.01705635,
      "epoch": 0.8793362592436722,
      "flos": 24499301230080.0,
      "grad_norm": 1.6138612460199269,
      "language_loss": 0.73218554,
      "learning_rate": 1.5067977639601014e-07,
      "loss": 0.75369042,
      "num_input_tokens_seen": 158303290,
      "step": 7313,
      "time_per_iteration": 2.7289884090423584
    },
    {
      "auxiliary_loss_clip": 0.01269335,
      "auxiliary_loss_mlp": 0.01022469,
      "balance_loss_clip": 1.04632056,
      "balance_loss_mlp": 1.01563203,
      "epoch": 0.8794565021343113,
      "flos": 14538399834240.0,
      "grad_norm": 3.1865915987335134,
      "language_loss": 0.71207809,
      "learning_rate": 1.5038328934600864e-07,
      "loss": 0.73499614,
      "num_input_tokens_seen": 158319925,
      "step": 7314,
      "time_per_iteration": 3.540827751159668
    },
    {
      "auxiliary_loss_clip": 0.01271439,
      "auxiliary_loss_mlp": 0.01027412,
      "balance_loss_clip": 1.04894388,
      "balance_loss_mlp": 1.02056623,
      "epoch": 0.8795767450249504,
      "flos": 39530286224640.0,
      "grad_norm": 1.8487222649044321,
      "language_loss": 0.70591855,
      "learning_rate": 1.5008708288139161e-07,
      "loss": 0.72890711,
      "num_input_tokens_seen": 158342285,
      "step": 7315,
      "time_per_iteration": 2.8140828609466553
    },
    {
      "auxiliary_loss_clip": 0.0122675,
      "auxiliary_loss_mlp": 0.01020278,
      "balance_loss_clip": 1.05126977,
      "balance_loss_mlp": 1.012941,
      "epoch": 0.8796969879155895,
      "flos": 22960672197120.0,
      "grad_norm": 2.176562650504814,
      "language_loss": 0.73531139,
      "learning_rate": 1.497911570470931e-07,
      "loss": 0.75778168,
      "num_input_tokens_seen": 158362290,
      "step": 7316,
      "time_per_iteration": 2.6969447135925293
    },
    {
      "auxiliary_loss_clip": 0.01312801,
      "auxiliary_loss_mlp": 0.01029162,
      "balance_loss_clip": 1.04350853,
      "balance_loss_mlp": 1.0223316,
      "epoch": 0.8798172308062285,
      "flos": 28362427004160.0,
      "grad_norm": 1.97484434722781,
      "language_loss": 0.85554618,
      "learning_rate": 1.494955118880048e-07,
      "loss": 0.87896585,
      "num_input_tokens_seen": 158383275,
      "step": 7317,
      "time_per_iteration": 2.7322680950164795
    },
    {
      "auxiliary_loss_clip": 0.01222855,
      "auxiliary_loss_mlp": 0.01022792,
      "balance_loss_clip": 1.04731596,
      "balance_loss_mlp": 1.01583934,
      "epoch": 0.8799374736968677,
      "flos": 23988974751360.0,
      "grad_norm": 1.7813382610153483,
      "language_loss": 0.72909725,
      "learning_rate": 1.4920014744897634e-07,
      "loss": 0.75155377,
      "num_input_tokens_seen": 158402690,
      "step": 7318,
      "time_per_iteration": 2.7537519931793213
    },
    {
      "auxiliary_loss_clip": 0.01267439,
      "auxiliary_loss_mlp": 0.01027926,
      "balance_loss_clip": 1.04510856,
      "balance_loss_mlp": 1.02099419,
      "epoch": 0.8800577165875068,
      "flos": 25630271832960.0,
      "grad_norm": 2.1973758673051678,
      "language_loss": 0.86115426,
      "learning_rate": 1.4890506377481392e-07,
      "loss": 0.88410795,
      "num_input_tokens_seen": 158421780,
      "step": 7319,
      "time_per_iteration": 2.6970462799072266
    },
    {
      "auxiliary_loss_clip": 0.01326207,
      "auxiliary_loss_mlp": 0.01023334,
      "balance_loss_clip": 1.04631019,
      "balance_loss_mlp": 1.01680732,
      "epoch": 0.8801779594781458,
      "flos": 23440331439360.0,
      "grad_norm": 1.5537291353551215,
      "language_loss": 0.63945633,
      "learning_rate": 1.486102609102815e-07,
      "loss": 0.66295171,
      "num_input_tokens_seen": 158442330,
      "step": 7320,
      "time_per_iteration": 2.8720717430114746
    },
    {
      "auxiliary_loss_clip": 0.01265032,
      "auxiliary_loss_mlp": 0.01026538,
      "balance_loss_clip": 1.04478431,
      "balance_loss_mlp": 1.0196805,
      "epoch": 0.880298202368785,
      "flos": 11508580656000.0,
      "grad_norm": 3.201891021006774,
      "language_loss": 0.85826665,
      "learning_rate": 1.483157389001004e-07,
      "loss": 0.88118231,
      "num_input_tokens_seen": 158459890,
      "step": 7321,
      "time_per_iteration": 2.567586660385132
    },
    {
      "auxiliary_loss_clip": 0.01270287,
      "auxiliary_loss_mlp": 0.0102519,
      "balance_loss_clip": 1.04191113,
      "balance_loss_mlp": 1.01755142,
      "epoch": 0.880418445259424,
      "flos": 22671447886080.0,
      "grad_norm": 2.4360468696608586,
      "language_loss": 0.78822601,
      "learning_rate": 1.4802149778894933e-07,
      "loss": 0.81118077,
      "num_input_tokens_seen": 158478680,
      "step": 7322,
      "time_per_iteration": 2.702939987182617
    },
    {
      "auxiliary_loss_clip": 0.01213956,
      "auxiliary_loss_mlp": 0.01023651,
      "balance_loss_clip": 1.04198229,
      "balance_loss_mlp": 1.01712751,
      "epoch": 0.8805386881500631,
      "flos": 20522158709760.0,
      "grad_norm": 2.0856646748944434,
      "language_loss": 0.87629378,
      "learning_rate": 1.4772753762146484e-07,
      "loss": 0.89866978,
      "num_input_tokens_seen": 158497935,
      "step": 7323,
      "time_per_iteration": 2.5900728702545166
    },
    {
      "auxiliary_loss_clip": 0.012178,
      "auxiliary_loss_mlp": 0.01029781,
      "balance_loss_clip": 1.04380751,
      "balance_loss_mlp": 1.02207398,
      "epoch": 0.8806589310407023,
      "flos": 36538891620480.0,
      "grad_norm": 2.553097955552744,
      "language_loss": 0.70695686,
      "learning_rate": 1.474338584422401e-07,
      "loss": 0.72943258,
      "num_input_tokens_seen": 158523145,
      "step": 7324,
      "time_per_iteration": 2.762192964553833
    },
    {
      "auxiliary_loss_clip": 0.01216424,
      "auxiliary_loss_mlp": 0.01021411,
      "balance_loss_clip": 1.04720426,
      "balance_loss_mlp": 1.01468205,
      "epoch": 0.8807791739313413,
      "flos": 23440187784960.0,
      "grad_norm": 1.7331008222598208,
      "language_loss": 0.75634396,
      "learning_rate": 1.4714046029582595e-07,
      "loss": 0.77872229,
      "num_input_tokens_seen": 158542210,
      "step": 7325,
      "time_per_iteration": 2.648693799972534
    },
    {
      "auxiliary_loss_clip": 0.01324753,
      "auxiliary_loss_mlp": 0.01026602,
      "balance_loss_clip": 1.04455805,
      "balance_loss_mlp": 1.01977158,
      "epoch": 0.8808994168219804,
      "flos": 25956843310080.0,
      "grad_norm": 1.87648745389448,
      "language_loss": 0.75970292,
      "learning_rate": 1.46847343226731e-07,
      "loss": 0.78321648,
      "num_input_tokens_seen": 158563250,
      "step": 7326,
      "time_per_iteration": 2.862873077392578
    },
    {
      "auxiliary_loss_clip": 0.01226931,
      "auxiliary_loss_mlp": 0.01024844,
      "balance_loss_clip": 1.04878378,
      "balance_loss_mlp": 1.01836848,
      "epoch": 0.8810196597126195,
      "flos": 17092079303040.0,
      "grad_norm": 2.2350771523976842,
      "language_loss": 0.69322914,
      "learning_rate": 1.465545072794203e-07,
      "loss": 0.71574688,
      "num_input_tokens_seen": 158581125,
      "step": 7327,
      "time_per_iteration": 2.616023540496826
    },
    {
      "auxiliary_loss_clip": 0.01270535,
      "auxiliary_loss_mlp": 0.01025444,
      "balance_loss_clip": 1.0454576,
      "balance_loss_mlp": 1.01862788,
      "epoch": 0.8811399026032586,
      "flos": 23002831785600.0,
      "grad_norm": 1.7175433447945296,
      "language_loss": 0.75800216,
      "learning_rate": 1.4626195249831774e-07,
      "loss": 0.78096199,
      "num_input_tokens_seen": 158602025,
      "step": 7328,
      "time_per_iteration": 2.787062406539917
    },
    {
      "auxiliary_loss_clip": 0.01219414,
      "auxiliary_loss_mlp": 0.01027146,
      "balance_loss_clip": 1.04444289,
      "balance_loss_mlp": 1.02027404,
      "epoch": 0.8812601454938976,
      "flos": 14463813242880.0,
      "grad_norm": 1.9193512076272574,
      "language_loss": 0.71863717,
      "learning_rate": 1.4596967892780244e-07,
      "loss": 0.74110276,
      "num_input_tokens_seen": 158618355,
      "step": 7329,
      "time_per_iteration": 2.642733097076416
    },
    {
      "auxiliary_loss_clip": 0.01171691,
      "auxiliary_loss_mlp": 0.01022251,
      "balance_loss_clip": 1.04912853,
      "balance_loss_mlp": 1.01611149,
      "epoch": 0.8813803883845368,
      "flos": 22493223578880.0,
      "grad_norm": 2.431684832132277,
      "language_loss": 0.74453938,
      "learning_rate": 1.4567768661221314e-07,
      "loss": 0.76647884,
      "num_input_tokens_seen": 158638925,
      "step": 7330,
      "time_per_iteration": 2.6536061763763428
    },
    {
      "auxiliary_loss_clip": 0.01227637,
      "auxiliary_loss_mlp": 0.02566425,
      "balance_loss_clip": 1.04911137,
      "balance_loss_mlp": 0.99992144,
      "epoch": 0.8815006312751759,
      "flos": 21506901045120.0,
      "grad_norm": 2.111432920466043,
      "language_loss": 0.74735117,
      "learning_rate": 1.4538597559584442e-07,
      "loss": 0.78529173,
      "num_input_tokens_seen": 158656715,
      "step": 7331,
      "time_per_iteration": 2.680534601211548
    },
    {
      "auxiliary_loss_clip": 0.01269249,
      "auxiliary_loss_mlp": 0.01021947,
      "balance_loss_clip": 1.04547513,
      "balance_loss_mlp": 1.0146451,
      "epoch": 0.8816208741658149,
      "flos": 22784566792320.0,
      "grad_norm": 2.304527689527815,
      "language_loss": 0.78853637,
      "learning_rate": 1.4509454592294823e-07,
      "loss": 0.81144834,
      "num_input_tokens_seen": 158677200,
      "step": 7332,
      "time_per_iteration": 2.7490921020507812
    },
    {
      "auxiliary_loss_clip": 0.01327597,
      "auxiliary_loss_mlp": 0.02567915,
      "balance_loss_clip": 1.04807258,
      "balance_loss_mlp": 0.99992251,
      "epoch": 0.8817411170564541,
      "flos": 17779409026560.0,
      "grad_norm": 2.17337299563662,
      "language_loss": 0.786129,
      "learning_rate": 1.448033976377354e-07,
      "loss": 0.82508409,
      "num_input_tokens_seen": 158692185,
      "step": 7333,
      "time_per_iteration": 2.734869956970215
    },
    {
      "auxiliary_loss_clip": 0.01222591,
      "auxiliary_loss_mlp": 0.01022938,
      "balance_loss_clip": 1.04508519,
      "balance_loss_mlp": 1.01653373,
      "epoch": 0.8818613599470931,
      "flos": 18551812112640.0,
      "grad_norm": 2.0485433799345847,
      "language_loss": 0.74170417,
      "learning_rate": 1.445125307843713e-07,
      "loss": 0.76415944,
      "num_input_tokens_seen": 158710410,
      "step": 7334,
      "time_per_iteration": 4.496086120605469
    },
    {
      "auxiliary_loss_clip": 0.01219292,
      "auxiliary_loss_mlp": 0.01025667,
      "balance_loss_clip": 1.04816508,
      "balance_loss_mlp": 1.01879168,
      "epoch": 0.8819816028377322,
      "flos": 27599792417280.0,
      "grad_norm": 1.6873821645974334,
      "language_loss": 0.75910187,
      "learning_rate": 1.442219454069813e-07,
      "loss": 0.78155148,
      "num_input_tokens_seen": 158731435,
      "step": 7335,
      "time_per_iteration": 2.6921231746673584
    },
    {
      "auxiliary_loss_clip": 0.01375336,
      "auxiliary_loss_mlp": 0.01031044,
      "balance_loss_clip": 1.04349041,
      "balance_loss_mlp": 1.0242914,
      "epoch": 0.8821018457283714,
      "flos": 23404600385280.0,
      "grad_norm": 2.8010235160232457,
      "language_loss": 0.66620779,
      "learning_rate": 1.4393164154964676e-07,
      "loss": 0.69027162,
      "num_input_tokens_seen": 158750965,
      "step": 7336,
      "time_per_iteration": 2.795656681060791
    },
    {
      "auxiliary_loss_clip": 0.01227566,
      "auxiliary_loss_mlp": 0.01029401,
      "balance_loss_clip": 1.05238199,
      "balance_loss_mlp": 1.02200365,
      "epoch": 0.8822220886190104,
      "flos": 29132459792640.0,
      "grad_norm": 1.7740600112751952,
      "language_loss": 0.94155294,
      "learning_rate": 1.4364161925640649e-07,
      "loss": 0.96412253,
      "num_input_tokens_seen": 158772365,
      "step": 7337,
      "time_per_iteration": 3.616431713104248
    },
    {
      "auxiliary_loss_clip": 0.01171591,
      "auxiliary_loss_mlp": 0.01024603,
      "balance_loss_clip": 1.04801071,
      "balance_loss_mlp": 1.01846933,
      "epoch": 0.8823423315096495,
      "flos": 20485422074880.0,
      "grad_norm": 2.186790461387461,
      "language_loss": 0.85071641,
      "learning_rate": 1.4335187857125663e-07,
      "loss": 0.87267834,
      "num_input_tokens_seen": 158791065,
      "step": 7338,
      "time_per_iteration": 2.6117806434631348
    },
    {
      "auxiliary_loss_clip": 0.01223571,
      "auxiliary_loss_mlp": 0.01029408,
      "balance_loss_clip": 1.04706883,
      "balance_loss_mlp": 1.02253568,
      "epoch": 0.8824625744002886,
      "flos": 24206377818240.0,
      "grad_norm": 1.7492575243104,
      "language_loss": 0.75840783,
      "learning_rate": 1.4306241953815023e-07,
      "loss": 0.78093755,
      "num_input_tokens_seen": 158812125,
      "step": 7339,
      "time_per_iteration": 2.7079453468322754
    },
    {
      "auxiliary_loss_clip": 0.01223791,
      "auxiliary_loss_mlp": 0.01029526,
      "balance_loss_clip": 1.04704714,
      "balance_loss_mlp": 1.02325284,
      "epoch": 0.8825828172909277,
      "flos": 24679500785280.0,
      "grad_norm": 1.846427176840634,
      "language_loss": 0.71412981,
      "learning_rate": 1.4277324220099862e-07,
      "loss": 0.73666304,
      "num_input_tokens_seen": 158834035,
      "step": 7340,
      "time_per_iteration": 3.556201696395874
    },
    {
      "auxiliary_loss_clip": 0.01317914,
      "auxiliary_loss_mlp": 0.01024885,
      "balance_loss_clip": 1.04182041,
      "balance_loss_mlp": 1.017905,
      "epoch": 0.8827030601815667,
      "flos": 22456163721600.0,
      "grad_norm": 2.0137296240438176,
      "language_loss": 0.74163628,
      "learning_rate": 1.4248434660366938e-07,
      "loss": 0.7650643,
      "num_input_tokens_seen": 158853510,
      "step": 7341,
      "time_per_iteration": 2.7116291522979736
    },
    {
      "auxiliary_loss_clip": 0.01271055,
      "auxiliary_loss_mlp": 0.01026551,
      "balance_loss_clip": 1.04832554,
      "balance_loss_mlp": 1.01978314,
      "epoch": 0.8828233030722058,
      "flos": 19865639877120.0,
      "grad_norm": 2.2357768874443646,
      "language_loss": 0.7030068,
      "learning_rate": 1.4219573278998808e-07,
      "loss": 0.72598284,
      "num_input_tokens_seen": 158871970,
      "step": 7342,
      "time_per_iteration": 2.6980764865875244
    },
    {
      "auxiliary_loss_clip": 0.01273312,
      "auxiliary_loss_mlp": 0.01024047,
      "balance_loss_clip": 1.04375625,
      "balance_loss_mlp": 1.01677847,
      "epoch": 0.882943545962845,
      "flos": 39347213581440.0,
      "grad_norm": 3.0571938823276494,
      "language_loss": 0.65116316,
      "learning_rate": 1.4190740080373685e-07,
      "loss": 0.67413676,
      "num_input_tokens_seen": 158892250,
      "step": 7343,
      "time_per_iteration": 2.7685272693634033
    },
    {
      "auxiliary_loss_clip": 0.01368345,
      "auxiliary_loss_mlp": 0.01023588,
      "balance_loss_clip": 1.04226232,
      "balance_loss_mlp": 1.0160836,
      "epoch": 0.883063788853484,
      "flos": 19054524908160.0,
      "grad_norm": 2.0678173602275627,
      "language_loss": 0.84038186,
      "learning_rate": 1.4161935068865538e-07,
      "loss": 0.8643012,
      "num_input_tokens_seen": 158907395,
      "step": 7344,
      "time_per_iteration": 2.6727406978607178
    },
    {
      "auxiliary_loss_clip": 0.01173889,
      "auxiliary_loss_mlp": 0.01025832,
      "balance_loss_clip": 1.04965568,
      "balance_loss_mlp": 1.01843464,
      "epoch": 0.8831840317441231,
      "flos": 18733196816640.0,
      "grad_norm": 2.062243846718292,
      "language_loss": 0.75656617,
      "learning_rate": 1.4133158248844113e-07,
      "loss": 0.77856332,
      "num_input_tokens_seen": 158926300,
      "step": 7345,
      "time_per_iteration": 2.6615543365478516
    },
    {
      "auxiliary_loss_clip": 0.0121909,
      "auxiliary_loss_mlp": 0.01023316,
      "balance_loss_clip": 1.04373741,
      "balance_loss_mlp": 1.01592803,
      "epoch": 0.8833042746347622,
      "flos": 26827712553600.0,
      "grad_norm": 1.9795344453013903,
      "language_loss": 0.73278463,
      "learning_rate": 1.4104409624674785e-07,
      "loss": 0.75520873,
      "num_input_tokens_seen": 158946085,
      "step": 7346,
      "time_per_iteration": 2.701359272003174
    },
    {
      "auxiliary_loss_clip": 0.01223717,
      "auxiliary_loss_mlp": 0.01025806,
      "balance_loss_clip": 1.05040801,
      "balance_loss_mlp": 1.01917505,
      "epoch": 0.8834245175254013,
      "flos": 26104077158400.0,
      "grad_norm": 1.825370580375947,
      "language_loss": 0.78759128,
      "learning_rate": 1.407568920071873e-07,
      "loss": 0.81008649,
      "num_input_tokens_seen": 158964950,
      "step": 7347,
      "time_per_iteration": 2.701831340789795
    },
    {
      "auxiliary_loss_clip": 0.0117801,
      "auxiliary_loss_mlp": 0.01027442,
      "balance_loss_clip": 1.05044532,
      "balance_loss_mlp": 1.01987791,
      "epoch": 0.8835447604160404,
      "flos": 30629036977920.0,
      "grad_norm": 2.016378527691623,
      "language_loss": 0.68550783,
      "learning_rate": 1.4046996981332782e-07,
      "loss": 0.70756239,
      "num_input_tokens_seen": 158984835,
      "step": 7348,
      "time_per_iteration": 2.7356677055358887
    },
    {
      "auxiliary_loss_clip": 0.01222156,
      "auxiliary_loss_mlp": 0.01024415,
      "balance_loss_clip": 1.04494643,
      "balance_loss_mlp": 1.01710176,
      "epoch": 0.8836650033066795,
      "flos": 24718356322560.0,
      "grad_norm": 3.1558971321353972,
      "language_loss": 0.78087866,
      "learning_rate": 1.4018332970869516e-07,
      "loss": 0.80334443,
      "num_input_tokens_seen": 159002775,
      "step": 7349,
      "time_per_iteration": 2.6898298263549805
    },
    {
      "auxiliary_loss_clip": 0.01275611,
      "auxiliary_loss_mlp": 0.01025983,
      "balance_loss_clip": 1.05000305,
      "balance_loss_mlp": 1.01872957,
      "epoch": 0.8837852461973186,
      "flos": 25413371556480.0,
      "grad_norm": 3.784461360428959,
      "language_loss": 0.85200697,
      "learning_rate": 1.398969717367733e-07,
      "loss": 0.87502295,
      "num_input_tokens_seen": 159024100,
      "step": 7350,
      "time_per_iteration": 2.748755931854248
    },
    {
      "auxiliary_loss_clip": 0.01263988,
      "auxiliary_loss_mlp": 0.0102679,
      "balance_loss_clip": 1.04526067,
      "balance_loss_mlp": 1.02009904,
      "epoch": 0.8839054890879576,
      "flos": 17822574195840.0,
      "grad_norm": 1.8965431264166648,
      "language_loss": 0.76458442,
      "learning_rate": 1.396108959410014e-07,
      "loss": 0.78749216,
      "num_input_tokens_seen": 159043315,
      "step": 7351,
      "time_per_iteration": 2.682246685028076
    },
    {
      "auxiliary_loss_clip": 0.01221967,
      "auxiliary_loss_mlp": 0.02567005,
      "balance_loss_clip": 1.04870272,
      "balance_loss_mlp": 0.99988085,
      "epoch": 0.8840257319785968,
      "flos": 23769021818880.0,
      "grad_norm": 2.044767085568208,
      "language_loss": 0.81362438,
      "learning_rate": 1.3932510236477745e-07,
      "loss": 0.8515141,
      "num_input_tokens_seen": 159063985,
      "step": 7352,
      "time_per_iteration": 2.709618330001831
    },
    {
      "auxiliary_loss_clip": 0.0122235,
      "auxiliary_loss_mlp": 0.01025382,
      "balance_loss_clip": 1.0452224,
      "balance_loss_mlp": 1.01800036,
      "epoch": 0.8841459748692359,
      "flos": 29059776622080.0,
      "grad_norm": 2.07152122771741,
      "language_loss": 0.56306332,
      "learning_rate": 1.3903959105145636e-07,
      "loss": 0.58554065,
      "num_input_tokens_seen": 159084475,
      "step": 7353,
      "time_per_iteration": 2.667526960372925
    },
    {
      "auxiliary_loss_clip": 0.01169792,
      "auxiliary_loss_mlp": 0.01024424,
      "balance_loss_clip": 1.04704106,
      "balance_loss_mlp": 1.0180676,
      "epoch": 0.8842662177598749,
      "flos": 24311523905280.0,
      "grad_norm": 2.0216512598166436,
      "language_loss": 0.832744,
      "learning_rate": 1.387543620443492e-07,
      "loss": 0.85468614,
      "num_input_tokens_seen": 159101320,
      "step": 7354,
      "time_per_iteration": 2.6250710487365723
    },
    {
      "auxiliary_loss_clip": 0.0117064,
      "auxiliary_loss_mlp": 0.01026102,
      "balance_loss_clip": 1.04852319,
      "balance_loss_mlp": 1.01972723,
      "epoch": 0.8843864606505141,
      "flos": 25007867942400.0,
      "grad_norm": 1.9882581166257396,
      "language_loss": 0.84315526,
      "learning_rate": 1.3846941538672606e-07,
      "loss": 0.86512268,
      "num_input_tokens_seen": 159120025,
      "step": 7355,
      "time_per_iteration": 2.6054844856262207
    },
    {
      "auxiliary_loss_clip": 0.01373217,
      "auxiliary_loss_mlp": 0.01028897,
      "balance_loss_clip": 1.0451951,
      "balance_loss_mlp": 1.02221513,
      "epoch": 0.8845067035411531,
      "flos": 28183915388160.0,
      "grad_norm": 2.2774921862776263,
      "language_loss": 0.8093859,
      "learning_rate": 1.3818475112181193e-07,
      "loss": 0.83340704,
      "num_input_tokens_seen": 159138820,
      "step": 7356,
      "time_per_iteration": 2.8032939434051514
    },
    {
      "auxiliary_loss_clip": 0.01168589,
      "auxiliary_loss_mlp": 0.01024801,
      "balance_loss_clip": 1.04653525,
      "balance_loss_mlp": 1.01822019,
      "epoch": 0.8846269464317922,
      "flos": 12853219311360.0,
      "grad_norm": 2.166088324066519,
      "language_loss": 0.79624707,
      "learning_rate": 1.3790036929279091e-07,
      "loss": 0.81818092,
      "num_input_tokens_seen": 159155975,
      "step": 7357,
      "time_per_iteration": 2.6044211387634277
    },
    {
      "auxiliary_loss_clip": 0.01220728,
      "auxiliary_loss_mlp": 0.02564581,
      "balance_loss_clip": 1.04760718,
      "balance_loss_mlp": 0.99988657,
      "epoch": 0.8847471893224313,
      "flos": 18624351628800.0,
      "grad_norm": 2.2066664756451293,
      "language_loss": 0.59407628,
      "learning_rate": 1.3761626994280363e-07,
      "loss": 0.63192934,
      "num_input_tokens_seen": 159173445,
      "step": 7358,
      "time_per_iteration": 2.738797903060913
    },
    {
      "auxiliary_loss_clip": 0.01326093,
      "auxiliary_loss_mlp": 0.0102699,
      "balance_loss_clip": 1.04482365,
      "balance_loss_mlp": 1.02018952,
      "epoch": 0.8848674322130704,
      "flos": 35769433449600.0,
      "grad_norm": 1.8035340292053146,
      "language_loss": 0.73662883,
      "learning_rate": 1.3733245311494735e-07,
      "loss": 0.76015967,
      "num_input_tokens_seen": 159196100,
      "step": 7359,
      "time_per_iteration": 3.657203435897827
    },
    {
      "auxiliary_loss_clip": 0.01222452,
      "auxiliary_loss_mlp": 0.01022843,
      "balance_loss_clip": 1.04832029,
      "balance_loss_mlp": 1.01594353,
      "epoch": 0.8849876751037095,
      "flos": 24243760897920.0,
      "grad_norm": 2.2718099462960883,
      "language_loss": 0.70425606,
      "learning_rate": 1.3704891885227676e-07,
      "loss": 0.72670901,
      "num_input_tokens_seen": 159216145,
      "step": 7360,
      "time_per_iteration": 2.628399610519409
    },
    {
      "auxiliary_loss_clip": 0.01325247,
      "auxiliary_loss_mlp": 0.01028858,
      "balance_loss_clip": 1.04190588,
      "balance_loss_mlp": 1.02125275,
      "epoch": 0.8851079179943486,
      "flos": 21500580251520.0,
      "grad_norm": 3.4571200140530163,
      "language_loss": 0.78032184,
      "learning_rate": 1.367656671978037e-07,
      "loss": 0.80386293,
      "num_input_tokens_seen": 159233610,
      "step": 7361,
      "time_per_iteration": 3.6607017517089844
    },
    {
      "auxiliary_loss_clip": 0.01278443,
      "auxiliary_loss_mlp": 0.01024152,
      "balance_loss_clip": 1.04628563,
      "balance_loss_mlp": 1.01733947,
      "epoch": 0.8852281608849877,
      "flos": 15300711198720.0,
      "grad_norm": 1.8920042621806061,
      "language_loss": 0.73377997,
      "learning_rate": 1.36482698194498e-07,
      "loss": 0.7568059,
      "num_input_tokens_seen": 159250155,
      "step": 7362,
      "time_per_iteration": 2.671027421951294
    },
    {
      "auxiliary_loss_clip": 0.01269704,
      "auxiliary_loss_mlp": 0.01027179,
      "balance_loss_clip": 1.04469025,
      "balance_loss_mlp": 1.02019882,
      "epoch": 0.8853484037756267,
      "flos": 23295719283840.0,
      "grad_norm": 1.8382819463688291,
      "language_loss": 0.72190213,
      "learning_rate": 1.3620001188528506e-07,
      "loss": 0.74487102,
      "num_input_tokens_seen": 159270875,
      "step": 7363,
      "time_per_iteration": 3.7116382122039795
    },
    {
      "auxiliary_loss_clip": 0.01223163,
      "auxiliary_loss_mlp": 0.01024252,
      "balance_loss_clip": 1.04526162,
      "balance_loss_mlp": 1.01751995,
      "epoch": 0.8854686466662659,
      "flos": 25114773795840.0,
      "grad_norm": 2.471766360138975,
      "language_loss": 0.73717701,
      "learning_rate": 1.3591760831304865e-07,
      "loss": 0.75965118,
      "num_input_tokens_seen": 159288565,
      "step": 7364,
      "time_per_iteration": 2.6521761417388916
    },
    {
      "auxiliary_loss_clip": 0.01172416,
      "auxiliary_loss_mlp": 0.01026014,
      "balance_loss_clip": 1.04905343,
      "balance_loss_mlp": 1.01887894,
      "epoch": 0.885588889556905,
      "flos": 21390873137280.0,
      "grad_norm": 1.7638770351518518,
      "language_loss": 0.79220796,
      "learning_rate": 1.356354875206287e-07,
      "loss": 0.81419218,
      "num_input_tokens_seen": 159306400,
      "step": 7365,
      "time_per_iteration": 2.5754919052124023
    },
    {
      "auxiliary_loss_clip": 0.01317416,
      "auxiliary_loss_mlp": 0.01025611,
      "balance_loss_clip": 1.04640532,
      "balance_loss_mlp": 1.01861966,
      "epoch": 0.885709132447544,
      "flos": 26906752431360.0,
      "grad_norm": 2.242904258333885,
      "language_loss": 0.69870919,
      "learning_rate": 1.3535364955082296e-07,
      "loss": 0.72213948,
      "num_input_tokens_seen": 159326250,
      "step": 7366,
      "time_per_iteration": 3.6201729774475098
    },
    {
      "auxiliary_loss_clip": 0.01173246,
      "auxiliary_loss_mlp": 0.01023771,
      "balance_loss_clip": 1.05132747,
      "balance_loss_mlp": 1.01695204,
      "epoch": 0.8858293753381832,
      "flos": 26103394800000.0,
      "grad_norm": 1.9766735866031175,
      "language_loss": 0.64991766,
      "learning_rate": 1.3507209444638613e-07,
      "loss": 0.67188787,
      "num_input_tokens_seen": 159348250,
      "step": 7367,
      "time_per_iteration": 2.654050588607788
    },
    {
      "auxiliary_loss_clip": 0.01221184,
      "auxiliary_loss_mlp": 0.01024346,
      "balance_loss_clip": 1.04892516,
      "balance_loss_mlp": 1.01761949,
      "epoch": 0.8859496182288222,
      "flos": 23292810282240.0,
      "grad_norm": 1.943862046912065,
      "language_loss": 0.74070781,
      "learning_rate": 1.347908222500298e-07,
      "loss": 0.76316315,
      "num_input_tokens_seen": 159368325,
      "step": 7368,
      "time_per_iteration": 2.601515531539917
    },
    {
      "auxiliary_loss_clip": 0.01316453,
      "auxiliary_loss_mlp": 0.01021728,
      "balance_loss_clip": 1.04671144,
      "balance_loss_mlp": 1.01526141,
      "epoch": 0.8860698611194613,
      "flos": 16872916469760.0,
      "grad_norm": 3.7059924241699727,
      "language_loss": 0.70000553,
      "learning_rate": 1.3450983300442276e-07,
      "loss": 0.72338736,
      "num_input_tokens_seen": 159387555,
      "step": 7369,
      "time_per_iteration": 2.747530221939087
    },
    {
      "auxiliary_loss_clip": 0.01224894,
      "auxiliary_loss_mlp": 0.01026136,
      "balance_loss_clip": 1.04837048,
      "balance_loss_mlp": 1.01958525,
      "epoch": 0.8861901040101005,
      "flos": 24681404206080.0,
      "grad_norm": 1.9746201122831275,
      "language_loss": 0.7377702,
      "learning_rate": 1.3422912675219068e-07,
      "loss": 0.76028049,
      "num_input_tokens_seen": 159407310,
      "step": 7370,
      "time_per_iteration": 2.6386117935180664
    },
    {
      "auxiliary_loss_clip": 0.01171878,
      "auxiliary_loss_mlp": 0.01028858,
      "balance_loss_clip": 1.05020213,
      "balance_loss_mlp": 1.02231956,
      "epoch": 0.8863103469007395,
      "flos": 24423026699520.0,
      "grad_norm": 1.5720991744848314,
      "language_loss": 0.79138064,
      "learning_rate": 1.339487035359166e-07,
      "loss": 0.81338805,
      "num_input_tokens_seen": 159427680,
      "step": 7371,
      "time_per_iteration": 2.623685359954834
    },
    {
      "auxiliary_loss_clip": 0.01274242,
      "auxiliary_loss_mlp": 0.02562072,
      "balance_loss_clip": 1.04907227,
      "balance_loss_mlp": 0.99987918,
      "epoch": 0.8864305897913786,
      "flos": 22053964158720.0,
      "grad_norm": 1.5892674647402252,
      "language_loss": 0.8470186,
      "learning_rate": 1.336685633981409e-07,
      "loss": 0.88538176,
      "num_input_tokens_seen": 159448765,
      "step": 7372,
      "time_per_iteration": 2.7033698558807373
    },
    {
      "auxiliary_loss_clip": 0.01225237,
      "auxiliary_loss_mlp": 0.01028289,
      "balance_loss_clip": 1.04810619,
      "balance_loss_mlp": 1.02048659,
      "epoch": 0.8865508326820177,
      "flos": 19099449843840.0,
      "grad_norm": 5.530329441921911,
      "language_loss": 0.74949569,
      "learning_rate": 1.333887063813597e-07,
      "loss": 0.77203095,
      "num_input_tokens_seen": 159466870,
      "step": 7373,
      "time_per_iteration": 2.651745557785034
    },
    {
      "auxiliary_loss_clip": 0.01272395,
      "auxiliary_loss_mlp": 0.0102725,
      "balance_loss_clip": 1.04475832,
      "balance_loss_mlp": 1.02088141,
      "epoch": 0.8866710755726568,
      "flos": 15414189240960.0,
      "grad_norm": 1.8177045008327999,
      "language_loss": 0.66448891,
      "learning_rate": 1.331091325280278e-07,
      "loss": 0.68748534,
      "num_input_tokens_seen": 159485840,
      "step": 7374,
      "time_per_iteration": 2.642770290374756
    },
    {
      "auxiliary_loss_clip": 0.01372449,
      "auxiliary_loss_mlp": 0.01029288,
      "balance_loss_clip": 1.04642582,
      "balance_loss_mlp": 1.02201009,
      "epoch": 0.8867913184632958,
      "flos": 20083689388800.0,
      "grad_norm": 2.3072593327739486,
      "language_loss": 0.78629494,
      "learning_rate": 1.3282984188055625e-07,
      "loss": 0.81031227,
      "num_input_tokens_seen": 159505630,
      "step": 7375,
      "time_per_iteration": 2.8295605182647705
    },
    {
      "auxiliary_loss_clip": 0.01173752,
      "auxiliary_loss_mlp": 0.01029572,
      "balance_loss_clip": 1.04912186,
      "balance_loss_mlp": 1.02263403,
      "epoch": 0.8869115613539349,
      "flos": 23365852588800.0,
      "grad_norm": 1.6721933654793701,
      "language_loss": 0.79528856,
      "learning_rate": 1.3255083448131288e-07,
      "loss": 0.81732172,
      "num_input_tokens_seen": 159524675,
      "step": 7376,
      "time_per_iteration": 2.616567611694336
    },
    {
      "auxiliary_loss_clip": 0.01224959,
      "auxiliary_loss_mlp": 0.01029621,
      "balance_loss_clip": 1.04564297,
      "balance_loss_mlp": 1.02215862,
      "epoch": 0.8870318042445741,
      "flos": 21286840371840.0,
      "grad_norm": 2.0741285204780335,
      "language_loss": 0.79201901,
      "learning_rate": 1.3227211037262365e-07,
      "loss": 0.81456488,
      "num_input_tokens_seen": 159541915,
      "step": 7377,
      "time_per_iteration": 2.633108615875244
    },
    {
      "auxiliary_loss_clip": 0.01274578,
      "auxiliary_loss_mlp": 0.01024058,
      "balance_loss_clip": 1.04264736,
      "balance_loss_mlp": 1.01700056,
      "epoch": 0.8871520471352131,
      "flos": 20010862563840.0,
      "grad_norm": 5.683834652954261,
      "language_loss": 0.85298836,
      "learning_rate": 1.319936695967696e-07,
      "loss": 0.87597477,
      "num_input_tokens_seen": 159559740,
      "step": 7378,
      "time_per_iteration": 2.6847503185272217
    },
    {
      "auxiliary_loss_clip": 0.01177562,
      "auxiliary_loss_mlp": 0.01026533,
      "balance_loss_clip": 1.04862046,
      "balance_loss_mlp": 1.01893377,
      "epoch": 0.8872722900258522,
      "flos": 22601422321920.0,
      "grad_norm": 2.204810601250175,
      "language_loss": 0.81790155,
      "learning_rate": 1.3171551219599097e-07,
      "loss": 0.83994251,
      "num_input_tokens_seen": 159578265,
      "step": 7379,
      "time_per_iteration": 2.639313220977783
    },
    {
      "auxiliary_loss_clip": 0.01174079,
      "auxiliary_loss_mlp": 0.01024247,
      "balance_loss_clip": 1.05052078,
      "balance_loss_mlp": 1.017398,
      "epoch": 0.8873925329164913,
      "flos": 22163276223360.0,
      "grad_norm": 2.3223994692888743,
      "language_loss": 0.7779513,
      "learning_rate": 1.3143763821248377e-07,
      "loss": 0.79993451,
      "num_input_tokens_seen": 159595350,
      "step": 7380,
      "time_per_iteration": 2.63461971282959
    },
    {
      "auxiliary_loss_clip": 0.01171851,
      "auxiliary_loss_mlp": 0.01022837,
      "balance_loss_clip": 1.04929662,
      "balance_loss_mlp": 1.01648617,
      "epoch": 0.8875127758071304,
      "flos": 19208223204480.0,
      "grad_norm": 1.7570825118254785,
      "language_loss": 0.7223503,
      "learning_rate": 1.3116004768840118e-07,
      "loss": 0.74429721,
      "num_input_tokens_seen": 159613725,
      "step": 7381,
      "time_per_iteration": 2.708155393600464
    },
    {
      "auxiliary_loss_clip": 0.01174309,
      "auxiliary_loss_mlp": 0.01026945,
      "balance_loss_clip": 1.04928803,
      "balance_loss_mlp": 1.01997471,
      "epoch": 0.8876330186977694,
      "flos": 18110900666880.0,
      "grad_norm": 1.7228635281586109,
      "language_loss": 0.74278444,
      "learning_rate": 1.3088274066585348e-07,
      "loss": 0.76479697,
      "num_input_tokens_seen": 159631335,
      "step": 7382,
      "time_per_iteration": 2.6291143894195557
    },
    {
      "auxiliary_loss_clip": 0.01324168,
      "auxiliary_loss_mlp": 0.01028012,
      "balance_loss_clip": 1.04281139,
      "balance_loss_mlp": 1.0216608,
      "epoch": 0.8877532615884086,
      "flos": 22009434272640.0,
      "grad_norm": 2.397269711014476,
      "language_loss": 0.90645951,
      "learning_rate": 1.3060571718690749e-07,
      "loss": 0.92998135,
      "num_input_tokens_seen": 159648830,
      "step": 7383,
      "time_per_iteration": 2.706164836883545
    },
    {
      "auxiliary_loss_clip": 0.01223266,
      "auxiliary_loss_mlp": 0.02505433,
      "balance_loss_clip": 1.00686336,
      "balance_loss_mlp": 0.99991113,
      "epoch": 0.8878735044790477,
      "flos": 72136924346880.0,
      "grad_norm": 0.7478632655142272,
      "language_loss": 0.56885737,
      "learning_rate": 1.3032897729358805e-07,
      "loss": 0.60614437,
      "num_input_tokens_seen": 159709785,
      "step": 7384,
      "time_per_iteration": 3.3171682357788086
    },
    {
      "auxiliary_loss_clip": 0.0141067,
      "auxiliary_loss_mlp": 0.02568327,
      "balance_loss_clip": 1.03763485,
      "balance_loss_mlp": 0.99990785,
      "epoch": 0.8879937473696867,
      "flos": 27526355061120.0,
      "grad_norm": 1.9570256220478626,
      "language_loss": 0.79973578,
      "learning_rate": 1.3005252102787645e-07,
      "loss": 0.83952576,
      "num_input_tokens_seen": 159728725,
      "step": 7385,
      "time_per_iteration": 3.78452730178833
    },
    {
      "auxiliary_loss_clip": 0.01226156,
      "auxiliary_loss_mlp": 0.01024222,
      "balance_loss_clip": 1.04868722,
      "balance_loss_mlp": 1.01737905,
      "epoch": 0.8881139902603259,
      "flos": 22234091886720.0,
      "grad_norm": 1.6440704825572914,
      "language_loss": 0.73643637,
      "learning_rate": 1.297763484317105e-07,
      "loss": 0.7589401,
      "num_input_tokens_seen": 159747020,
      "step": 7386,
      "time_per_iteration": 2.780329704284668
    },
    {
      "auxiliary_loss_clip": 0.01270272,
      "auxiliary_loss_mlp": 0.02567712,
      "balance_loss_clip": 1.04007924,
      "balance_loss_mlp": 0.99988687,
      "epoch": 0.888234233150965,
      "flos": 20299548170880.0,
      "grad_norm": 2.2110640799479606,
      "language_loss": 0.70196527,
      "learning_rate": 1.2950045954698551e-07,
      "loss": 0.74034512,
      "num_input_tokens_seen": 159764855,
      "step": 7387,
      "time_per_iteration": 3.6594748497009277
    },
    {
      "auxiliary_loss_clip": 0.01314605,
      "auxiliary_loss_mlp": 0.01023852,
      "balance_loss_clip": 1.0441674,
      "balance_loss_mlp": 1.01731575,
      "epoch": 0.888354476041604,
      "flos": 18147996437760.0,
      "grad_norm": 1.5793746909982582,
      "language_loss": 0.75635505,
      "learning_rate": 1.2922485441555343e-07,
      "loss": 0.77973968,
      "num_input_tokens_seen": 159783935,
      "step": 7388,
      "time_per_iteration": 2.7026126384735107
    },
    {
      "auxiliary_loss_clip": 0.01168897,
      "auxiliary_loss_mlp": 0.01021112,
      "balance_loss_clip": 1.04615521,
      "balance_loss_mlp": 1.01436794,
      "epoch": 0.8884747189322432,
      "flos": 22014282608640.0,
      "grad_norm": 1.857981985745319,
      "language_loss": 0.81832492,
      "learning_rate": 1.2894953307922363e-07,
      "loss": 0.84022498,
      "num_input_tokens_seen": 159802895,
      "step": 7389,
      "time_per_iteration": 3.522310733795166
    },
    {
      "auxiliary_loss_clip": 0.01318562,
      "auxiliary_loss_mlp": 0.01025566,
      "balance_loss_clip": 1.04318261,
      "balance_loss_mlp": 1.01822257,
      "epoch": 0.8885949618228822,
      "flos": 19786779567360.0,
      "grad_norm": 1.8772549921267627,
      "language_loss": 0.83897412,
      "learning_rate": 1.2867449557976208e-07,
      "loss": 0.86241543,
      "num_input_tokens_seen": 159820995,
      "step": 7390,
      "time_per_iteration": 2.7546064853668213
    },
    {
      "auxiliary_loss_clip": 0.01219927,
      "auxiliary_loss_mlp": 0.01025838,
      "balance_loss_clip": 1.0475992,
      "balance_loss_mlp": 1.01885879,
      "epoch": 0.8887152047135213,
      "flos": 20047599198720.0,
      "grad_norm": 1.8307414463301817,
      "language_loss": 0.76026374,
      "learning_rate": 1.283997419588916e-07,
      "loss": 0.7827214,
      "num_input_tokens_seen": 159840465,
      "step": 7391,
      "time_per_iteration": 2.625218152999878
    },
    {
      "auxiliary_loss_clip": 0.01226953,
      "auxiliary_loss_mlp": 0.01024091,
      "balance_loss_clip": 1.04817867,
      "balance_loss_mlp": 1.01749301,
      "epoch": 0.8888354476041604,
      "flos": 18588117784320.0,
      "grad_norm": 1.853844851391151,
      "language_loss": 0.62224114,
      "learning_rate": 1.2812527225829216e-07,
      "loss": 0.64475155,
      "num_input_tokens_seen": 159858690,
      "step": 7392,
      "time_per_iteration": 3.48183274269104
    },
    {
      "auxiliary_loss_clip": 0.01233141,
      "auxiliary_loss_mlp": 0.01033556,
      "balance_loss_clip": 1.05200112,
      "balance_loss_mlp": 1.02604282,
      "epoch": 0.8889556904947995,
      "flos": 21689794120320.0,
      "grad_norm": 2.1583089960833193,
      "language_loss": 0.76693892,
      "learning_rate": 1.2785108651960052e-07,
      "loss": 0.78960586,
      "num_input_tokens_seen": 159880325,
      "step": 7393,
      "time_per_iteration": 2.651188850402832
    },
    {
      "auxiliary_loss_clip": 0.01226966,
      "auxiliary_loss_mlp": 0.01030258,
      "balance_loss_clip": 1.04833615,
      "balance_loss_mlp": 1.02361155,
      "epoch": 0.8890759333854386,
      "flos": 27381204201600.0,
      "grad_norm": 2.1210235880176023,
      "language_loss": 0.80867559,
      "learning_rate": 1.2757718478441094e-07,
      "loss": 0.83124781,
      "num_input_tokens_seen": 159901070,
      "step": 7394,
      "time_per_iteration": 2.7574827671051025
    },
    {
      "auxiliary_loss_clip": 0.01267276,
      "auxiliary_loss_mlp": 0.01023652,
      "balance_loss_clip": 1.04236984,
      "balance_loss_mlp": 1.01677024,
      "epoch": 0.8891961762760777,
      "flos": 24498834353280.0,
      "grad_norm": 1.6745013988603412,
      "language_loss": 0.77232307,
      "learning_rate": 1.2730356709427302e-07,
      "loss": 0.79523242,
      "num_input_tokens_seen": 159919750,
      "step": 7395,
      "time_per_iteration": 2.6885931491851807
    },
    {
      "auxiliary_loss_clip": 0.01227591,
      "auxiliary_loss_mlp": 0.01026653,
      "balance_loss_clip": 1.05237246,
      "balance_loss_mlp": 1.01975381,
      "epoch": 0.8893164191667168,
      "flos": 41499770895360.0,
      "grad_norm": 2.30470123456414,
      "language_loss": 0.59780079,
      "learning_rate": 1.2703023349069542e-07,
      "loss": 0.62034321,
      "num_input_tokens_seen": 159944600,
      "step": 7396,
      "time_per_iteration": 2.7803781032562256
    },
    {
      "auxiliary_loss_clip": 0.01219151,
      "auxiliary_loss_mlp": 0.01028198,
      "balance_loss_clip": 1.0474894,
      "balance_loss_mlp": 1.02140939,
      "epoch": 0.8894366620573558,
      "flos": 33583623120000.0,
      "grad_norm": 4.347554356809147,
      "language_loss": 0.61937535,
      "learning_rate": 1.2675718401514223e-07,
      "loss": 0.6418488,
      "num_input_tokens_seen": 159968780,
      "step": 7397,
      "time_per_iteration": 2.744286060333252
    },
    {
      "auxiliary_loss_clip": 0.01271171,
      "auxiliary_loss_mlp": 0.01024472,
      "balance_loss_clip": 1.04589295,
      "balance_loss_mlp": 1.01775777,
      "epoch": 0.889556904947995,
      "flos": 16909832672640.0,
      "grad_norm": 2.3748885280668177,
      "language_loss": 0.74425548,
      "learning_rate": 1.264844187090346e-07,
      "loss": 0.76721191,
      "num_input_tokens_seen": 159985905,
      "step": 7398,
      "time_per_iteration": 2.6506099700927734
    },
    {
      "auxiliary_loss_clip": 0.01270697,
      "auxiliary_loss_mlp": 0.01025497,
      "balance_loss_clip": 1.04484403,
      "balance_loss_mlp": 1.01882672,
      "epoch": 0.889677147838634,
      "flos": 26030855283840.0,
      "grad_norm": 2.940849952269287,
      "language_loss": 0.75230479,
      "learning_rate": 1.262119376137516e-07,
      "loss": 0.77526671,
      "num_input_tokens_seen": 160006965,
      "step": 7399,
      "time_per_iteration": 2.812908172607422
    },
    {
      "auxiliary_loss_clip": 0.01217014,
      "auxiliary_loss_mlp": 0.01023723,
      "balance_loss_clip": 1.04489338,
      "balance_loss_mlp": 1.01677871,
      "epoch": 0.8897973907292731,
      "flos": 26468283110400.0,
      "grad_norm": 1.8035758718627675,
      "language_loss": 0.8520726,
      "learning_rate": 1.2593974077062707e-07,
      "loss": 0.87447995,
      "num_input_tokens_seen": 160028585,
      "step": 7400,
      "time_per_iteration": 2.6679067611694336
    },
    {
      "auxiliary_loss_clip": 0.01313194,
      "auxiliary_loss_mlp": 0.01026012,
      "balance_loss_clip": 1.04267728,
      "balance_loss_mlp": 1.01901698,
      "epoch": 0.8899176336199123,
      "flos": 26249694894720.0,
      "grad_norm": 1.862212924776728,
      "language_loss": 0.63347864,
      "learning_rate": 1.2566782822095423e-07,
      "loss": 0.65687066,
      "num_input_tokens_seen": 160048840,
      "step": 7401,
      "time_per_iteration": 2.790428638458252
    },
    {
      "auxiliary_loss_clip": 0.01326654,
      "auxiliary_loss_mlp": 0.01026937,
      "balance_loss_clip": 1.0483135,
      "balance_loss_mlp": 1.02000189,
      "epoch": 0.8900378765105513,
      "flos": 20811742156800.0,
      "grad_norm": 2.303497908134964,
      "language_loss": 0.71178484,
      "learning_rate": 1.2539620000598162e-07,
      "loss": 0.73532069,
      "num_input_tokens_seen": 160068175,
      "step": 7402,
      "time_per_iteration": 2.827194929122925
    },
    {
      "auxiliary_loss_clip": 0.01170249,
      "auxiliary_loss_mlp": 0.01022435,
      "balance_loss_clip": 1.04714298,
      "balance_loss_mlp": 1.01524055,
      "epoch": 0.8901581194011904,
      "flos": 16472333018880.0,
      "grad_norm": 4.370343002639944,
      "language_loss": 0.7984035,
      "learning_rate": 1.2512485616691492e-07,
      "loss": 0.82033032,
      "num_input_tokens_seen": 160085230,
      "step": 7403,
      "time_per_iteration": 2.6182937622070312
    },
    {
      "auxiliary_loss_clip": 0.01324041,
      "auxiliary_loss_mlp": 0.01025883,
      "balance_loss_clip": 1.04316187,
      "balance_loss_mlp": 1.01836658,
      "epoch": 0.8902783622918296,
      "flos": 35155253773440.0,
      "grad_norm": 1.5065724449459545,
      "language_loss": 0.80750293,
      "learning_rate": 1.2485379674491681e-07,
      "loss": 0.83100218,
      "num_input_tokens_seen": 160111425,
      "step": 7404,
      "time_per_iteration": 2.7968783378601074
    },
    {
      "auxiliary_loss_clip": 0.01272329,
      "auxiliary_loss_mlp": 0.01025533,
      "balance_loss_clip": 1.04860485,
      "balance_loss_mlp": 1.01876199,
      "epoch": 0.8903986051824686,
      "flos": 17201068145280.0,
      "grad_norm": 2.5600525529138545,
      "language_loss": 0.79320669,
      "learning_rate": 1.2458302178110657e-07,
      "loss": 0.8161853,
      "num_input_tokens_seen": 160129790,
      "step": 7405,
      "time_per_iteration": 2.6930782794952393
    },
    {
      "auxiliary_loss_clip": 0.01310812,
      "auxiliary_loss_mlp": 0.01020861,
      "balance_loss_clip": 1.0416193,
      "balance_loss_mlp": 1.01431596,
      "epoch": 0.8905188480731077,
      "flos": 25483863997440.0,
      "grad_norm": 3.3962297712759426,
      "language_loss": 0.82558924,
      "learning_rate": 1.2431253131656118e-07,
      "loss": 0.84890592,
      "num_input_tokens_seen": 160149265,
      "step": 7406,
      "time_per_iteration": 2.6981492042541504
    },
    {
      "auxiliary_loss_clip": 0.01264509,
      "auxiliary_loss_mlp": 0.01022906,
      "balance_loss_clip": 1.04533482,
      "balance_loss_mlp": 1.01601005,
      "epoch": 0.8906390909637467,
      "flos": 23365888502400.0,
      "grad_norm": 1.9821885283209937,
      "language_loss": 0.7669971,
      "learning_rate": 1.240423253923133e-07,
      "loss": 0.78987134,
      "num_input_tokens_seen": 160168870,
      "step": 7407,
      "time_per_iteration": 2.655198574066162
    },
    {
      "auxiliary_loss_clip": 0.01224023,
      "auxiliary_loss_mlp": 0.01025153,
      "balance_loss_clip": 1.04646039,
      "balance_loss_mlp": 1.01833475,
      "epoch": 0.8907593338543859,
      "flos": 21068790860160.0,
      "grad_norm": 3.3860200820568456,
      "language_loss": 0.69736761,
      "learning_rate": 1.237724040493533e-07,
      "loss": 0.71985936,
      "num_input_tokens_seen": 160187495,
      "step": 7408,
      "time_per_iteration": 2.674344778060913
    },
    {
      "auxiliary_loss_clip": 0.0118061,
      "auxiliary_loss_mlp": 0.01032179,
      "balance_loss_clip": 1.05338907,
      "balance_loss_mlp": 1.02431107,
      "epoch": 0.8908795767450249,
      "flos": 21869562712320.0,
      "grad_norm": 3.883372888213518,
      "language_loss": 0.72724199,
      "learning_rate": 1.2350276732862773e-07,
      "loss": 0.74936992,
      "num_input_tokens_seen": 160208520,
      "step": 7409,
      "time_per_iteration": 2.6162221431732178
    },
    {
      "auxiliary_loss_clip": 0.0111539,
      "auxiliary_loss_mlp": 0.01000824,
      "balance_loss_clip": 1.00669837,
      "balance_loss_mlp": 0.99992961,
      "epoch": 0.890999819635664,
      "flos": 66307869348480.0,
      "grad_norm": 0.8361473658592499,
      "language_loss": 0.56688654,
      "learning_rate": 1.2323341527103993e-07,
      "loss": 0.58804864,
      "num_input_tokens_seen": 160263720,
      "step": 7410,
      "time_per_iteration": 3.1401209831237793
    },
    {
      "auxiliary_loss_clip": 0.01172418,
      "auxiliary_loss_mlp": 0.01020804,
      "balance_loss_clip": 1.04901159,
      "balance_loss_mlp": 1.014256,
      "epoch": 0.8911200625263032,
      "flos": 26869908055680.0,
      "grad_norm": 1.9486360456092378,
      "language_loss": 0.85221815,
      "learning_rate": 1.2296434791745135e-07,
      "loss": 0.8741504,
      "num_input_tokens_seen": 160282170,
      "step": 7411,
      "time_per_iteration": 3.4504992961883545
    },
    {
      "auxiliary_loss_clip": 0.01222463,
      "auxiliary_loss_mlp": 0.01024034,
      "balance_loss_clip": 1.0473907,
      "balance_loss_mlp": 1.01689315,
      "epoch": 0.8912403054169422,
      "flos": 20885825957760.0,
      "grad_norm": 1.7648082287906823,
      "language_loss": 0.76615512,
      "learning_rate": 1.2269556530867875e-07,
      "loss": 0.78862011,
      "num_input_tokens_seen": 160300725,
      "step": 7412,
      "time_per_iteration": 2.6642749309539795
    },
    {
      "auxiliary_loss_clip": 0.01178127,
      "auxiliary_loss_mlp": 0.01032211,
      "balance_loss_clip": 1.05094457,
      "balance_loss_mlp": 1.02477527,
      "epoch": 0.8913605483075813,
      "flos": 27016567286400.0,
      "grad_norm": 1.9456588301723987,
      "language_loss": 0.82048386,
      "learning_rate": 1.2242706748549614e-07,
      "loss": 0.84258723,
      "num_input_tokens_seen": 160318720,
      "step": 7413,
      "time_per_iteration": 3.5924248695373535
    },
    {
      "auxiliary_loss_clip": 0.01273805,
      "auxiliary_loss_mlp": 0.01023415,
      "balance_loss_clip": 1.04250765,
      "balance_loss_mlp": 1.01646769,
      "epoch": 0.8914807911982204,
      "flos": 23621500661760.0,
      "grad_norm": 1.8760207925346373,
      "language_loss": 0.82348728,
      "learning_rate": 1.2215885448863473e-07,
      "loss": 0.84645951,
      "num_input_tokens_seen": 160339595,
      "step": 7414,
      "time_per_iteration": 2.6440443992614746
    },
    {
      "auxiliary_loss_clip": 0.01269853,
      "auxiliary_loss_mlp": 0.01029593,
      "balance_loss_clip": 1.04681301,
      "balance_loss_mlp": 1.02293777,
      "epoch": 0.8916010340888595,
      "flos": 24462277286400.0,
      "grad_norm": 1.9237207411845587,
      "language_loss": 0.80587387,
      "learning_rate": 1.2189092635878152e-07,
      "loss": 0.82886839,
      "num_input_tokens_seen": 160361045,
      "step": 7415,
      "time_per_iteration": 3.5700223445892334
    },
    {
      "auxiliary_loss_clip": 0.01312938,
      "auxiliary_loss_mlp": 0.01023724,
      "balance_loss_clip": 1.04140592,
      "balance_loss_mlp": 1.01643741,
      "epoch": 0.8917212769794985,
      "flos": 21215773313280.0,
      "grad_norm": 1.638336111605152,
      "language_loss": 0.77326506,
      "learning_rate": 1.216232831365822e-07,
      "loss": 0.79663169,
      "num_input_tokens_seen": 160379990,
      "step": 7416,
      "time_per_iteration": 2.6726181507110596
    },
    {
      "auxiliary_loss_clip": 0.0128123,
      "auxiliary_loss_mlp": 0.01030351,
      "balance_loss_clip": 1.04786587,
      "balance_loss_mlp": 1.02295673,
      "epoch": 0.8918415198701377,
      "flos": 25513992529920.0,
      "grad_norm": 1.7894264856593751,
      "language_loss": 0.80885303,
      "learning_rate": 1.2135592486263678e-07,
      "loss": 0.83196884,
      "num_input_tokens_seen": 160399240,
      "step": 7417,
      "time_per_iteration": 2.746429443359375
    },
    {
      "auxiliary_loss_clip": 0.01269303,
      "auxiliary_loss_mlp": 0.01025822,
      "balance_loss_clip": 1.04512143,
      "balance_loss_mlp": 1.01934338,
      "epoch": 0.8919617627607768,
      "flos": 37853006693760.0,
      "grad_norm": 2.08542561000155,
      "language_loss": 0.61509508,
      "learning_rate": 1.2108885157750415e-07,
      "loss": 0.63804632,
      "num_input_tokens_seen": 160421600,
      "step": 7418,
      "time_per_iteration": 3.74702787399292
    },
    {
      "auxiliary_loss_clip": 0.01314252,
      "auxiliary_loss_mlp": 0.02564575,
      "balance_loss_clip": 1.04745388,
      "balance_loss_mlp": 0.99989754,
      "epoch": 0.8920820056514158,
      "flos": 26213676531840.0,
      "grad_norm": 1.8798390599378836,
      "language_loss": 0.80302495,
      "learning_rate": 1.2082206332169897e-07,
      "loss": 0.84181321,
      "num_input_tokens_seen": 160441695,
      "step": 7419,
      "time_per_iteration": 2.6902048587799072
    },
    {
      "auxiliary_loss_clip": 0.01267332,
      "auxiliary_loss_mlp": 0.01025504,
      "balance_loss_clip": 1.04717016,
      "balance_loss_mlp": 1.01877463,
      "epoch": 0.892202248542055,
      "flos": 17383135207680.0,
      "grad_norm": 4.467434391869748,
      "language_loss": 0.73576033,
      "learning_rate": 1.2055556013569225e-07,
      "loss": 0.75868869,
      "num_input_tokens_seen": 160457205,
      "step": 7420,
      "time_per_iteration": 2.6753575801849365
    },
    {
      "auxiliary_loss_clip": 0.01170964,
      "auxiliary_loss_mlp": 0.01026626,
      "balance_loss_clip": 1.04681921,
      "balance_loss_mlp": 1.01980174,
      "epoch": 0.892322491432694,
      "flos": 21324223451520.0,
      "grad_norm": 1.699026548396117,
      "language_loss": 0.82066101,
      "learning_rate": 1.2028934205991315e-07,
      "loss": 0.84263688,
      "num_input_tokens_seen": 160476525,
      "step": 7421,
      "time_per_iteration": 2.681149482727051
    },
    {
      "auxiliary_loss_clip": 0.01221524,
      "auxiliary_loss_mlp": 0.01022456,
      "balance_loss_clip": 1.04450989,
      "balance_loss_mlp": 1.01566696,
      "epoch": 0.8924427343233331,
      "flos": 24029374573440.0,
      "grad_norm": 1.703928870901403,
      "language_loss": 0.76672852,
      "learning_rate": 1.2002340913474607e-07,
      "loss": 0.78916824,
      "num_input_tokens_seen": 160500160,
      "step": 7422,
      "time_per_iteration": 2.721454381942749
    },
    {
      "auxiliary_loss_clip": 0.01177081,
      "auxiliary_loss_mlp": 0.01023707,
      "balance_loss_clip": 1.05048084,
      "balance_loss_mlp": 1.0166322,
      "epoch": 0.8925629772139723,
      "flos": 30008069631360.0,
      "grad_norm": 2.8019508743824373,
      "language_loss": 0.74046415,
      "learning_rate": 1.1975776140053317e-07,
      "loss": 0.76247203,
      "num_input_tokens_seen": 160520130,
      "step": 7423,
      "time_per_iteration": 2.637305736541748
    },
    {
      "auxiliary_loss_clip": 0.01370897,
      "auxiliary_loss_mlp": 0.01028989,
      "balance_loss_clip": 1.04411769,
      "balance_loss_mlp": 1.02126408,
      "epoch": 0.8926832201046113,
      "flos": 22601709630720.0,
      "grad_norm": 2.319736781769992,
      "language_loss": 0.73782921,
      "learning_rate": 1.194923988975729e-07,
      "loss": 0.76182806,
      "num_input_tokens_seen": 160539730,
      "step": 7424,
      "time_per_iteration": 2.7085745334625244
    },
    {
      "auxiliary_loss_clip": 0.01315229,
      "auxiliary_loss_mlp": 0.01023509,
      "balance_loss_clip": 1.04383707,
      "balance_loss_mlp": 1.01618361,
      "epoch": 0.8928034629952504,
      "flos": 13297722117120.0,
      "grad_norm": 3.8024022098676156,
      "language_loss": 0.7325964,
      "learning_rate": 1.192273216661206e-07,
      "loss": 0.75598377,
      "num_input_tokens_seen": 160557820,
      "step": 7425,
      "time_per_iteration": 2.649780511856079
    },
    {
      "auxiliary_loss_clip": 0.01326143,
      "auxiliary_loss_mlp": 0.00999753,
      "balance_loss_clip": 1.00782645,
      "balance_loss_mlp": 0.99890107,
      "epoch": 0.8929237058858895,
      "flos": 54854556744960.0,
      "grad_norm": 0.766753728525189,
      "language_loss": 0.57428825,
      "learning_rate": 1.189625297463881e-07,
      "loss": 0.59754723,
      "num_input_tokens_seen": 160619510,
      "step": 7426,
      "time_per_iteration": 3.362109661102295
    },
    {
      "auxiliary_loss_clip": 0.01417386,
      "auxiliary_loss_mlp": 0.01024467,
      "balance_loss_clip": 1.03876984,
      "balance_loss_mlp": 1.01738906,
      "epoch": 0.8930439487765286,
      "flos": 28883850785280.0,
      "grad_norm": 2.0346585891092484,
      "language_loss": 0.79946548,
      "learning_rate": 1.1869802317854394e-07,
      "loss": 0.82388401,
      "num_input_tokens_seen": 160643295,
      "step": 7427,
      "time_per_iteration": 3.2331345081329346
    },
    {
      "auxiliary_loss_clip": 0.01370542,
      "auxiliary_loss_mlp": 0.010323,
      "balance_loss_clip": 1.04273367,
      "balance_loss_mlp": 1.02475119,
      "epoch": 0.8931641916671677,
      "flos": 22419283432320.0,
      "grad_norm": 2.0801419532988135,
      "language_loss": 0.72116125,
      "learning_rate": 1.1843380200271425e-07,
      "loss": 0.74518967,
      "num_input_tokens_seen": 160662495,
      "step": 7428,
      "time_per_iteration": 2.7760350704193115
    },
    {
      "auxiliary_loss_clip": 0.01218198,
      "auxiliary_loss_mlp": 0.01021527,
      "balance_loss_clip": 1.0446049,
      "balance_loss_mlp": 1.01488113,
      "epoch": 0.8932844345578068,
      "flos": 25843149786240.0,
      "grad_norm": 1.983653176616118,
      "language_loss": 0.80606115,
      "learning_rate": 1.181698662589805e-07,
      "loss": 0.82845843,
      "num_input_tokens_seen": 160682080,
      "step": 7429,
      "time_per_iteration": 2.7385826110839844
    },
    {
      "auxiliary_loss_clip": 0.01221169,
      "auxiliary_loss_mlp": 0.01027838,
      "balance_loss_clip": 1.04786956,
      "balance_loss_mlp": 1.02086425,
      "epoch": 0.8934046774484459,
      "flos": 22925803069440.0,
      "grad_norm": 2.4862890489443203,
      "language_loss": 0.76037651,
      "learning_rate": 1.1790621598738249e-07,
      "loss": 0.7828666,
      "num_input_tokens_seen": 160700395,
      "step": 7430,
      "time_per_iteration": 2.606382131576538
    },
    {
      "auxiliary_loss_clip": 0.01168182,
      "auxiliary_loss_mlp": 0.01026161,
      "balance_loss_clip": 1.04766989,
      "balance_loss_mlp": 1.01964951,
      "epoch": 0.8935249203390849,
      "flos": 24462097718400.0,
      "grad_norm": 1.950574595682022,
      "language_loss": 0.74693412,
      "learning_rate": 1.1764285122791461e-07,
      "loss": 0.76887751,
      "num_input_tokens_seen": 160721115,
      "step": 7431,
      "time_per_iteration": 2.6603198051452637
    },
    {
      "auxiliary_loss_clip": 0.01221331,
      "auxiliary_loss_mlp": 0.0102677,
      "balance_loss_clip": 1.04529023,
      "balance_loss_mlp": 1.01995122,
      "epoch": 0.8936451632297241,
      "flos": 15742735966080.0,
      "grad_norm": 2.09888445231494,
      "language_loss": 0.76835513,
      "learning_rate": 1.173797720205294e-07,
      "loss": 0.79083622,
      "num_input_tokens_seen": 160739150,
      "step": 7432,
      "time_per_iteration": 2.595998764038086
    },
    {
      "auxiliary_loss_clip": 0.01223638,
      "auxiliary_loss_mlp": 0.01024068,
      "balance_loss_clip": 1.04829931,
      "balance_loss_mlp": 1.01697826,
      "epoch": 0.8937654061203631,
      "flos": 35115500396160.0,
      "grad_norm": 2.735672757818518,
      "language_loss": 0.71787536,
      "learning_rate": 1.1711697840513602e-07,
      "loss": 0.74035239,
      "num_input_tokens_seen": 160758585,
      "step": 7433,
      "time_per_iteration": 2.7270569801330566
    },
    {
      "auxiliary_loss_clip": 0.01215977,
      "auxiliary_loss_mlp": 0.01020575,
      "balance_loss_clip": 1.04396331,
      "balance_loss_mlp": 1.01388431,
      "epoch": 0.8938856490110022,
      "flos": 16107444708480.0,
      "grad_norm": 3.200135976877204,
      "language_loss": 0.71203959,
      "learning_rate": 1.1685447042160012e-07,
      "loss": 0.73440516,
      "num_input_tokens_seen": 160776620,
      "step": 7434,
      "time_per_iteration": 2.7879831790924072
    },
    {
      "auxiliary_loss_clip": 0.01176022,
      "auxiliary_loss_mlp": 0.01023175,
      "balance_loss_clip": 1.04978931,
      "balance_loss_mlp": 1.01614475,
      "epoch": 0.8940058919016414,
      "flos": 20704189858560.0,
      "grad_norm": 1.7274605621155177,
      "language_loss": 0.71718419,
      "learning_rate": 1.1659224810974367e-07,
      "loss": 0.73917615,
      "num_input_tokens_seen": 160796580,
      "step": 7435,
      "time_per_iteration": 2.6253976821899414
    },
    {
      "auxiliary_loss_clip": 0.01268511,
      "auxiliary_loss_mlp": 0.01024557,
      "balance_loss_clip": 1.04817533,
      "balance_loss_mlp": 1.01754212,
      "epoch": 0.8941261347922804,
      "flos": 25229041937280.0,
      "grad_norm": 1.4708108284795507,
      "language_loss": 0.68520498,
      "learning_rate": 1.1633031150934591e-07,
      "loss": 0.70813566,
      "num_input_tokens_seen": 160819610,
      "step": 7436,
      "time_per_iteration": 2.757740020751953
    },
    {
      "auxiliary_loss_clip": 0.01222455,
      "auxiliary_loss_mlp": 0.01031542,
      "balance_loss_clip": 1.04774427,
      "balance_loss_mlp": 1.02439213,
      "epoch": 0.8942463776829195,
      "flos": 19537236806400.0,
      "grad_norm": 1.8880088633221186,
      "language_loss": 0.80113721,
      "learning_rate": 1.1606866066014176e-07,
      "loss": 0.82367718,
      "num_input_tokens_seen": 160838660,
      "step": 7437,
      "time_per_iteration": 3.513551712036133
    },
    {
      "auxiliary_loss_clip": 0.01316803,
      "auxiliary_loss_mlp": 0.01025939,
      "balance_loss_clip": 1.04494524,
      "balance_loss_mlp": 1.01954019,
      "epoch": 0.8943666205735585,
      "flos": 22301567585280.0,
      "grad_norm": 2.6227735301043844,
      "language_loss": 0.75460011,
      "learning_rate": 1.1580729560182434e-07,
      "loss": 0.77802753,
      "num_input_tokens_seen": 160854515,
      "step": 7438,
      "time_per_iteration": 2.679056406021118
    },
    {
      "auxiliary_loss_clip": 0.01172035,
      "auxiliary_loss_mlp": 0.02563212,
      "balance_loss_clip": 1.04852223,
      "balance_loss_mlp": 0.99990857,
      "epoch": 0.8944868634641977,
      "flos": 18912893581440.0,
      "grad_norm": 1.8291268290991458,
      "language_loss": 0.71089709,
      "learning_rate": 1.1554621637404171e-07,
      "loss": 0.74824953,
      "num_input_tokens_seen": 160872605,
      "step": 7439,
      "time_per_iteration": 3.6175131797790527
    },
    {
      "auxiliary_loss_clip": 0.01221475,
      "auxiliary_loss_mlp": 0.0102661,
      "balance_loss_clip": 1.04681945,
      "balance_loss_mlp": 1.0192728,
      "epoch": 0.8946071063548368,
      "flos": 14460904241280.0,
      "grad_norm": 3.811906606866283,
      "language_loss": 0.61113703,
      "learning_rate": 1.1528542301639999e-07,
      "loss": 0.63361788,
      "num_input_tokens_seen": 160889395,
      "step": 7440,
      "time_per_iteration": 2.655818223953247
    },
    {
      "auxiliary_loss_clip": 0.0131884,
      "auxiliary_loss_mlp": 0.0102258,
      "balance_loss_clip": 1.04246283,
      "balance_loss_mlp": 1.01547503,
      "epoch": 0.8947273492454758,
      "flos": 20084084438400.0,
      "grad_norm": 3.158675711061776,
      "language_loss": 0.82595748,
      "learning_rate": 1.1502491556846105e-07,
      "loss": 0.84937167,
      "num_input_tokens_seen": 160907890,
      "step": 7441,
      "time_per_iteration": 3.508023977279663
    },
    {
      "auxiliary_loss_clip": 0.01267541,
      "auxiliary_loss_mlp": 0.01022525,
      "balance_loss_clip": 1.04558384,
      "balance_loss_mlp": 1.01584959,
      "epoch": 0.894847592136115,
      "flos": 18550555136640.0,
      "grad_norm": 2.2824985270352522,
      "language_loss": 0.81598788,
      "learning_rate": 1.1476469406974331e-07,
      "loss": 0.83888853,
      "num_input_tokens_seen": 160923490,
      "step": 7442,
      "time_per_iteration": 2.6619369983673096
    },
    {
      "auxiliary_loss_clip": 0.01171449,
      "auxiliary_loss_mlp": 0.01022726,
      "balance_loss_clip": 1.04948092,
      "balance_loss_mlp": 1.01604474,
      "epoch": 0.894967835026754,
      "flos": 23478468704640.0,
      "grad_norm": 1.8470569852025782,
      "language_loss": 0.7680034,
      "learning_rate": 1.1450475855972341e-07,
      "loss": 0.78994513,
      "num_input_tokens_seen": 160944280,
      "step": 7443,
      "time_per_iteration": 2.6106791496276855
    },
    {
      "auxiliary_loss_clip": 0.01270434,
      "auxiliary_loss_mlp": 0.02565798,
      "balance_loss_clip": 1.04399228,
      "balance_loss_mlp": 0.99991864,
      "epoch": 0.8950880779173931,
      "flos": 15188310564480.0,
      "grad_norm": 2.872988061616327,
      "language_loss": 0.70730484,
      "learning_rate": 1.1424510907783158e-07,
      "loss": 0.74566716,
      "num_input_tokens_seen": 160961560,
      "step": 7444,
      "time_per_iteration": 3.613391399383545
    },
    {
      "auxiliary_loss_clip": 0.0127495,
      "auxiliary_loss_mlp": 0.01023224,
      "balance_loss_clip": 1.04331303,
      "balance_loss_mlp": 1.01696849,
      "epoch": 0.8952083208080323,
      "flos": 22091957769600.0,
      "grad_norm": 1.8258074210867783,
      "language_loss": 0.83136523,
      "learning_rate": 1.1398574566345787e-07,
      "loss": 0.85434699,
      "num_input_tokens_seen": 160982195,
      "step": 7445,
      "time_per_iteration": 2.8397269248962402
    },
    {
      "auxiliary_loss_clip": 0.01274431,
      "auxiliary_loss_mlp": 0.01024811,
      "balance_loss_clip": 1.04286206,
      "balance_loss_mlp": 1.01725268,
      "epoch": 0.8953285636986713,
      "flos": 23254026572160.0,
      "grad_norm": 2.0306358963170177,
      "language_loss": 0.82266021,
      "learning_rate": 1.1372666835594702e-07,
      "loss": 0.84565258,
      "num_input_tokens_seen": 161000520,
      "step": 7446,
      "time_per_iteration": 2.719120740890503
    },
    {
      "auxiliary_loss_clip": 0.01267771,
      "auxiliary_loss_mlp": 0.01021069,
      "balance_loss_clip": 1.04481506,
      "balance_loss_mlp": 1.01432824,
      "epoch": 0.8954488065893104,
      "flos": 16362661818240.0,
      "grad_norm": 2.3062669365785657,
      "language_loss": 0.71871161,
      "learning_rate": 1.1346787719460071e-07,
      "loss": 0.74160004,
      "num_input_tokens_seen": 161019405,
      "step": 7447,
      "time_per_iteration": 2.6243927478790283
    },
    {
      "auxiliary_loss_clip": 0.01263673,
      "auxiliary_loss_mlp": 0.01023111,
      "balance_loss_clip": 1.0434978,
      "balance_loss_mlp": 1.01615214,
      "epoch": 0.8955690494799495,
      "flos": 18257883120000.0,
      "grad_norm": 1.8309592158875168,
      "language_loss": 0.72581756,
      "learning_rate": 1.1320937221867732e-07,
      "loss": 0.74868536,
      "num_input_tokens_seen": 161036985,
      "step": 7448,
      "time_per_iteration": 2.708491563796997
    },
    {
      "auxiliary_loss_clip": 0.01271219,
      "auxiliary_loss_mlp": 0.01024573,
      "balance_loss_clip": 1.04415321,
      "balance_loss_mlp": 1.01870489,
      "epoch": 0.8956892923705886,
      "flos": 25447486498560.0,
      "grad_norm": 1.859633369194452,
      "language_loss": 0.79760867,
      "learning_rate": 1.1295115346739192e-07,
      "loss": 0.82056659,
      "num_input_tokens_seen": 161056985,
      "step": 7449,
      "time_per_iteration": 2.71567964553833
    },
    {
      "auxiliary_loss_clip": 0.01275894,
      "auxiliary_loss_mlp": 0.01025998,
      "balance_loss_clip": 1.04665494,
      "balance_loss_mlp": 1.01846981,
      "epoch": 0.8958095352612276,
      "flos": 52661883939840.0,
      "grad_norm": 2.7568388437982865,
      "language_loss": 0.72908819,
      "learning_rate": 1.1269322097991629e-07,
      "loss": 0.75210714,
      "num_input_tokens_seen": 161080270,
      "step": 7450,
      "time_per_iteration": 2.920205593109131
    },
    {
      "auxiliary_loss_clip": 0.01226686,
      "auxiliary_loss_mlp": 0.01024164,
      "balance_loss_clip": 1.04909861,
      "balance_loss_mlp": 1.01661849,
      "epoch": 0.8959297781518668,
      "flos": 23186335392000.0,
      "grad_norm": 3.3742909957778724,
      "language_loss": 0.68125933,
      "learning_rate": 1.1243557479537846e-07,
      "loss": 0.7037679,
      "num_input_tokens_seen": 161100160,
      "step": 7451,
      "time_per_iteration": 2.64849853515625
    },
    {
      "auxiliary_loss_clip": 0.01171413,
      "auxiliary_loss_mlp": 0.01026197,
      "balance_loss_clip": 1.04722667,
      "balance_loss_mlp": 1.01920867,
      "epoch": 0.8960500210425059,
      "flos": 20334309557760.0,
      "grad_norm": 2.1579578568283626,
      "language_loss": 0.68830919,
      "learning_rate": 1.121782149528634e-07,
      "loss": 0.71028537,
      "num_input_tokens_seen": 161117260,
      "step": 7452,
      "time_per_iteration": 2.588974714279175
    },
    {
      "auxiliary_loss_clip": 0.0118148,
      "auxiliary_loss_mlp": 0.01025433,
      "balance_loss_clip": 1.05143917,
      "balance_loss_mlp": 1.01872802,
      "epoch": 0.8961702639331449,
      "flos": 19901694153600.0,
      "grad_norm": 2.066919308968981,
      "language_loss": 0.7885235,
      "learning_rate": 1.1192114149141208e-07,
      "loss": 0.81059265,
      "num_input_tokens_seen": 161136895,
      "step": 7453,
      "time_per_iteration": 2.738297700881958
    },
    {
      "auxiliary_loss_clip": 0.01272996,
      "auxiliary_loss_mlp": 0.01025128,
      "balance_loss_clip": 1.04311335,
      "balance_loss_mlp": 1.0181067,
      "epoch": 0.8962905068237841,
      "flos": 12896348567040.0,
      "grad_norm": 2.9272641775470047,
      "language_loss": 0.65946448,
      "learning_rate": 1.1166435445002197e-07,
      "loss": 0.6824457,
      "num_input_tokens_seen": 161154565,
      "step": 7454,
      "time_per_iteration": 2.6650853157043457
    },
    {
      "auxiliary_loss_clip": 0.01226697,
      "auxiliary_loss_mlp": 0.01028852,
      "balance_loss_clip": 1.05113947,
      "balance_loss_mlp": 1.02147293,
      "epoch": 0.8964107497144231,
      "flos": 23440331439360.0,
      "grad_norm": 2.1144603638334383,
      "language_loss": 0.68307322,
      "learning_rate": 1.1140785386764818e-07,
      "loss": 0.70562875,
      "num_input_tokens_seen": 161173265,
      "step": 7455,
      "time_per_iteration": 2.694751024246216
    },
    {
      "auxiliary_loss_clip": 0.0121878,
      "auxiliary_loss_mlp": 0.01026497,
      "balance_loss_clip": 1.04538798,
      "balance_loss_mlp": 1.01955903,
      "epoch": 0.8965309926050622,
      "flos": 19500176949120.0,
      "grad_norm": 2.273352555165449,
      "language_loss": 0.69960946,
      "learning_rate": 1.1115163978320153e-07,
      "loss": 0.72206223,
      "num_input_tokens_seen": 161191995,
      "step": 7456,
      "time_per_iteration": 2.660167932510376
    },
    {
      "auxiliary_loss_clip": 0.01228694,
      "auxiliary_loss_mlp": 0.02567451,
      "balance_loss_clip": 1.04931855,
      "balance_loss_mlp": 0.99991131,
      "epoch": 0.8966512354957014,
      "flos": 28658008022400.0,
      "grad_norm": 2.0431072746153283,
      "language_loss": 0.82509387,
      "learning_rate": 1.1089571223554917e-07,
      "loss": 0.86305535,
      "num_input_tokens_seen": 161212880,
      "step": 7457,
      "time_per_iteration": 2.7142207622528076
    },
    {
      "auxiliary_loss_clip": 0.01222832,
      "auxiliary_loss_mlp": 0.01026911,
      "balance_loss_clip": 1.0447551,
      "balance_loss_mlp": 1.02036095,
      "epoch": 0.8967714783863404,
      "flos": 23370916406400.0,
      "grad_norm": 1.946244636229548,
      "language_loss": 0.85356247,
      "learning_rate": 1.1064007126351537e-07,
      "loss": 0.87605989,
      "num_input_tokens_seen": 161233595,
      "step": 7458,
      "time_per_iteration": 2.6650357246398926
    },
    {
      "auxiliary_loss_clip": 0.01268553,
      "auxiliary_loss_mlp": 0.01029939,
      "balance_loss_clip": 1.04831374,
      "balance_loss_mlp": 1.02261615,
      "epoch": 0.8968917212769795,
      "flos": 24535175938560.0,
      "grad_norm": 2.271810247489334,
      "language_loss": 0.76382416,
      "learning_rate": 1.1038471690588003e-07,
      "loss": 0.78680909,
      "num_input_tokens_seen": 161252740,
      "step": 7459,
      "time_per_iteration": 2.730560541152954
    },
    {
      "auxiliary_loss_clip": 0.01376834,
      "auxiliary_loss_mlp": 0.01024605,
      "balance_loss_clip": 1.04808187,
      "balance_loss_mlp": 1.01780701,
      "epoch": 0.8970119641676186,
      "flos": 23475416048640.0,
      "grad_norm": 2.1198183809265445,
      "language_loss": 0.79906881,
      "learning_rate": 1.1012964920138145e-07,
      "loss": 0.82308328,
      "num_input_tokens_seen": 161272325,
      "step": 7460,
      "time_per_iteration": 2.7862727642059326
    },
    {
      "auxiliary_loss_clip": 0.0126908,
      "auxiliary_loss_mlp": 0.01025106,
      "balance_loss_clip": 1.04215586,
      "balance_loss_mlp": 1.01826978,
      "epoch": 0.8971322070582577,
      "flos": 24538192680960.0,
      "grad_norm": 1.99096799982235,
      "language_loss": 0.759341,
      "learning_rate": 1.0987486818871205e-07,
      "loss": 0.78228283,
      "num_input_tokens_seen": 161295915,
      "step": 7461,
      "time_per_iteration": 2.7263381481170654
    },
    {
      "auxiliary_loss_clip": 0.01224118,
      "auxiliary_loss_mlp": 0.02568371,
      "balance_loss_clip": 1.05024731,
      "balance_loss_mlp": 0.99993587,
      "epoch": 0.8972524499488967,
      "flos": 21797454159360.0,
      "grad_norm": 2.2555295089053096,
      "language_loss": 0.73140419,
      "learning_rate": 1.0962037390652245e-07,
      "loss": 0.76932907,
      "num_input_tokens_seen": 161314935,
      "step": 7462,
      "time_per_iteration": 2.6176726818084717
    },
    {
      "auxiliary_loss_clip": 0.01279495,
      "auxiliary_loss_mlp": 0.01023731,
      "balance_loss_clip": 1.05000925,
      "balance_loss_mlp": 1.01652527,
      "epoch": 0.8973726928395359,
      "flos": 21726243446400.0,
      "grad_norm": 1.865926534015747,
      "language_loss": 0.71854395,
      "learning_rate": 1.0936616639341911e-07,
      "loss": 0.74157619,
      "num_input_tokens_seen": 161335225,
      "step": 7463,
      "time_per_iteration": 3.650050401687622
    },
    {
      "auxiliary_loss_clip": 0.0110757,
      "auxiliary_loss_mlp": 0.01002146,
      "balance_loss_clip": 1.00658059,
      "balance_loss_mlp": 1.00127554,
      "epoch": 0.897492935730175,
      "flos": 53837100097920.0,
      "grad_norm": 0.7515293860408219,
      "language_loss": 0.5469805,
      "learning_rate": 1.0911224568796473e-07,
      "loss": 0.56807768,
      "num_input_tokens_seen": 161393420,
      "step": 7464,
      "time_per_iteration": 3.2196896076202393
    },
    {
      "auxiliary_loss_clip": 0.01220109,
      "auxiliary_loss_mlp": 0.01028935,
      "balance_loss_clip": 1.04888582,
      "balance_loss_mlp": 1.02193773,
      "epoch": 0.897613178620814,
      "flos": 18290346036480.0,
      "grad_norm": 2.105001595506374,
      "language_loss": 0.7086823,
      "learning_rate": 1.0885861182867984e-07,
      "loss": 0.73117268,
      "num_input_tokens_seen": 161411525,
      "step": 7465,
      "time_per_iteration": 3.6072964668273926
    },
    {
      "auxiliary_loss_clip": 0.01274021,
      "auxiliary_loss_mlp": 0.01023225,
      "balance_loss_clip": 1.0458889,
      "balance_loss_mlp": 1.0162816,
      "epoch": 0.8977334215114532,
      "flos": 32993718059520.0,
      "grad_norm": 6.704834944609961,
      "language_loss": 0.70965993,
      "learning_rate": 1.0860526485403942e-07,
      "loss": 0.7326324,
      "num_input_tokens_seen": 161432800,
      "step": 7466,
      "time_per_iteration": 2.7508723735809326
    },
    {
      "auxiliary_loss_clip": 0.01172502,
      "auxiliary_loss_mlp": 0.01025707,
      "balance_loss_clip": 1.04863405,
      "balance_loss_mlp": 1.01880217,
      "epoch": 0.8978536644020922,
      "flos": 15195636938880.0,
      "grad_norm": 1.853415290471086,
      "language_loss": 0.7722007,
      "learning_rate": 1.0835220480247675e-07,
      "loss": 0.79418278,
      "num_input_tokens_seen": 161451295,
      "step": 7467,
      "time_per_iteration": 3.4982364177703857
    },
    {
      "auxiliary_loss_clip": 0.01268937,
      "auxiliary_loss_mlp": 0.01031744,
      "balance_loss_clip": 1.04581821,
      "balance_loss_mlp": 1.024544,
      "epoch": 0.8979739072927313,
      "flos": 18004389863040.0,
      "grad_norm": 2.6024824638014614,
      "language_loss": 0.83871281,
      "learning_rate": 1.0809943171238067e-07,
      "loss": 0.86171961,
      "num_input_tokens_seen": 161469220,
      "step": 7468,
      "time_per_iteration": 2.6138577461242676
    },
    {
      "auxiliary_loss_clip": 0.01280853,
      "auxiliary_loss_mlp": 0.01029097,
      "balance_loss_clip": 1.04820573,
      "balance_loss_mlp": 1.02118993,
      "epoch": 0.8980941501833704,
      "flos": 22271546793600.0,
      "grad_norm": 2.6663712577394048,
      "language_loss": 0.63135219,
      "learning_rate": 1.078469456220965e-07,
      "loss": 0.65445167,
      "num_input_tokens_seen": 161489375,
      "step": 7469,
      "time_per_iteration": 2.682008981704712
    },
    {
      "auxiliary_loss_clip": 0.01122736,
      "auxiliary_loss_mlp": 0.01027548,
      "balance_loss_clip": 1.0467689,
      "balance_loss_mlp": 1.02028561,
      "epoch": 0.8982143930740095,
      "flos": 37560729726720.0,
      "grad_norm": 1.8064353644501332,
      "language_loss": 0.69543916,
      "learning_rate": 1.0759474656992606e-07,
      "loss": 0.71694201,
      "num_input_tokens_seen": 161512145,
      "step": 7470,
      "time_per_iteration": 3.6774489879608154
    },
    {
      "auxiliary_loss_clip": 0.01274846,
      "auxiliary_loss_mlp": 0.01024037,
      "balance_loss_clip": 1.0433954,
      "balance_loss_mlp": 1.01627624,
      "epoch": 0.8983346359646486,
      "flos": 18076893465600.0,
      "grad_norm": 2.5389913183402952,
      "language_loss": 0.77677435,
      "learning_rate": 1.0734283459412785e-07,
      "loss": 0.7997632,
      "num_input_tokens_seen": 161528995,
      "step": 7471,
      "time_per_iteration": 2.635111093521118
    },
    {
      "auxiliary_loss_clip": 0.01374277,
      "auxiliary_loss_mlp": 0.01022286,
      "balance_loss_clip": 1.04308391,
      "balance_loss_mlp": 1.0146805,
      "epoch": 0.8984548788552876,
      "flos": 20558895344640.0,
      "grad_norm": 1.634634097166387,
      "language_loss": 0.80165792,
      "learning_rate": 1.0709120973291707e-07,
      "loss": 0.82562363,
      "num_input_tokens_seen": 161548775,
      "step": 7472,
      "time_per_iteration": 2.7310078144073486
    },
    {
      "auxiliary_loss_clip": 0.01174979,
      "auxiliary_loss_mlp": 0.01027048,
      "balance_loss_clip": 1.0497508,
      "balance_loss_mlp": 1.02002656,
      "epoch": 0.8985751217459268,
      "flos": 17785442511360.0,
      "grad_norm": 3.6204237901506695,
      "language_loss": 0.77568388,
      "learning_rate": 1.0683987202446475e-07,
      "loss": 0.79770416,
      "num_input_tokens_seen": 161566960,
      "step": 7473,
      "time_per_iteration": 2.5579640865325928
    },
    {
      "auxiliary_loss_clip": 0.01227096,
      "auxiliary_loss_mlp": 0.01027612,
      "balance_loss_clip": 1.04763186,
      "balance_loss_mlp": 1.02024531,
      "epoch": 0.8986953646365659,
      "flos": 21617003208960.0,
      "grad_norm": 2.3505575453807133,
      "language_loss": 0.69899327,
      "learning_rate": 1.0658882150689862e-07,
      "loss": 0.72154033,
      "num_input_tokens_seen": 161585820,
      "step": 7474,
      "time_per_iteration": 2.6168203353881836
    },
    {
      "auxiliary_loss_clip": 0.01322075,
      "auxiliary_loss_mlp": 0.01021835,
      "balance_loss_clip": 1.04575276,
      "balance_loss_mlp": 1.01477814,
      "epoch": 0.8988156075272049,
      "flos": 14027355083520.0,
      "grad_norm": 2.454648144626863,
      "language_loss": 0.78675234,
      "learning_rate": 1.0633805821830288e-07,
      "loss": 0.81019151,
      "num_input_tokens_seen": 161602505,
      "step": 7475,
      "time_per_iteration": 2.6135222911834717
    },
    {
      "auxiliary_loss_clip": 0.01274225,
      "auxiliary_loss_mlp": 0.01023593,
      "balance_loss_clip": 1.04665256,
      "balance_loss_mlp": 1.0167861,
      "epoch": 0.8989358504178441,
      "flos": 29059202004480.0,
      "grad_norm": 4.106360224198777,
      "language_loss": 0.83086103,
      "learning_rate": 1.0608758219671753e-07,
      "loss": 0.85383922,
      "num_input_tokens_seen": 161621545,
      "step": 7476,
      "time_per_iteration": 2.738093852996826
    },
    {
      "auxiliary_loss_clip": 0.01278029,
      "auxiliary_loss_mlp": 0.01027867,
      "balance_loss_clip": 1.04658771,
      "balance_loss_mlp": 1.02100646,
      "epoch": 0.8990560933084831,
      "flos": 20230420446720.0,
      "grad_norm": 1.5428873629979003,
      "language_loss": 0.7042864,
      "learning_rate": 1.0583739348014065e-07,
      "loss": 0.72734535,
      "num_input_tokens_seen": 161642630,
      "step": 7477,
      "time_per_iteration": 2.6709177494049072
    },
    {
      "auxiliary_loss_clip": 0.01176369,
      "auxiliary_loss_mlp": 0.01030368,
      "balance_loss_clip": 1.0514394,
      "balance_loss_mlp": 1.02276576,
      "epoch": 0.8991763361991222,
      "flos": 25520672459520.0,
      "grad_norm": 2.29154686661569,
      "language_loss": 0.84668732,
      "learning_rate": 1.0558749210652518e-07,
      "loss": 0.86875468,
      "num_input_tokens_seen": 161662560,
      "step": 7478,
      "time_per_iteration": 2.8208038806915283
    },
    {
      "auxiliary_loss_clip": 0.01324278,
      "auxiliary_loss_mlp": 0.01031149,
      "balance_loss_clip": 1.0448091,
      "balance_loss_mlp": 1.02405643,
      "epoch": 0.8992965790897613,
      "flos": 25119191168640.0,
      "grad_norm": 1.736854734918298,
      "language_loss": 0.85555029,
      "learning_rate": 1.053378781137808e-07,
      "loss": 0.87910461,
      "num_input_tokens_seen": 161683480,
      "step": 7479,
      "time_per_iteration": 2.7536044120788574
    },
    {
      "auxiliary_loss_clip": 0.01173407,
      "auxiliary_loss_mlp": 0.01030559,
      "balance_loss_clip": 1.04563189,
      "balance_loss_mlp": 1.02324831,
      "epoch": 0.8994168219804004,
      "flos": 16070815814400.0,
      "grad_norm": 1.9355726600668008,
      "language_loss": 0.77589869,
      "learning_rate": 1.0508855153977392e-07,
      "loss": 0.79793835,
      "num_input_tokens_seen": 161699945,
      "step": 7480,
      "time_per_iteration": 2.6563968658447266
    },
    {
      "auxiliary_loss_clip": 0.01221793,
      "auxiliary_loss_mlp": 0.01022343,
      "balance_loss_clip": 1.04486907,
      "balance_loss_mlp": 1.01520276,
      "epoch": 0.8995370648710395,
      "flos": 24825764966400.0,
      "grad_norm": 2.8920908665007956,
      "language_loss": 0.67190742,
      "learning_rate": 1.0483951242232669e-07,
      "loss": 0.69434875,
      "num_input_tokens_seen": 161720420,
      "step": 7481,
      "time_per_iteration": 2.628389596939087
    },
    {
      "auxiliary_loss_clip": 0.01060649,
      "auxiliary_loss_mlp": 0.00997951,
      "balance_loss_clip": 1.00633025,
      "balance_loss_mlp": 0.99707472,
      "epoch": 0.8996573077616786,
      "flos": 63116238378240.0,
      "grad_norm": 0.9696953615608013,
      "language_loss": 0.57700312,
      "learning_rate": 1.0459076079921936e-07,
      "loss": 0.59758902,
      "num_input_tokens_seen": 161773080,
      "step": 7482,
      "time_per_iteration": 3.224740982055664
    },
    {
      "auxiliary_loss_clip": 0.0126797,
      "auxiliary_loss_mlp": 0.01028726,
      "balance_loss_clip": 1.04575109,
      "balance_loss_mlp": 1.02119517,
      "epoch": 0.8997775506523177,
      "flos": 18219674027520.0,
      "grad_norm": 2.481346458767828,
      "language_loss": 0.85167187,
      "learning_rate": 1.0434229670818618e-07,
      "loss": 0.87463892,
      "num_input_tokens_seen": 161789755,
      "step": 7483,
      "time_per_iteration": 2.6706936359405518
    },
    {
      "auxiliary_loss_clip": 0.01265196,
      "auxiliary_loss_mlp": 0.01028356,
      "balance_loss_clip": 1.04362881,
      "balance_loss_mlp": 1.02185631,
      "epoch": 0.8998977935429567,
      "flos": 24166768095360.0,
      "grad_norm": 2.369761021471158,
      "language_loss": 0.8015337,
      "learning_rate": 1.0409412018691944e-07,
      "loss": 0.82446927,
      "num_input_tokens_seen": 161810220,
      "step": 7484,
      "time_per_iteration": 2.6955106258392334
    },
    {
      "auxiliary_loss_clip": 0.01270777,
      "auxiliary_loss_mlp": 0.01031502,
      "balance_loss_clip": 1.04529834,
      "balance_loss_mlp": 1.02454019,
      "epoch": 0.9000180364335959,
      "flos": 20773030273920.0,
      "grad_norm": 1.8416445492711069,
      "language_loss": 0.75114036,
      "learning_rate": 1.0384623127306724e-07,
      "loss": 0.77416313,
      "num_input_tokens_seen": 161827565,
      "step": 7485,
      "time_per_iteration": 2.709456443786621
    },
    {
      "auxiliary_loss_clip": 0.01320474,
      "auxiliary_loss_mlp": 0.0102721,
      "balance_loss_clip": 1.0423305,
      "balance_loss_mlp": 1.02038848,
      "epoch": 0.900138279324235,
      "flos": 19205745166080.0,
      "grad_norm": 2.0439805831730005,
      "language_loss": 0.79504788,
      "learning_rate": 1.0359863000423397e-07,
      "loss": 0.81852472,
      "num_input_tokens_seen": 161845700,
      "step": 7486,
      "time_per_iteration": 2.7102298736572266
    },
    {
      "auxiliary_loss_clip": 0.01173825,
      "auxiliary_loss_mlp": 0.0102376,
      "balance_loss_clip": 1.04847646,
      "balance_loss_mlp": 1.01726627,
      "epoch": 0.900258522214874,
      "flos": 28731158069760.0,
      "grad_norm": 1.5551303531018585,
      "language_loss": 0.7188834,
      "learning_rate": 1.0335131641798112e-07,
      "loss": 0.74085927,
      "num_input_tokens_seen": 161867660,
      "step": 7487,
      "time_per_iteration": 2.7350993156433105
    },
    {
      "auxiliary_loss_clip": 0.01165521,
      "auxiliary_loss_mlp": 0.01002408,
      "balance_loss_clip": 1.00582445,
      "balance_loss_mlp": 1.0015738,
      "epoch": 0.9003787651055132,
      "flos": 58280685655680.0,
      "grad_norm": 0.8358813554376188,
      "language_loss": 0.55542231,
      "learning_rate": 1.0310429055182512e-07,
      "loss": 0.57710159,
      "num_input_tokens_seen": 161921980,
      "step": 7488,
      "time_per_iteration": 3.0950400829315186
    },
    {
      "auxiliary_loss_clip": 0.01321466,
      "auxiliary_loss_mlp": 0.01032557,
      "balance_loss_clip": 1.044451,
      "balance_loss_mlp": 1.02547002,
      "epoch": 0.9004990079961522,
      "flos": 25556475340800.0,
      "grad_norm": 1.8380410420872626,
      "language_loss": 0.73928571,
      "learning_rate": 1.0285755244324024e-07,
      "loss": 0.76282597,
      "num_input_tokens_seen": 161942725,
      "step": 7489,
      "time_per_iteration": 2.766084909439087
    },
    {
      "auxiliary_loss_clip": 0.01272954,
      "auxiliary_loss_mlp": 0.0256223,
      "balance_loss_clip": 1.0436542,
      "balance_loss_mlp": 0.99991298,
      "epoch": 0.9006192508867913,
      "flos": 23335185352320.0,
      "grad_norm": 1.6905846878227084,
      "language_loss": 0.69032073,
      "learning_rate": 1.0261110212965629e-07,
      "loss": 0.72867262,
      "num_input_tokens_seen": 161964520,
      "step": 7490,
      "time_per_iteration": 3.6712255477905273
    },
    {
      "auxiliary_loss_clip": 0.01271008,
      "auxiliary_loss_mlp": 0.01028271,
      "balance_loss_clip": 1.0458647,
      "balance_loss_mlp": 1.021613,
      "epoch": 0.9007394937774305,
      "flos": 18040300485120.0,
      "grad_norm": 2.047707951794858,
      "language_loss": 0.79010326,
      "learning_rate": 1.023649396484596e-07,
      "loss": 0.81309605,
      "num_input_tokens_seen": 161983575,
      "step": 7491,
      "time_per_iteration": 3.613166570663452
    },
    {
      "auxiliary_loss_clip": 0.01173823,
      "auxiliary_loss_mlp": 0.01023711,
      "balance_loss_clip": 1.04942274,
      "balance_loss_mlp": 1.01693368,
      "epoch": 0.9008597366680695,
      "flos": 43068456633600.0,
      "grad_norm": 3.8193669455907067,
      "language_loss": 0.67765427,
      "learning_rate": 1.0211906503699275e-07,
      "loss": 0.69962966,
      "num_input_tokens_seen": 162006550,
      "step": 7492,
      "time_per_iteration": 3.7057671546936035
    },
    {
      "auxiliary_loss_clip": 0.0122452,
      "auxiliary_loss_mlp": 0.01029053,
      "balance_loss_clip": 1.05028403,
      "balance_loss_mlp": 1.02162313,
      "epoch": 0.9009799795587086,
      "flos": 14939055112320.0,
      "grad_norm": 3.1905690613604545,
      "language_loss": 0.82415223,
      "learning_rate": 1.0187347833255455e-07,
      "loss": 0.84668791,
      "num_input_tokens_seen": 162022455,
      "step": 7493,
      "time_per_iteration": 2.625375270843506
    },
    {
      "auxiliary_loss_clip": 0.01171429,
      "auxiliary_loss_mlp": 0.01031196,
      "balance_loss_clip": 1.04959643,
      "balance_loss_mlp": 1.02459192,
      "epoch": 0.9011002224493477,
      "flos": 21579584215680.0,
      "grad_norm": 1.8358632121547218,
      "language_loss": 0.7933234,
      "learning_rate": 1.0162817957240056e-07,
      "loss": 0.8153497,
      "num_input_tokens_seen": 162042350,
      "step": 7494,
      "time_per_iteration": 2.580335855484009
    },
    {
      "auxiliary_loss_clip": 0.01114411,
      "auxiliary_loss_mlp": 0.00998736,
      "balance_loss_clip": 1.00681853,
      "balance_loss_mlp": 0.9978773,
      "epoch": 0.9012204653399868,
      "flos": 71166367883520.0,
      "grad_norm": 0.886071762828719,
      "language_loss": 0.63015676,
      "learning_rate": 1.0138316879374253e-07,
      "loss": 0.65128821,
      "num_input_tokens_seen": 162111640,
      "step": 7495,
      "time_per_iteration": 3.3735499382019043
    },
    {
      "auxiliary_loss_clip": 0.01270135,
      "auxiliary_loss_mlp": 0.01030145,
      "balance_loss_clip": 1.04762268,
      "balance_loss_mlp": 1.02316213,
      "epoch": 0.9013407082306258,
      "flos": 15594963413760.0,
      "grad_norm": 2.202398254874286,
      "language_loss": 0.74664831,
      "learning_rate": 1.0113844603374833e-07,
      "loss": 0.76965117,
      "num_input_tokens_seen": 162128165,
      "step": 7496,
      "time_per_iteration": 2.5921730995178223
    },
    {
      "auxiliary_loss_clip": 0.01272149,
      "auxiliary_loss_mlp": 0.01027984,
      "balance_loss_clip": 1.04500937,
      "balance_loss_mlp": 1.02024722,
      "epoch": 0.901460951121265,
      "flos": 15049157276160.0,
      "grad_norm": 2.790567710390583,
      "language_loss": 0.72166419,
      "learning_rate": 1.0089401132954178e-07,
      "loss": 0.7446655,
      "num_input_tokens_seen": 162146145,
      "step": 7497,
      "time_per_iteration": 3.555449962615967
    },
    {
      "auxiliary_loss_clip": 0.01270994,
      "auxiliary_loss_mlp": 0.01025916,
      "balance_loss_clip": 1.04588842,
      "balance_loss_mlp": 1.01949978,
      "epoch": 0.9015811940119041,
      "flos": 22236857233920.0,
      "grad_norm": 1.9105340086231528,
      "language_loss": 0.72360879,
      "learning_rate": 1.006498647182037e-07,
      "loss": 0.74657786,
      "num_input_tokens_seen": 162164800,
      "step": 7498,
      "time_per_iteration": 2.696904420852661
    },
    {
      "auxiliary_loss_clip": 0.01421118,
      "auxiliary_loss_mlp": 0.01031219,
      "balance_loss_clip": 1.03952694,
      "balance_loss_mlp": 1.02384579,
      "epoch": 0.9017014369025431,
      "flos": 24973824827520.0,
      "grad_norm": 2.341608087218323,
      "language_loss": 0.71763724,
      "learning_rate": 1.004060062367713e-07,
      "loss": 0.74216062,
      "num_input_tokens_seen": 162185895,
      "step": 7499,
      "time_per_iteration": 2.809385299682617
    },
    {
      "auxiliary_loss_clip": 0.0122483,
      "auxiliary_loss_mlp": 0.01030025,
      "balance_loss_clip": 1.04810238,
      "balance_loss_mlp": 1.02252114,
      "epoch": 0.9018216797931822,
      "flos": 18114168804480.0,
      "grad_norm": 1.8700754611425907,
      "language_loss": 0.69837832,
      "learning_rate": 1.0016243592223728e-07,
      "loss": 0.72092688,
      "num_input_tokens_seen": 162206295,
      "step": 7500,
      "time_per_iteration": 2.6733274459838867
    },
    {
      "auxiliary_loss_clip": 0.01416003,
      "auxiliary_loss_mlp": 0.01023758,
      "balance_loss_clip": 1.04120326,
      "balance_loss_mlp": 1.01676369,
      "epoch": 0.9019419226838213,
      "flos": 37268452759680.0,
      "grad_norm": 2.0774841038783625,
      "language_loss": 0.65575558,
      "learning_rate": 9.991915381155114e-08,
      "loss": 0.68015319,
      "num_input_tokens_seen": 162229275,
      "step": 7501,
      "time_per_iteration": 2.8910129070281982
    },
    {
      "auxiliary_loss_clip": 0.01226024,
      "auxiliary_loss_mlp": 0.0103036,
      "balance_loss_clip": 1.04775643,
      "balance_loss_mlp": 1.02271295,
      "epoch": 0.9020621655744604,
      "flos": 23441121538560.0,
      "grad_norm": 2.484042692348454,
      "language_loss": 0.75234711,
      "learning_rate": 9.967615994161871e-08,
      "loss": 0.77491093,
      "num_input_tokens_seen": 162248935,
      "step": 7502,
      "time_per_iteration": 2.695986747741699
    },
    {
      "auxiliary_loss_clip": 0.0117186,
      "auxiliary_loss_mlp": 0.01028289,
      "balance_loss_clip": 1.04894328,
      "balance_loss_mlp": 1.02152956,
      "epoch": 0.9021824084650995,
      "flos": 22857465444480.0,
      "grad_norm": 1.8063508278906053,
      "language_loss": 0.78607059,
      "learning_rate": 9.943345434930161e-08,
      "loss": 0.80807209,
      "num_input_tokens_seen": 162269185,
      "step": 7503,
      "time_per_iteration": 2.640256643295288
    },
    {
      "auxiliary_loss_clip": 0.01320915,
      "auxiliary_loss_mlp": 0.01027645,
      "balance_loss_clip": 1.04855204,
      "balance_loss_mlp": 1.02068353,
      "epoch": 0.9023026513557386,
      "flos": 22127581082880.0,
      "grad_norm": 4.223339988793139,
      "language_loss": 0.69531602,
      "learning_rate": 9.919103707141885e-08,
      "loss": 0.71880162,
      "num_input_tokens_seen": 162288065,
      "step": 7504,
      "time_per_iteration": 2.8407580852508545
    },
    {
      "auxiliary_loss_clip": 0.01224573,
      "auxiliary_loss_mlp": 0.01022137,
      "balance_loss_clip": 1.04934037,
      "balance_loss_mlp": 1.01476097,
      "epoch": 0.9024228942463777,
      "flos": 24199087357440.0,
      "grad_norm": 2.1875777265237044,
      "language_loss": 0.76885998,
      "learning_rate": 9.89489081447441e-08,
      "loss": 0.79132712,
      "num_input_tokens_seen": 162305265,
      "step": 7505,
      "time_per_iteration": 2.5926425457000732
    },
    {
      "auxiliary_loss_clip": 0.01269371,
      "auxiliary_loss_mlp": 0.01026283,
      "balance_loss_clip": 1.0434649,
      "balance_loss_mlp": 1.01894617,
      "epoch": 0.9025431371370167,
      "flos": 25008262992000.0,
      "grad_norm": 1.8559899479002695,
      "language_loss": 0.83010793,
      "learning_rate": 9.870706760600844e-08,
      "loss": 0.85306448,
      "num_input_tokens_seen": 162325215,
      "step": 7506,
      "time_per_iteration": 2.7051470279693604
    },
    {
      "auxiliary_loss_clip": 0.0127367,
      "auxiliary_loss_mlp": 0.01030707,
      "balance_loss_clip": 1.04827189,
      "balance_loss_mlp": 1.02350426,
      "epoch": 0.9026633800276559,
      "flos": 18952862440320.0,
      "grad_norm": 1.9146514529508956,
      "language_loss": 0.72864532,
      "learning_rate": 9.846551549189918e-08,
      "loss": 0.75168914,
      "num_input_tokens_seen": 162344820,
      "step": 7507,
      "time_per_iteration": 2.7136874198913574
    },
    {
      "auxiliary_loss_clip": 0.01268434,
      "auxiliary_loss_mlp": 0.01026358,
      "balance_loss_clip": 1.04670918,
      "balance_loss_mlp": 1.01930678,
      "epoch": 0.902783622918295,
      "flos": 32416059536640.0,
      "grad_norm": 3.5028851586095753,
      "language_loss": 0.68788987,
      "learning_rate": 9.822425183905902e-08,
      "loss": 0.71083772,
      "num_input_tokens_seen": 162365345,
      "step": 7508,
      "time_per_iteration": 2.733797550201416
    },
    {
      "auxiliary_loss_clip": 0.01219844,
      "auxiliary_loss_mlp": 0.01001584,
      "balance_loss_clip": 1.00651741,
      "balance_loss_mlp": 1.0006901,
      "epoch": 0.902903865808934,
      "flos": 63717453244800.0,
      "grad_norm": 0.9199189282899558,
      "language_loss": 0.75053149,
      "learning_rate": 9.798327668408823e-08,
      "loss": 0.77274579,
      "num_input_tokens_seen": 162426980,
      "step": 7509,
      "time_per_iteration": 3.4521212577819824
    },
    {
      "auxiliary_loss_clip": 0.01175083,
      "auxiliary_loss_mlp": 0.01024423,
      "balance_loss_clip": 1.04833353,
      "balance_loss_mlp": 1.01678717,
      "epoch": 0.9030241086995732,
      "flos": 23804034600960.0,
      "grad_norm": 2.0839712458436406,
      "language_loss": 0.68922776,
      "learning_rate": 9.774259006354158e-08,
      "loss": 0.71122277,
      "num_input_tokens_seen": 162447050,
      "step": 7510,
      "time_per_iteration": 2.569611072540283
    },
    {
      "auxiliary_loss_clip": 0.01276853,
      "auxiliary_loss_mlp": 0.01024999,
      "balance_loss_clip": 1.04596734,
      "balance_loss_mlp": 1.01830554,
      "epoch": 0.9031443515902122,
      "flos": 26395887248640.0,
      "grad_norm": 3.3142292181327955,
      "language_loss": 0.76575935,
      "learning_rate": 9.750219201393184e-08,
      "loss": 0.78877783,
      "num_input_tokens_seen": 162467015,
      "step": 7511,
      "time_per_iteration": 2.7352912425994873
    },
    {
      "auxiliary_loss_clip": 0.01221613,
      "auxiliary_loss_mlp": 0.0103017,
      "balance_loss_clip": 1.04768825,
      "balance_loss_mlp": 1.02313066,
      "epoch": 0.9032645944808513,
      "flos": 24939350749440.0,
      "grad_norm": 2.267444862976855,
      "language_loss": 0.7820434,
      "learning_rate": 9.726208257172697e-08,
      "loss": 0.80456126,
      "num_input_tokens_seen": 162488710,
      "step": 7512,
      "time_per_iteration": 2.672280788421631
    },
    {
      "auxiliary_loss_clip": 0.01172341,
      "auxiliary_loss_mlp": 0.01026582,
      "balance_loss_clip": 1.04909003,
      "balance_loss_mlp": 1.01955509,
      "epoch": 0.9033848373714904,
      "flos": 21178821196800.0,
      "grad_norm": 2.0059591334146423,
      "language_loss": 0.74614179,
      "learning_rate": 9.702226177335115e-08,
      "loss": 0.76813108,
      "num_input_tokens_seen": 162507205,
      "step": 7513,
      "time_per_iteration": 2.6243467330932617
    },
    {
      "auxiliary_loss_clip": 0.01272383,
      "auxiliary_loss_mlp": 0.01029451,
      "balance_loss_clip": 1.04838562,
      "balance_loss_mlp": 1.02253723,
      "epoch": 0.9035050802621295,
      "flos": 26286359702400.0,
      "grad_norm": 1.6510426508389227,
      "language_loss": 0.7270925,
      "learning_rate": 9.67827296551853e-08,
      "loss": 0.75011086,
      "num_input_tokens_seen": 162528490,
      "step": 7514,
      "time_per_iteration": 2.667962074279785
    },
    {
      "auxiliary_loss_clip": 0.01267996,
      "auxiliary_loss_mlp": 0.02563967,
      "balance_loss_clip": 1.04453218,
      "balance_loss_mlp": 0.99993706,
      "epoch": 0.9036253231527686,
      "flos": 24204546224640.0,
      "grad_norm": 1.926111310567,
      "language_loss": 0.68498391,
      "learning_rate": 9.65434862535659e-08,
      "loss": 0.7233035,
      "num_input_tokens_seen": 162547860,
      "step": 7515,
      "time_per_iteration": 3.6485180854797363
    },
    {
      "auxiliary_loss_clip": 0.01278333,
      "auxiliary_loss_mlp": 0.01032321,
      "balance_loss_clip": 1.04795146,
      "balance_loss_mlp": 1.02523732,
      "epoch": 0.9037455660434077,
      "flos": 18072655660800.0,
      "grad_norm": 3.258028218521879,
      "language_loss": 0.65159637,
      "learning_rate": 9.630453160478635e-08,
      "loss": 0.67470294,
      "num_input_tokens_seen": 162563215,
      "step": 7516,
      "time_per_iteration": 2.614276885986328
    },
    {
      "auxiliary_loss_clip": 0.01364258,
      "auxiliary_loss_mlp": 0.01025226,
      "balance_loss_clip": 1.0404532,
      "balance_loss_mlp": 1.01873195,
      "epoch": 0.9038658089340468,
      "flos": 24060795995520.0,
      "grad_norm": 2.2533589307828583,
      "language_loss": 0.82512647,
      "learning_rate": 9.60658657450959e-08,
      "loss": 0.84902132,
      "num_input_tokens_seen": 162583515,
      "step": 7517,
      "time_per_iteration": 3.716310977935791
    },
    {
      "auxiliary_loss_clip": 0.01261956,
      "auxiliary_loss_mlp": 0.01023044,
      "balance_loss_clip": 1.04209471,
      "balance_loss_mlp": 1.01663923,
      "epoch": 0.9039860518246858,
      "flos": 21834298535040.0,
      "grad_norm": 1.7407568061763332,
      "language_loss": 0.79613137,
      "learning_rate": 9.582748871069979e-08,
      "loss": 0.81898141,
      "num_input_tokens_seen": 162602955,
      "step": 7518,
      "time_per_iteration": 3.6157875061035156
    },
    {
      "auxiliary_loss_clip": 0.0127528,
      "auxiliary_loss_mlp": 0.02562681,
      "balance_loss_clip": 1.04572988,
      "balance_loss_mlp": 0.99992156,
      "epoch": 0.904106294715325,
      "flos": 26614870513920.0,
      "grad_norm": 2.0423994017439346,
      "language_loss": 0.8309468,
      "learning_rate": 9.558940053775954e-08,
      "loss": 0.86932635,
      "num_input_tokens_seen": 162621595,
      "step": 7519,
      "time_per_iteration": 2.762542247772217
    },
    {
      "auxiliary_loss_clip": 0.01220029,
      "auxiliary_loss_mlp": 0.01026948,
      "balance_loss_clip": 1.04784071,
      "balance_loss_mlp": 1.0196104,
      "epoch": 0.904226537605964,
      "flos": 17785693906560.0,
      "grad_norm": 1.8162696981499886,
      "language_loss": 0.68008351,
      "learning_rate": 9.535160126239294e-08,
      "loss": 0.70255327,
      "num_input_tokens_seen": 162638220,
      "step": 7520,
      "time_per_iteration": 2.566317081451416
    },
    {
      "auxiliary_loss_clip": 0.01220145,
      "auxiliary_loss_mlp": 0.01024635,
      "balance_loss_clip": 1.04828095,
      "balance_loss_mlp": 1.01810551,
      "epoch": 0.9043467804966031,
      "flos": 24790428961920.0,
      "grad_norm": 1.616475689565195,
      "language_loss": 0.70962441,
      "learning_rate": 9.511409092067424e-08,
      "loss": 0.73207223,
      "num_input_tokens_seen": 162658575,
      "step": 7521,
      "time_per_iteration": 2.6775503158569336
    },
    {
      "auxiliary_loss_clip": 0.0126808,
      "auxiliary_loss_mlp": 0.01027256,
      "balance_loss_clip": 1.04801273,
      "balance_loss_mlp": 1.02000213,
      "epoch": 0.9044670233872423,
      "flos": 22632125472000.0,
      "grad_norm": 1.9555799208537195,
      "language_loss": 0.67362118,
      "learning_rate": 9.487686954863327e-08,
      "loss": 0.69657457,
      "num_input_tokens_seen": 162678295,
      "step": 7522,
      "time_per_iteration": 2.6533730030059814
    },
    {
      "auxiliary_loss_clip": 0.01218324,
      "auxiliary_loss_mlp": 0.01026586,
      "balance_loss_clip": 1.04736614,
      "balance_loss_mlp": 1.01982069,
      "epoch": 0.9045872662778813,
      "flos": 23771320289280.0,
      "grad_norm": 4.1337302987647195,
      "language_loss": 0.77862471,
      "learning_rate": 9.46399371822566e-08,
      "loss": 0.80107379,
      "num_input_tokens_seen": 162698070,
      "step": 7523,
      "time_per_iteration": 3.852790355682373
    },
    {
      "auxiliary_loss_clip": 0.0117365,
      "auxiliary_loss_mlp": 0.01026104,
      "balance_loss_clip": 1.04917502,
      "balance_loss_mlp": 1.0189724,
      "epoch": 0.9047075091685204,
      "flos": 15191039998080.0,
      "grad_norm": 2.3770772638479345,
      "language_loss": 0.72253102,
      "learning_rate": 9.440329385748657e-08,
      "loss": 0.74452853,
      "num_input_tokens_seen": 162715140,
      "step": 7524,
      "time_per_iteration": 2.613947629928589
    },
    {
      "auxiliary_loss_clip": 0.01320628,
      "auxiliary_loss_mlp": 0.01018819,
      "balance_loss_clip": 1.0473876,
      "balance_loss_mlp": 1.01230681,
      "epoch": 0.9048277520591596,
      "flos": 18003707504640.0,
      "grad_norm": 1.7895748133024145,
      "language_loss": 0.70810604,
      "learning_rate": 9.416693961022137e-08,
      "loss": 0.73150051,
      "num_input_tokens_seen": 162733390,
      "step": 7525,
      "time_per_iteration": 2.6340103149414062
    },
    {
      "auxiliary_loss_clip": 0.0140844,
      "auxiliary_loss_mlp": 0.01024267,
      "balance_loss_clip": 1.03884923,
      "balance_loss_mlp": 1.01777649,
      "epoch": 0.9049479949497986,
      "flos": 21872471713920.0,
      "grad_norm": 1.7101327101743922,
      "language_loss": 0.77054024,
      "learning_rate": 9.393087447631654e-08,
      "loss": 0.79486734,
      "num_input_tokens_seen": 162751670,
      "step": 7526,
      "time_per_iteration": 2.783454179763794
    },
    {
      "auxiliary_loss_clip": 0.01170557,
      "auxiliary_loss_mlp": 0.01024528,
      "balance_loss_clip": 1.04522061,
      "balance_loss_mlp": 1.01833785,
      "epoch": 0.9050682378404377,
      "flos": 20773928113920.0,
      "grad_norm": 1.6104582193015373,
      "language_loss": 0.73013073,
      "learning_rate": 9.36950984915823e-08,
      "loss": 0.75208151,
      "num_input_tokens_seen": 162770025,
      "step": 7527,
      "time_per_iteration": 2.612396240234375
    },
    {
      "auxiliary_loss_clip": 0.01172889,
      "auxiliary_loss_mlp": 0.01023577,
      "balance_loss_clip": 1.04921722,
      "balance_loss_mlp": 1.01656771,
      "epoch": 0.9051884807310768,
      "flos": 21580015178880.0,
      "grad_norm": 2.0611069313049315,
      "language_loss": 0.6941216,
      "learning_rate": 9.345961169178607e-08,
      "loss": 0.71608627,
      "num_input_tokens_seen": 162789710,
      "step": 7528,
      "time_per_iteration": 2.6093201637268066
    },
    {
      "auxiliary_loss_clip": 0.01321558,
      "auxiliary_loss_mlp": 0.01025123,
      "balance_loss_clip": 1.05115247,
      "balance_loss_mlp": 1.01768184,
      "epoch": 0.9053087236217159,
      "flos": 21908059113600.0,
      "grad_norm": 1.470980672708247,
      "language_loss": 0.72562009,
      "learning_rate": 9.322441411265081e-08,
      "loss": 0.74908692,
      "num_input_tokens_seen": 162810695,
      "step": 7529,
      "time_per_iteration": 2.702442169189453
    },
    {
      "auxiliary_loss_clip": 0.01271687,
      "auxiliary_loss_mlp": 0.0102733,
      "balance_loss_clip": 1.04596186,
      "balance_loss_mlp": 1.02031446,
      "epoch": 0.9054289665123549,
      "flos": 17055809544960.0,
      "grad_norm": 1.9431566291543405,
      "language_loss": 0.73084533,
      "learning_rate": 9.298950578985554e-08,
      "loss": 0.75383544,
      "num_input_tokens_seen": 162827770,
      "step": 7530,
      "time_per_iteration": 2.749070405960083
    },
    {
      "auxiliary_loss_clip": 0.0122059,
      "auxiliary_loss_mlp": 0.02567315,
      "balance_loss_clip": 1.04880631,
      "balance_loss_mlp": 0.99987406,
      "epoch": 0.905549209402994,
      "flos": 20777268078720.0,
      "grad_norm": 1.6875204207844825,
      "language_loss": 0.70667458,
      "learning_rate": 9.275488675903665e-08,
      "loss": 0.74455369,
      "num_input_tokens_seen": 162846715,
      "step": 7531,
      "time_per_iteration": 2.6348352432250977
    },
    {
      "auxiliary_loss_clip": 0.01374524,
      "auxiliary_loss_mlp": 0.0102455,
      "balance_loss_clip": 1.04560602,
      "balance_loss_mlp": 1.01750445,
      "epoch": 0.9056694522936332,
      "flos": 21686813291520.0,
      "grad_norm": 4.383111815625869,
      "language_loss": 0.73818779,
      "learning_rate": 9.252055705578454e-08,
      "loss": 0.76217854,
      "num_input_tokens_seen": 162866215,
      "step": 7532,
      "time_per_iteration": 2.759201765060425
    },
    {
      "auxiliary_loss_clip": 0.01222197,
      "auxiliary_loss_mlp": 0.01023092,
      "balance_loss_clip": 1.04643273,
      "balance_loss_mlp": 1.01691997,
      "epoch": 0.9057896951842722,
      "flos": 29569133433600.0,
      "grad_norm": 2.03275959456739,
      "language_loss": 0.72189224,
      "learning_rate": 9.228651671564747e-08,
      "loss": 0.74434519,
      "num_input_tokens_seen": 162888245,
      "step": 7533,
      "time_per_iteration": 2.6896867752075195
    },
    {
      "auxiliary_loss_clip": 0.01365183,
      "auxiliary_loss_mlp": 0.01025712,
      "balance_loss_clip": 1.04592824,
      "balance_loss_mlp": 1.01929235,
      "epoch": 0.9059099380749113,
      "flos": 27892248952320.0,
      "grad_norm": 1.5411700860182593,
      "language_loss": 0.78083223,
      "learning_rate": 9.205276577412901e-08,
      "loss": 0.80474114,
      "num_input_tokens_seen": 162911025,
      "step": 7534,
      "time_per_iteration": 2.7731142044067383
    },
    {
      "auxiliary_loss_clip": 0.01278015,
      "auxiliary_loss_mlp": 0.02566229,
      "balance_loss_clip": 1.04577529,
      "balance_loss_mlp": 0.99991494,
      "epoch": 0.9060301809655504,
      "flos": 17748993185280.0,
      "grad_norm": 2.912690853902717,
      "language_loss": 0.7669524,
      "learning_rate": 9.181930426668905e-08,
      "loss": 0.80539489,
      "num_input_tokens_seen": 162927820,
      "step": 7535,
      "time_per_iteration": 2.662313461303711
    },
    {
      "auxiliary_loss_clip": 0.01274775,
      "auxiliary_loss_mlp": 0.01026629,
      "balance_loss_clip": 1.04468393,
      "balance_loss_mlp": 1.0198189,
      "epoch": 0.9061504238561895,
      "flos": 31759432963200.0,
      "grad_norm": 1.561676719377741,
      "language_loss": 0.67874438,
      "learning_rate": 9.158613222874346e-08,
      "loss": 0.70175838,
      "num_input_tokens_seen": 162949445,
      "step": 7536,
      "time_per_iteration": 2.8976027965545654
    },
    {
      "auxiliary_loss_clip": 0.01269385,
      "auxiliary_loss_mlp": 0.01021929,
      "balance_loss_clip": 1.0441885,
      "balance_loss_mlp": 1.01536047,
      "epoch": 0.9062706667468285,
      "flos": 20048066075520.0,
      "grad_norm": 1.622584414445404,
      "language_loss": 0.82166004,
      "learning_rate": 9.135324969566394e-08,
      "loss": 0.84457326,
      "num_input_tokens_seen": 162968945,
      "step": 7537,
      "time_per_iteration": 2.9210472106933594
    },
    {
      "auxiliary_loss_clip": 0.01228134,
      "auxiliary_loss_mlp": 0.01025804,
      "balance_loss_clip": 1.05016541,
      "balance_loss_mlp": 1.01877403,
      "epoch": 0.9063909096374677,
      "flos": 18437292576000.0,
      "grad_norm": 1.7968038019788235,
      "language_loss": 0.75206029,
      "learning_rate": 9.112065670277913e-08,
      "loss": 0.77459967,
      "num_input_tokens_seen": 162985310,
      "step": 7538,
      "time_per_iteration": 2.6140897274017334
    },
    {
      "auxiliary_loss_clip": 0.0126993,
      "auxiliary_loss_mlp": 0.01024484,
      "balance_loss_clip": 1.04369044,
      "balance_loss_mlp": 1.01828527,
      "epoch": 0.9065111525281068,
      "flos": 33547353361920.0,
      "grad_norm": 1.8906081070582756,
      "language_loss": 0.72817999,
      "learning_rate": 9.088835328537303e-08,
      "loss": 0.75112402,
      "num_input_tokens_seen": 163006900,
      "step": 7539,
      "time_per_iteration": 2.768148899078369
    },
    {
      "auxiliary_loss_clip": 0.01273126,
      "auxiliary_loss_mlp": 0.01027882,
      "balance_loss_clip": 1.04627967,
      "balance_loss_mlp": 1.02052987,
      "epoch": 0.9066313954187458,
      "flos": 23367863750400.0,
      "grad_norm": 3.851998002201823,
      "language_loss": 0.71569967,
      "learning_rate": 9.065633947868568e-08,
      "loss": 0.73870975,
      "num_input_tokens_seen": 163026505,
      "step": 7540,
      "time_per_iteration": 2.661752700805664
    },
    {
      "auxiliary_loss_clip": 0.01316493,
      "auxiliary_loss_mlp": 0.02562514,
      "balance_loss_clip": 1.04697859,
      "balance_loss_mlp": 0.99989033,
      "epoch": 0.906751638309385,
      "flos": 26249623067520.0,
      "grad_norm": 2.184165836940251,
      "language_loss": 0.80139935,
      "learning_rate": 9.042461531791379e-08,
      "loss": 0.84018946,
      "num_input_tokens_seen": 163044925,
      "step": 7541,
      "time_per_iteration": 3.6434593200683594
    },
    {
      "auxiliary_loss_clip": 0.01168725,
      "auxiliary_loss_mlp": 0.01022461,
      "balance_loss_clip": 1.04812527,
      "balance_loss_mlp": 1.01584768,
      "epoch": 0.906871881200024,
      "flos": 16544477485440.0,
      "grad_norm": 1.7210103265438566,
      "language_loss": 0.7800318,
      "learning_rate": 9.019318083820903e-08,
      "loss": 0.80194366,
      "num_input_tokens_seen": 163063505,
      "step": 7542,
      "time_per_iteration": 2.6327402591705322
    },
    {
      "auxiliary_loss_clip": 0.01220124,
      "auxiliary_loss_mlp": 0.01025125,
      "balance_loss_clip": 1.04714084,
      "balance_loss_mlp": 1.01820779,
      "epoch": 0.9069921240906631,
      "flos": 24605129675520.0,
      "grad_norm": 1.735459062756827,
      "language_loss": 0.84965932,
      "learning_rate": 8.996203607468045e-08,
      "loss": 0.8721118,
      "num_input_tokens_seen": 163082505,
      "step": 7543,
      "time_per_iteration": 3.666023015975952
    },
    {
      "auxiliary_loss_clip": 0.01219714,
      "auxiliary_loss_mlp": 0.01024743,
      "balance_loss_clip": 1.0433743,
      "balance_loss_mlp": 1.01756048,
      "epoch": 0.9071123669813023,
      "flos": 25374731500800.0,
      "grad_norm": 1.4884574503709511,
      "language_loss": 0.7543931,
      "learning_rate": 8.973118106239241e-08,
      "loss": 0.77683771,
      "num_input_tokens_seen": 163105110,
      "step": 7544,
      "time_per_iteration": 2.736302137374878
    },
    {
      "auxiliary_loss_clip": 0.01422947,
      "auxiliary_loss_mlp": 0.01024403,
      "balance_loss_clip": 1.03852344,
      "balance_loss_mlp": 1.01784635,
      "epoch": 0.9072326098719413,
      "flos": 26725798690560.0,
      "grad_norm": 1.9766408173081598,
      "language_loss": 0.94701767,
      "learning_rate": 8.95006158363656e-08,
      "loss": 0.97149122,
      "num_input_tokens_seen": 163125295,
      "step": 7545,
      "time_per_iteration": 3.7341220378875732
    },
    {
      "auxiliary_loss_clip": 0.01229075,
      "auxiliary_loss_mlp": 0.01029251,
      "balance_loss_clip": 1.05173576,
      "balance_loss_mlp": 1.0213834,
      "epoch": 0.9073528527625804,
      "flos": 23878800760320.0,
      "grad_norm": 2.237316622818048,
      "language_loss": 0.77604473,
      "learning_rate": 8.9270340431576e-08,
      "loss": 0.79862809,
      "num_input_tokens_seen": 163144385,
      "step": 7546,
      "time_per_iteration": 2.614132881164551
    },
    {
      "auxiliary_loss_clip": 0.012232,
      "auxiliary_loss_mlp": 0.01023188,
      "balance_loss_clip": 1.04672861,
      "balance_loss_mlp": 1.01600885,
      "epoch": 0.9074730956532195,
      "flos": 37852144767360.0,
      "grad_norm": 6.2876781967164685,
      "language_loss": 0.73682415,
      "learning_rate": 8.904035488295658e-08,
      "loss": 0.75928807,
      "num_input_tokens_seen": 163163885,
      "step": 7547,
      "time_per_iteration": 2.7679173946380615
    },
    {
      "auxiliary_loss_clip": 0.01113774,
      "auxiliary_loss_mlp": 0.02505062,
      "balance_loss_clip": 1.00644636,
      "balance_loss_mlp": 0.99993652,
      "epoch": 0.9075933385438586,
      "flos": 65173307385600.0,
      "grad_norm": 0.6749293508983584,
      "language_loss": 0.53234601,
      "learning_rate": 8.881065922539632e-08,
      "loss": 0.56853437,
      "num_input_tokens_seen": 163224325,
      "step": 7548,
      "time_per_iteration": 3.128706932067871
    },
    {
      "auxiliary_loss_clip": 0.01313854,
      "auxiliary_loss_mlp": 0.01025462,
      "balance_loss_clip": 1.04576433,
      "balance_loss_mlp": 1.01911664,
      "epoch": 0.9077135814344977,
      "flos": 19931571290880.0,
      "grad_norm": 2.0164824056689676,
      "language_loss": 0.73640478,
      "learning_rate": 8.85812534937389e-08,
      "loss": 0.75979793,
      "num_input_tokens_seen": 163242425,
      "step": 7549,
      "time_per_iteration": 3.5682995319366455
    },
    {
      "auxiliary_loss_clip": 0.01129517,
      "auxiliary_loss_mlp": 0.01032331,
      "balance_loss_clip": 1.05081034,
      "balance_loss_mlp": 1.02491367,
      "epoch": 0.9078338243251368,
      "flos": 17529650784000.0,
      "grad_norm": 2.6455919073185403,
      "language_loss": 0.67957497,
      "learning_rate": 8.835213772278583e-08,
      "loss": 0.70119351,
      "num_input_tokens_seen": 163259280,
      "step": 7550,
      "time_per_iteration": 2.634061813354492
    },
    {
      "auxiliary_loss_clip": 0.01317977,
      "auxiliary_loss_mlp": 0.01020371,
      "balance_loss_clip": 1.04795575,
      "balance_loss_mlp": 1.01370764,
      "epoch": 0.9079540672157759,
      "flos": 28803410277120.0,
      "grad_norm": 1.9942270889796516,
      "language_loss": 0.78826094,
      "learning_rate": 8.812331194729373e-08,
      "loss": 0.81164443,
      "num_input_tokens_seen": 163278925,
      "step": 7551,
      "time_per_iteration": 2.731265068054199
    },
    {
      "auxiliary_loss_clip": 0.01179608,
      "auxiliary_loss_mlp": 0.01032637,
      "balance_loss_clip": 1.05346274,
      "balance_loss_mlp": 1.02544546,
      "epoch": 0.9080743101064149,
      "flos": 23513840622720.0,
      "grad_norm": 1.9572778563972915,
      "language_loss": 0.72511387,
      "learning_rate": 8.789477620197461e-08,
      "loss": 0.74723637,
      "num_input_tokens_seen": 163298450,
      "step": 7552,
      "time_per_iteration": 2.718407392501831
    },
    {
      "auxiliary_loss_clip": 0.0127123,
      "auxiliary_loss_mlp": 0.01024389,
      "balance_loss_clip": 1.04592037,
      "balance_loss_mlp": 1.01729333,
      "epoch": 0.9081945529970541,
      "flos": 22778102344320.0,
      "grad_norm": 3.2894664677818555,
      "language_loss": 0.78862953,
      "learning_rate": 8.766653052149831e-08,
      "loss": 0.81158578,
      "num_input_tokens_seen": 163313635,
      "step": 7553,
      "time_per_iteration": 2.6193885803222656
    },
    {
      "auxiliary_loss_clip": 0.01267607,
      "auxiliary_loss_mlp": 0.01027356,
      "balance_loss_clip": 1.04634404,
      "balance_loss_mlp": 1.02039158,
      "epoch": 0.9083147958876931,
      "flos": 18873714821760.0,
      "grad_norm": 2.638039523908064,
      "language_loss": 0.74514705,
      "learning_rate": 8.743857494048823e-08,
      "loss": 0.76809669,
      "num_input_tokens_seen": 163330450,
      "step": 7554,
      "time_per_iteration": 2.6555545330047607
    },
    {
      "auxiliary_loss_clip": 0.01322544,
      "auxiliary_loss_mlp": 0.01029135,
      "balance_loss_clip": 1.04556346,
      "balance_loss_mlp": 1.02213192,
      "epoch": 0.9084350387783322,
      "flos": 18909374048640.0,
      "grad_norm": 2.312431465750946,
      "language_loss": 0.62570572,
      "learning_rate": 8.721090949352605e-08,
      "loss": 0.64922249,
      "num_input_tokens_seen": 163346690,
      "step": 7555,
      "time_per_iteration": 2.6748809814453125
    },
    {
      "auxiliary_loss_clip": 0.01132511,
      "auxiliary_loss_mlp": 0.01027904,
      "balance_loss_clip": 1.05053866,
      "balance_loss_mlp": 1.01997089,
      "epoch": 0.9085552816689714,
      "flos": 20595488325120.0,
      "grad_norm": 3.316448742562777,
      "language_loss": 0.73387957,
      "learning_rate": 8.698353421514793e-08,
      "loss": 0.75548375,
      "num_input_tokens_seen": 163365065,
      "step": 7556,
      "time_per_iteration": 2.62380313873291
    },
    {
      "auxiliary_loss_clip": 0.01219026,
      "auxiliary_loss_mlp": 0.01028861,
      "balance_loss_clip": 1.04772437,
      "balance_loss_mlp": 1.02256036,
      "epoch": 0.9086755245596104,
      "flos": 18113163223680.0,
      "grad_norm": 3.6852032908939534,
      "language_loss": 0.80065662,
      "learning_rate": 8.67564491398467e-08,
      "loss": 0.8231355,
      "num_input_tokens_seen": 163382070,
      "step": 7557,
      "time_per_iteration": 2.596177816390991
    },
    {
      "auxiliary_loss_clip": 0.01226796,
      "auxiliary_loss_mlp": 0.01024392,
      "balance_loss_clip": 1.04792023,
      "balance_loss_mlp": 1.01774907,
      "epoch": 0.9087957674502495,
      "flos": 19129793857920.0,
      "grad_norm": 1.746330257033229,
      "language_loss": 0.74076021,
      "learning_rate": 8.652965430207104e-08,
      "loss": 0.76327211,
      "num_input_tokens_seen": 163399975,
      "step": 7558,
      "time_per_iteration": 2.6998112201690674
    },
    {
      "auxiliary_loss_clip": 0.01224474,
      "auxiliary_loss_mlp": 0.01025148,
      "balance_loss_clip": 1.04616725,
      "balance_loss_mlp": 1.01824021,
      "epoch": 0.9089160103408886,
      "flos": 18109930999680.0,
      "grad_norm": 2.1652503606870024,
      "language_loss": 0.65710032,
      "learning_rate": 8.630314973622521e-08,
      "loss": 0.67959654,
      "num_input_tokens_seen": 163417520,
      "step": 7559,
      "time_per_iteration": 2.5725057125091553
    },
    {
      "auxiliary_loss_clip": 0.01217325,
      "auxiliary_loss_mlp": 0.01020826,
      "balance_loss_clip": 1.04820657,
      "balance_loss_mlp": 1.01443005,
      "epoch": 0.9090362532315277,
      "flos": 33364855336320.0,
      "grad_norm": 2.2282944982148742,
      "language_loss": 0.7084614,
      "learning_rate": 8.607693547666995e-08,
      "loss": 0.73084289,
      "num_input_tokens_seen": 163440060,
      "step": 7560,
      "time_per_iteration": 2.7540194988250732
    },
    {
      "auxiliary_loss_clip": 0.01223674,
      "auxiliary_loss_mlp": 0.00998436,
      "balance_loss_clip": 1.00601661,
      "balance_loss_mlp": 0.9975422,
      "epoch": 0.9091564961221668,
      "flos": 71480585082240.0,
      "grad_norm": 0.8753520999168358,
      "language_loss": 0.57909131,
      "learning_rate": 8.585101155772201e-08,
      "loss": 0.6013124,
      "num_input_tokens_seen": 163502180,
      "step": 7561,
      "time_per_iteration": 3.3690836429595947
    },
    {
      "auxiliary_loss_clip": 0.01268782,
      "auxiliary_loss_mlp": 0.01028951,
      "balance_loss_clip": 1.04199398,
      "balance_loss_mlp": 1.0218966,
      "epoch": 0.9092767390128058,
      "flos": 24712574232960.0,
      "grad_norm": 1.8271349804239712,
      "language_loss": 0.68634492,
      "learning_rate": 8.562537801365377e-08,
      "loss": 0.70932233,
      "num_input_tokens_seen": 163521915,
      "step": 7562,
      "time_per_iteration": 2.684295892715454
    },
    {
      "auxiliary_loss_clip": 0.01175099,
      "auxiliary_loss_mlp": 0.01025681,
      "balance_loss_clip": 1.04906631,
      "balance_loss_mlp": 1.01854038,
      "epoch": 0.909396981903445,
      "flos": 23586487879680.0,
      "grad_norm": 1.8996454409213293,
      "language_loss": 0.70146048,
      "learning_rate": 8.540003487869362e-08,
      "loss": 0.7234683,
      "num_input_tokens_seen": 163543585,
      "step": 7563,
      "time_per_iteration": 2.6528258323669434
    },
    {
      "auxiliary_loss_clip": 0.0130819,
      "auxiliary_loss_mlp": 0.01024826,
      "balance_loss_clip": 1.04217672,
      "balance_loss_mlp": 1.01728308,
      "epoch": 0.909517224794084,
      "flos": 23404169422080.0,
      "grad_norm": 2.6679176820698154,
      "language_loss": 0.79693192,
      "learning_rate": 8.517498218702557e-08,
      "loss": 0.82026207,
      "num_input_tokens_seen": 163561515,
      "step": 7564,
      "time_per_iteration": 2.718322277069092
    },
    {
      "auxiliary_loss_clip": 0.01318145,
      "auxiliary_loss_mlp": 0.01027685,
      "balance_loss_clip": 1.0426718,
      "balance_loss_mlp": 1.02080667,
      "epoch": 0.9096374676847231,
      "flos": 19208618254080.0,
      "grad_norm": 1.8253110696969452,
      "language_loss": 0.69941449,
      "learning_rate": 8.49502199727905e-08,
      "loss": 0.72287273,
      "num_input_tokens_seen": 163579540,
      "step": 7565,
      "time_per_iteration": 2.7079927921295166
    },
    {
      "auxiliary_loss_clip": 0.01218919,
      "auxiliary_loss_mlp": 0.0102565,
      "balance_loss_clip": 1.04411864,
      "balance_loss_mlp": 1.01868796,
      "epoch": 0.9097577105753623,
      "flos": 33292495388160.0,
      "grad_norm": 5.409002431113223,
      "language_loss": 0.66274464,
      "learning_rate": 8.472574827008428e-08,
      "loss": 0.68519038,
      "num_input_tokens_seen": 163600425,
      "step": 7566,
      "time_per_iteration": 2.705028533935547
    },
    {
      "auxiliary_loss_clip": 0.01221618,
      "auxiliary_loss_mlp": 0.01023114,
      "balance_loss_clip": 1.0463016,
      "balance_loss_mlp": 1.01636922,
      "epoch": 0.9098779534660013,
      "flos": 21906443001600.0,
      "grad_norm": 1.7284709362899893,
      "language_loss": 0.8378911,
      "learning_rate": 8.450156711295942e-08,
      "loss": 0.86033845,
      "num_input_tokens_seen": 163620595,
      "step": 7567,
      "time_per_iteration": 3.4975571632385254
    },
    {
      "auxiliary_loss_clip": 0.01270807,
      "auxiliary_loss_mlp": 0.01026162,
      "balance_loss_clip": 1.04833293,
      "balance_loss_mlp": 1.01934922,
      "epoch": 0.9099981963566404,
      "flos": 25730354102400.0,
      "grad_norm": 2.6023457574424933,
      "language_loss": 0.86574459,
      "learning_rate": 8.427767653542383e-08,
      "loss": 0.88871431,
      "num_input_tokens_seen": 163635765,
      "step": 7568,
      "time_per_iteration": 2.7338473796844482
    },
    {
      "auxiliary_loss_clip": 0.01364454,
      "auxiliary_loss_mlp": 0.01021988,
      "balance_loss_clip": 1.04014897,
      "balance_loss_mlp": 1.01549399,
      "epoch": 0.9101184392472795,
      "flos": 21069437304960.0,
      "grad_norm": 1.8365415681018058,
      "language_loss": 0.70247614,
      "learning_rate": 8.405407657144125e-08,
      "loss": 0.72634053,
      "num_input_tokens_seen": 163654925,
      "step": 7569,
      "time_per_iteration": 3.701667308807373
    },
    {
      "auxiliary_loss_clip": 0.0126646,
      "auxiliary_loss_mlp": 0.01027003,
      "balance_loss_clip": 1.04488301,
      "balance_loss_mlp": 1.02087891,
      "epoch": 0.9102386821379186,
      "flos": 24752614919040.0,
      "grad_norm": 1.9026518190379227,
      "language_loss": 0.72676343,
      "learning_rate": 8.383076725493232e-08,
      "loss": 0.74969804,
      "num_input_tokens_seen": 163672245,
      "step": 7570,
      "time_per_iteration": 2.7026431560516357
    },
    {
      "auxiliary_loss_clip": 0.01223504,
      "auxiliary_loss_mlp": 0.0103187,
      "balance_loss_clip": 1.04710507,
      "balance_loss_mlp": 1.02467322,
      "epoch": 0.9103589250285576,
      "flos": 22562818179840.0,
      "grad_norm": 1.8950795801555236,
      "language_loss": 0.67913073,
      "learning_rate": 8.360774861977216e-08,
      "loss": 0.70168447,
      "num_input_tokens_seen": 163691365,
      "step": 7571,
      "time_per_iteration": 3.508399248123169
    },
    {
      "auxiliary_loss_clip": 0.01270709,
      "auxiliary_loss_mlp": 0.01024657,
      "balance_loss_clip": 1.04189897,
      "balance_loss_mlp": 1.01797831,
      "epoch": 0.9104791679191968,
      "flos": 25373474524800.0,
      "grad_norm": 1.8796840095762815,
      "language_loss": 0.74920994,
      "learning_rate": 8.338502069979281e-08,
      "loss": 0.77216363,
      "num_input_tokens_seen": 163711675,
      "step": 7572,
      "time_per_iteration": 2.721954584121704
    },
    {
      "auxiliary_loss_clip": 0.01222514,
      "auxiliary_loss_mlp": 0.01020859,
      "balance_loss_clip": 1.0442766,
      "balance_loss_mlp": 1.01397419,
      "epoch": 0.9105994108098359,
      "flos": 14426681558400.0,
      "grad_norm": 2.9587753239208303,
      "language_loss": 0.79728049,
      "learning_rate": 8.316258352878214e-08,
      "loss": 0.81971425,
      "num_input_tokens_seen": 163728095,
      "step": 7573,
      "time_per_iteration": 2.5853769779205322
    },
    {
      "auxiliary_loss_clip": 0.01228134,
      "auxiliary_loss_mlp": 0.01027036,
      "balance_loss_clip": 1.04793572,
      "balance_loss_mlp": 1.01939499,
      "epoch": 0.9107196537004749,
      "flos": 26718292748160.0,
      "grad_norm": 2.198098666131195,
      "language_loss": 0.71420968,
      "learning_rate": 8.294043714048338e-08,
      "loss": 0.73676133,
      "num_input_tokens_seen": 163747175,
      "step": 7574,
      "time_per_iteration": 2.6365127563476562
    },
    {
      "auxiliary_loss_clip": 0.01166723,
      "auxiliary_loss_mlp": 0.01002299,
      "balance_loss_clip": 1.00583196,
      "balance_loss_mlp": 1.00139904,
      "epoch": 0.9108398965911141,
      "flos": 66532634703360.0,
      "grad_norm": 0.7523857416289549,
      "language_loss": 0.60413384,
      "learning_rate": 8.271858156859624e-08,
      "loss": 0.62582409,
      "num_input_tokens_seen": 163812545,
      "step": 7575,
      "time_per_iteration": 4.1903674602508545
    },
    {
      "auxiliary_loss_clip": 0.01170997,
      "auxiliary_loss_mlp": 0.01025543,
      "balance_loss_clip": 1.04779804,
      "balance_loss_mlp": 1.01871836,
      "epoch": 0.9109601394817531,
      "flos": 25411073086080.0,
      "grad_norm": 1.7321912407217455,
      "language_loss": 0.7372154,
      "learning_rate": 8.249701684677557e-08,
      "loss": 0.75918078,
      "num_input_tokens_seen": 163833870,
      "step": 7576,
      "time_per_iteration": 2.6117520332336426
    },
    {
      "auxiliary_loss_clip": 0.01219749,
      "auxiliary_loss_mlp": 0.01026075,
      "balance_loss_clip": 1.04853559,
      "balance_loss_mlp": 1.01942623,
      "epoch": 0.9110803823723922,
      "flos": 22747794243840.0,
      "grad_norm": 2.148220495032876,
      "language_loss": 0.81175327,
      "learning_rate": 8.227574300863294e-08,
      "loss": 0.83421153,
      "num_input_tokens_seen": 163854040,
      "step": 7577,
      "time_per_iteration": 2.799767017364502
    },
    {
      "auxiliary_loss_clip": 0.0127621,
      "auxiliary_loss_mlp": 0.01021739,
      "balance_loss_clip": 1.0472393,
      "balance_loss_mlp": 1.01436865,
      "epoch": 0.9112006252630314,
      "flos": 48469924131840.0,
      "grad_norm": 1.7632201900926827,
      "language_loss": 0.69543993,
      "learning_rate": 8.205476008773548e-08,
      "loss": 0.71841943,
      "num_input_tokens_seen": 163878040,
      "step": 7578,
      "time_per_iteration": 2.939774751663208
    },
    {
      "auxiliary_loss_clip": 0.0131313,
      "auxiliary_loss_mlp": 0.01023895,
      "balance_loss_clip": 1.04533935,
      "balance_loss_mlp": 1.01738286,
      "epoch": 0.9113208681536704,
      "flos": 30009649829760.0,
      "grad_norm": 2.3702936812089965,
      "language_loss": 0.82174873,
      "learning_rate": 8.183406811760596e-08,
      "loss": 0.84511894,
      "num_input_tokens_seen": 163897770,
      "step": 7579,
      "time_per_iteration": 2.7141034603118896
    },
    {
      "auxiliary_loss_clip": 0.01313498,
      "auxiliary_loss_mlp": 0.01028268,
      "balance_loss_clip": 1.04076576,
      "balance_loss_mlp": 1.0216434,
      "epoch": 0.9114411110443095,
      "flos": 25594971742080.0,
      "grad_norm": 1.7151870792096728,
      "language_loss": 0.73817432,
      "learning_rate": 8.161366713172313e-08,
      "loss": 0.76159203,
      "num_input_tokens_seen": 163920160,
      "step": 7580,
      "time_per_iteration": 2.7537527084350586
    },
    {
      "auxiliary_loss_clip": 0.01328735,
      "auxiliary_loss_mlp": 0.01030544,
      "balance_loss_clip": 1.04292941,
      "balance_loss_mlp": 1.02293861,
      "epoch": 0.9115613539349486,
      "flos": 18399729928320.0,
      "grad_norm": 2.8179307646447134,
      "language_loss": 0.843714,
      "learning_rate": 8.139355716352137e-08,
      "loss": 0.86730683,
      "num_input_tokens_seen": 163935000,
      "step": 7581,
      "time_per_iteration": 2.6194937229156494
    },
    {
      "auxiliary_loss_clip": 0.01177247,
      "auxiliary_loss_mlp": 0.01025168,
      "balance_loss_clip": 1.04569149,
      "balance_loss_mlp": 1.0182178,
      "epoch": 0.9116815968255877,
      "flos": 21726171619200.0,
      "grad_norm": 2.231809196817548,
      "language_loss": 0.70172942,
      "learning_rate": 8.117373824639196e-08,
      "loss": 0.72375357,
      "num_input_tokens_seen": 163955265,
      "step": 7582,
      "time_per_iteration": 2.8070926666259766
    },
    {
      "auxiliary_loss_clip": 0.0106029,
      "auxiliary_loss_mlp": 0.01000923,
      "balance_loss_clip": 1.00628459,
      "balance_loss_mlp": 1.00001085,
      "epoch": 0.9118018397162267,
      "flos": 65363526835200.0,
      "grad_norm": 0.722798772391123,
      "language_loss": 0.59256959,
      "learning_rate": 8.095421041368067e-08,
      "loss": 0.61318171,
      "num_input_tokens_seen": 164014680,
      "step": 7583,
      "time_per_iteration": 3.074864149093628
    },
    {
      "auxiliary_loss_clip": 0.01266702,
      "auxiliary_loss_mlp": 0.02565753,
      "balance_loss_clip": 1.04652286,
      "balance_loss_mlp": 0.99993289,
      "epoch": 0.9119220826068659,
      "flos": 20922885815040.0,
      "grad_norm": 2.793469101684912,
      "language_loss": 0.70766878,
      "learning_rate": 8.073497369868999e-08,
      "loss": 0.74599332,
      "num_input_tokens_seen": 164033140,
      "step": 7584,
      "time_per_iteration": 2.6740458011627197
    },
    {
      "auxiliary_loss_clip": 0.01279687,
      "auxiliary_loss_mlp": 0.01031242,
      "balance_loss_clip": 1.0461117,
      "balance_loss_mlp": 1.02388096,
      "epoch": 0.912042325497505,
      "flos": 28366449327360.0,
      "grad_norm": 1.6338775113822197,
      "language_loss": 0.75442481,
      "learning_rate": 8.051602813467772e-08,
      "loss": 0.77753413,
      "num_input_tokens_seen": 164054995,
      "step": 7585,
      "time_per_iteration": 2.680936336517334
    },
    {
      "auxiliary_loss_clip": 0.01225294,
      "auxiliary_loss_mlp": 0.01025821,
      "balance_loss_clip": 1.04777169,
      "balance_loss_mlp": 1.01919246,
      "epoch": 0.912162568388144,
      "flos": 17566782468480.0,
      "grad_norm": 1.7443691225399316,
      "language_loss": 0.71268517,
      "learning_rate": 8.029737375485756e-08,
      "loss": 0.73519635,
      "num_input_tokens_seen": 164074225,
      "step": 7586,
      "time_per_iteration": 2.652338743209839
    },
    {
      "auxiliary_loss_clip": 0.01174135,
      "auxiliary_loss_mlp": 0.01024755,
      "balance_loss_clip": 1.04968262,
      "balance_loss_mlp": 1.01773024,
      "epoch": 0.9122828112787832,
      "flos": 19827897661440.0,
      "grad_norm": 1.8193009354767153,
      "language_loss": 0.72410667,
      "learning_rate": 8.007901059239986e-08,
      "loss": 0.74609554,
      "num_input_tokens_seen": 164093505,
      "step": 7587,
      "time_per_iteration": 2.5703210830688477
    },
    {
      "auxiliary_loss_clip": 0.0126985,
      "auxiliary_loss_mlp": 0.01024699,
      "balance_loss_clip": 1.04460573,
      "balance_loss_mlp": 1.01810646,
      "epoch": 0.9124030541694222,
      "flos": 20813789232000.0,
      "grad_norm": 1.863465015382947,
      "language_loss": 0.80414486,
      "learning_rate": 7.986093868042964e-08,
      "loss": 0.82709032,
      "num_input_tokens_seen": 164113750,
      "step": 7588,
      "time_per_iteration": 2.7107160091400146
    },
    {
      "auxiliary_loss_clip": 0.01216704,
      "auxiliary_loss_mlp": 0.01023968,
      "balance_loss_clip": 1.045524,
      "balance_loss_mlp": 1.01763558,
      "epoch": 0.9125232970600613,
      "flos": 25192305302400.0,
      "grad_norm": 2.23688370226302,
      "language_loss": 0.67891347,
      "learning_rate": 7.964315805202826e-08,
      "loss": 0.70132017,
      "num_input_tokens_seen": 164134330,
      "step": 7589,
      "time_per_iteration": 2.6821858882904053
    },
    {
      "auxiliary_loss_clip": 0.01277922,
      "auxiliary_loss_mlp": 0.01031071,
      "balance_loss_clip": 1.04904437,
      "balance_loss_mlp": 1.02395105,
      "epoch": 0.9126435399507005,
      "flos": 19719591177600.0,
      "grad_norm": 2.3722250203650734,
      "language_loss": 0.73303896,
      "learning_rate": 7.942566874023304e-08,
      "loss": 0.75612891,
      "num_input_tokens_seen": 164153515,
      "step": 7590,
      "time_per_iteration": 2.7011170387268066
    },
    {
      "auxiliary_loss_clip": 0.01171452,
      "auxiliary_loss_mlp": 0.01022796,
      "balance_loss_clip": 1.04415214,
      "balance_loss_mlp": 1.01564896,
      "epoch": 0.9127637828413395,
      "flos": 19573614305280.0,
      "grad_norm": 2.206756551964659,
      "language_loss": 0.69669676,
      "learning_rate": 7.920847077803649e-08,
      "loss": 0.71863925,
      "num_input_tokens_seen": 164171305,
      "step": 7591,
      "time_per_iteration": 2.661616325378418
    },
    {
      "auxiliary_loss_clip": 0.0136406,
      "auxiliary_loss_mlp": 0.01022923,
      "balance_loss_clip": 1.03674638,
      "balance_loss_mlp": 1.01576447,
      "epoch": 0.9128840257319786,
      "flos": 20230635928320.0,
      "grad_norm": 1.8607159988264674,
      "language_loss": 0.82206601,
      "learning_rate": 7.899156419838826e-08,
      "loss": 0.84593582,
      "num_input_tokens_seen": 164190275,
      "step": 7592,
      "time_per_iteration": 2.684455156326294
    },
    {
      "auxiliary_loss_clip": 0.0132002,
      "auxiliary_loss_mlp": 0.01028305,
      "balance_loss_clip": 1.04515386,
      "balance_loss_mlp": 1.02085483,
      "epoch": 0.9130042686226177,
      "flos": 24858658846080.0,
      "grad_norm": 2.2945439229076503,
      "language_loss": 0.65886915,
      "learning_rate": 7.87749490341918e-08,
      "loss": 0.68235242,
      "num_input_tokens_seen": 164210550,
      "step": 7593,
      "time_per_iteration": 3.6518964767456055
    },
    {
      "auxiliary_loss_clip": 0.01177561,
      "auxiliary_loss_mlp": 0.01024168,
      "balance_loss_clip": 1.0508132,
      "balance_loss_mlp": 1.01685464,
      "epoch": 0.9131245115132568,
      "flos": 23581747284480.0,
      "grad_norm": 2.3246822823315876,
      "language_loss": 0.8341319,
      "learning_rate": 7.855862531830836e-08,
      "loss": 0.8561492,
      "num_input_tokens_seen": 164226660,
      "step": 7594,
      "time_per_iteration": 2.5909533500671387
    },
    {
      "auxiliary_loss_clip": 0.01223324,
      "auxiliary_loss_mlp": 0.01022077,
      "balance_loss_clip": 1.04658842,
      "balance_loss_mlp": 1.01484668,
      "epoch": 0.9132447544038959,
      "flos": 19931607204480.0,
      "grad_norm": 1.6112898141343444,
      "language_loss": 0.72555327,
      "learning_rate": 7.834259308355373e-08,
      "loss": 0.7480073,
      "num_input_tokens_seen": 164245425,
      "step": 7595,
      "time_per_iteration": 3.569679021835327
    },
    {
      "auxiliary_loss_clip": 0.01406309,
      "auxiliary_loss_mlp": 0.01026775,
      "balance_loss_clip": 1.03910446,
      "balance_loss_mlp": 1.01967907,
      "epoch": 0.9133649972945349,
      "flos": 21981747864960.0,
      "grad_norm": 2.4725091642506385,
      "language_loss": 0.75350952,
      "learning_rate": 7.812685236269989e-08,
      "loss": 0.77784038,
      "num_input_tokens_seen": 164264085,
      "step": 7596,
      "time_per_iteration": 2.7518177032470703
    },
    {
      "auxiliary_loss_clip": 0.01215967,
      "auxiliary_loss_mlp": 0.01003579,
      "balance_loss_clip": 1.00674522,
      "balance_loss_mlp": 1.00267911,
      "epoch": 0.9134852401851741,
      "flos": 71240523511680.0,
      "grad_norm": 0.7914697998193851,
      "language_loss": 0.58644938,
      "learning_rate": 7.791140318847445e-08,
      "loss": 0.60864484,
      "num_input_tokens_seen": 164322220,
      "step": 7597,
      "time_per_iteration": 4.187063932418823
    },
    {
      "auxiliary_loss_clip": 0.01271561,
      "auxiliary_loss_mlp": 0.01025065,
      "balance_loss_clip": 1.04943204,
      "balance_loss_mlp": 1.0185535,
      "epoch": 0.9136054830758131,
      "flos": 23626923615360.0,
      "grad_norm": 1.4269760798260303,
      "language_loss": 0.80353773,
      "learning_rate": 7.769624559356081e-08,
      "loss": 0.82650399,
      "num_input_tokens_seen": 164345615,
      "step": 7598,
      "time_per_iteration": 2.7301881313323975
    },
    {
      "auxiliary_loss_clip": 0.01224185,
      "auxiliary_loss_mlp": 0.01027511,
      "balance_loss_clip": 1.0500071,
      "balance_loss_mlp": 1.02021241,
      "epoch": 0.9137257259664522,
      "flos": 23438858981760.0,
      "grad_norm": 2.722172589798478,
      "language_loss": 0.75797731,
      "learning_rate": 7.748137961059842e-08,
      "loss": 0.78049421,
      "num_input_tokens_seen": 164359595,
      "step": 7599,
      "time_per_iteration": 2.625005006790161
    },
    {
      "auxiliary_loss_clip": 0.01170589,
      "auxiliary_loss_mlp": 0.01023943,
      "balance_loss_clip": 1.04934156,
      "balance_loss_mlp": 1.0173533,
      "epoch": 0.9138459688570914,
      "flos": 19127854523520.0,
      "grad_norm": 5.965510157326781,
      "language_loss": 0.6555562,
      "learning_rate": 7.726680527218211e-08,
      "loss": 0.67750156,
      "num_input_tokens_seen": 164376635,
      "step": 7600,
      "time_per_iteration": 2.5623602867126465
    },
    {
      "auxiliary_loss_clip": 0.01170545,
      "auxiliary_loss_mlp": 0.01027772,
      "balance_loss_clip": 1.04601717,
      "balance_loss_mlp": 1.02122736,
      "epoch": 0.9139662117477304,
      "flos": 46281240714240.0,
      "grad_norm": 1.7458680156916204,
      "language_loss": 0.75511897,
      "learning_rate": 7.70525226108627e-08,
      "loss": 0.77710217,
      "num_input_tokens_seen": 164400305,
      "step": 7601,
      "time_per_iteration": 3.6371474266052246
    },
    {
      "auxiliary_loss_clip": 0.01226283,
      "auxiliary_loss_mlp": 0.01021644,
      "balance_loss_clip": 1.05041134,
      "balance_loss_mlp": 1.01469684,
      "epoch": 0.9140864546383695,
      "flos": 22273198819200.0,
      "grad_norm": 1.7951408337135109,
      "language_loss": 0.79891354,
      "learning_rate": 7.683853165914666e-08,
      "loss": 0.82139277,
      "num_input_tokens_seen": 164418075,
      "step": 7602,
      "time_per_iteration": 2.585395097732544
    },
    {
      "auxiliary_loss_clip": 0.01370677,
      "auxiliary_loss_mlp": 0.01024552,
      "balance_loss_clip": 1.04246283,
      "balance_loss_mlp": 1.01766753,
      "epoch": 0.9142066975290086,
      "flos": 17530009920000.0,
      "grad_norm": 1.9368947697342884,
      "language_loss": 0.77466702,
      "learning_rate": 7.662483244949602e-08,
      "loss": 0.79861927,
      "num_input_tokens_seen": 164435335,
      "step": 7603,
      "time_per_iteration": 2.7954394817352295
    },
    {
      "auxiliary_loss_clip": 0.01308838,
      "auxiliary_loss_mlp": 0.01025001,
      "balance_loss_clip": 1.04317975,
      "balance_loss_mlp": 1.0179584,
      "epoch": 0.9143269404196477,
      "flos": 17712148809600.0,
      "grad_norm": 2.3273893559971244,
      "language_loss": 0.80810618,
      "learning_rate": 7.641142501432951e-08,
      "loss": 0.83144456,
      "num_input_tokens_seen": 164451530,
      "step": 7604,
      "time_per_iteration": 2.7048652172088623
    },
    {
      "auxiliary_loss_clip": 0.01265099,
      "auxiliary_loss_mlp": 0.01024582,
      "balance_loss_clip": 1.04204416,
      "balance_loss_mlp": 1.01786733,
      "epoch": 0.9144471833102867,
      "flos": 33323414019840.0,
      "grad_norm": 1.8161032181983994,
      "language_loss": 0.73622668,
      "learning_rate": 7.619830938602013e-08,
      "loss": 0.75912356,
      "num_input_tokens_seen": 164472755,
      "step": 7605,
      "time_per_iteration": 2.8178369998931885
    },
    {
      "auxiliary_loss_clip": 0.01221206,
      "auxiliary_loss_mlp": 0.01026796,
      "balance_loss_clip": 1.04672039,
      "balance_loss_mlp": 1.01995647,
      "epoch": 0.9145674262009259,
      "flos": 21068970428160.0,
      "grad_norm": 2.15143373880573,
      "language_loss": 0.82386214,
      "learning_rate": 7.598548559689777e-08,
      "loss": 0.84634221,
      "num_input_tokens_seen": 164491155,
      "step": 7606,
      "time_per_iteration": 2.6555545330047607
    },
    {
      "auxiliary_loss_clip": 0.01314776,
      "auxiliary_loss_mlp": 0.01023959,
      "balance_loss_clip": 1.0421927,
      "balance_loss_mlp": 1.01727104,
      "epoch": 0.914687669091565,
      "flos": 16800269212800.0,
      "grad_norm": 3.297006619404753,
      "language_loss": 0.81293416,
      "learning_rate": 7.577295367924751e-08,
      "loss": 0.83632147,
      "num_input_tokens_seen": 164507555,
      "step": 7607,
      "time_per_iteration": 2.652890205383301
    },
    {
      "auxiliary_loss_clip": 0.01276852,
      "auxiliary_loss_mlp": 0.01027591,
      "balance_loss_clip": 1.04977846,
      "balance_loss_mlp": 1.02049839,
      "epoch": 0.914807911982204,
      "flos": 25773627012480.0,
      "grad_norm": 1.6464482731844308,
      "language_loss": 0.82236636,
      "learning_rate": 7.556071366531002e-08,
      "loss": 0.84541082,
      "num_input_tokens_seen": 164528525,
      "step": 7608,
      "time_per_iteration": 2.7772531509399414
    },
    {
      "auxiliary_loss_clip": 0.01221777,
      "auxiliary_loss_mlp": 0.01026886,
      "balance_loss_clip": 1.04893374,
      "balance_loss_mlp": 1.01985228,
      "epoch": 0.9149281548728432,
      "flos": 19208043636480.0,
      "grad_norm": 2.174642256299152,
      "language_loss": 0.78717339,
      "learning_rate": 7.53487655872822e-08,
      "loss": 0.80966002,
      "num_input_tokens_seen": 164547695,
      "step": 7609,
      "time_per_iteration": 2.59081768989563
    },
    {
      "auxiliary_loss_clip": 0.01374214,
      "auxiliary_loss_mlp": 0.01025727,
      "balance_loss_clip": 1.04117811,
      "balance_loss_mlp": 1.01899457,
      "epoch": 0.9150483977634822,
      "flos": 26870554500480.0,
      "grad_norm": 1.862325600411536,
      "language_loss": 0.74073219,
      "learning_rate": 7.513710947731656e-08,
      "loss": 0.76473165,
      "num_input_tokens_seen": 164568905,
      "step": 7610,
      "time_per_iteration": 2.803830623626709
    },
    {
      "auxiliary_loss_clip": 0.01266132,
      "auxiliary_loss_mlp": 0.0103444,
      "balance_loss_clip": 1.04510164,
      "balance_loss_mlp": 1.0274632,
      "epoch": 0.9151686406541213,
      "flos": 21908956953600.0,
      "grad_norm": 1.6853970045568514,
      "language_loss": 0.85172498,
      "learning_rate": 7.492574536752095e-08,
      "loss": 0.87473071,
      "num_input_tokens_seen": 164588895,
      "step": 7611,
      "time_per_iteration": 2.647684335708618
    },
    {
      "auxiliary_loss_clip": 0.01217942,
      "auxiliary_loss_mlp": 0.01025701,
      "balance_loss_clip": 1.04822779,
      "balance_loss_mlp": 1.018888,
      "epoch": 0.9152888835447605,
      "flos": 27308556944640.0,
      "grad_norm": 2.256371791795798,
      "language_loss": 0.78216332,
      "learning_rate": 7.471467328995907e-08,
      "loss": 0.80459976,
      "num_input_tokens_seen": 164607705,
      "step": 7612,
      "time_per_iteration": 2.688703775405884
    },
    {
      "auxiliary_loss_clip": 0.01423659,
      "auxiliary_loss_mlp": 0.01023334,
      "balance_loss_clip": 1.03873563,
      "balance_loss_mlp": 1.01669741,
      "epoch": 0.9154091264353995,
      "flos": 13370728510080.0,
      "grad_norm": 2.8761015641131626,
      "language_loss": 0.60217571,
      "learning_rate": 7.450389327665018e-08,
      "loss": 0.62664557,
      "num_input_tokens_seen": 164625540,
      "step": 7613,
      "time_per_iteration": 2.875971794128418
    },
    {
      "auxiliary_loss_clip": 0.01329267,
      "auxiliary_loss_mlp": 0.01027098,
      "balance_loss_clip": 1.05176497,
      "balance_loss_mlp": 1.02010345,
      "epoch": 0.9155293693260386,
      "flos": 20193037367040.0,
      "grad_norm": 2.3366484079259062,
      "language_loss": 0.67810369,
      "learning_rate": 7.429340535957029e-08,
      "loss": 0.70166731,
      "num_input_tokens_seen": 164640735,
      "step": 7614,
      "time_per_iteration": 3.444566011428833
    },
    {
      "auxiliary_loss_clip": 0.01273595,
      "auxiliary_loss_mlp": 0.01022533,
      "balance_loss_clip": 1.04636383,
      "balance_loss_mlp": 1.01592028,
      "epoch": 0.9156496122166777,
      "flos": 19354990176000.0,
      "grad_norm": 6.301799501942848,
      "language_loss": 0.7108407,
      "learning_rate": 7.40832095706494e-08,
      "loss": 0.73380196,
      "num_input_tokens_seen": 164657430,
      "step": 7615,
      "time_per_iteration": 2.6831982135772705
    },
    {
      "auxiliary_loss_clip": 0.01324428,
      "auxiliary_loss_mlp": 0.01029485,
      "balance_loss_clip": 1.0449338,
      "balance_loss_mlp": 1.02277327,
      "epoch": 0.9157698551073168,
      "flos": 21107287261440.0,
      "grad_norm": 2.134890897511472,
      "language_loss": 0.80454981,
      "learning_rate": 7.387330594177443e-08,
      "loss": 0.82808888,
      "num_input_tokens_seen": 164679505,
      "step": 7616,
      "time_per_iteration": 2.6993863582611084
    },
    {
      "auxiliary_loss_clip": 0.01318063,
      "auxiliary_loss_mlp": 0.01021223,
      "balance_loss_clip": 1.04342687,
      "balance_loss_mlp": 1.01438308,
      "epoch": 0.9158900979979558,
      "flos": 25193167228800.0,
      "grad_norm": 2.1260339494485607,
      "language_loss": 0.79404509,
      "learning_rate": 7.366369450478749e-08,
      "loss": 0.81743801,
      "num_input_tokens_seen": 164700615,
      "step": 7617,
      "time_per_iteration": 2.7368505001068115
    },
    {
      "auxiliary_loss_clip": 0.01316713,
      "auxiliary_loss_mlp": 0.01026839,
      "balance_loss_clip": 1.04425704,
      "balance_loss_mlp": 1.02024627,
      "epoch": 0.916010340888595,
      "flos": 30146648302080.0,
      "grad_norm": 2.948781655402579,
      "language_loss": 0.66694444,
      "learning_rate": 7.345437529148646e-08,
      "loss": 0.69037998,
      "num_input_tokens_seen": 164719625,
      "step": 7618,
      "time_per_iteration": 2.7175252437591553
    },
    {
      "auxiliary_loss_clip": 0.01321973,
      "auxiliary_loss_mlp": 0.01022858,
      "balance_loss_clip": 1.0441258,
      "balance_loss_mlp": 1.01638806,
      "epoch": 0.9161305837792341,
      "flos": 17091827907840.0,
      "grad_norm": 1.9744831754307444,
      "language_loss": 0.72649473,
      "learning_rate": 7.324534833362483e-08,
      "loss": 0.74994302,
      "num_input_tokens_seen": 164737200,
      "step": 7619,
      "time_per_iteration": 2.7575128078460693
    },
    {
      "auxiliary_loss_clip": 0.01269583,
      "auxiliary_loss_mlp": 0.01027513,
      "balance_loss_clip": 1.04737139,
      "balance_loss_mlp": 1.02086377,
      "epoch": 0.9162508266698731,
      "flos": 22893699288960.0,
      "grad_norm": 1.7112773577309563,
      "language_loss": 0.68870497,
      "learning_rate": 7.303661366291192e-08,
      "loss": 0.71167588,
      "num_input_tokens_seen": 164757870,
      "step": 7620,
      "time_per_iteration": 3.8878109455108643
    },
    {
      "auxiliary_loss_clip": 0.01372102,
      "auxiliary_loss_mlp": 0.01031197,
      "balance_loss_clip": 1.04147673,
      "balance_loss_mlp": 1.02438462,
      "epoch": 0.9163710695605123,
      "flos": 19974808287360.0,
      "grad_norm": 1.6767195427056085,
      "language_loss": 0.81866676,
      "learning_rate": 7.28281713110126e-08,
      "loss": 0.84269977,
      "num_input_tokens_seen": 164775945,
      "step": 7621,
      "time_per_iteration": 3.678748846054077
    },
    {
      "auxiliary_loss_clip": 0.01274193,
      "auxiliary_loss_mlp": 0.0102691,
      "balance_loss_clip": 1.04878068,
      "balance_loss_mlp": 1.0201118,
      "epoch": 0.9164913124511513,
      "flos": 22783812606720.0,
      "grad_norm": 2.1156315246112043,
      "language_loss": 0.77235758,
      "learning_rate": 7.262002130954759e-08,
      "loss": 0.79536867,
      "num_input_tokens_seen": 164794400,
      "step": 7622,
      "time_per_iteration": 2.7222959995269775
    },
    {
      "auxiliary_loss_clip": 0.01375073,
      "auxiliary_loss_mlp": 0.01026671,
      "balance_loss_clip": 1.04267216,
      "balance_loss_mlp": 1.02001905,
      "epoch": 0.9166115553417904,
      "flos": 24900854348160.0,
      "grad_norm": 1.7454492833268143,
      "language_loss": 0.78861547,
      "learning_rate": 7.241216369009296e-08,
      "loss": 0.81263292,
      "num_input_tokens_seen": 164814585,
      "step": 7623,
      "time_per_iteration": 3.6829864978790283
    },
    {
      "auxiliary_loss_clip": 0.01172497,
      "auxiliary_loss_mlp": 0.01031247,
      "balance_loss_clip": 1.04746616,
      "balance_loss_mlp": 1.02421403,
      "epoch": 0.9167317982324296,
      "flos": 25702919089920.0,
      "grad_norm": 1.9275015611482877,
      "language_loss": 0.66671312,
      "learning_rate": 7.220459848418037e-08,
      "loss": 0.68875057,
      "num_input_tokens_seen": 164834660,
      "step": 7624,
      "time_per_iteration": 2.61224627494812
    },
    {
      "auxiliary_loss_clip": 0.01172284,
      "auxiliary_loss_mlp": 0.01026252,
      "balance_loss_clip": 1.05051088,
      "balance_loss_mlp": 1.01933455,
      "epoch": 0.9168520411230686,
      "flos": 15632813370240.0,
      "grad_norm": 2.1029756957926056,
      "language_loss": 0.79768789,
      "learning_rate": 7.199732572329708e-08,
      "loss": 0.81967324,
      "num_input_tokens_seen": 164852560,
      "step": 7625,
      "time_per_iteration": 2.562105417251587
    },
    {
      "auxiliary_loss_clip": 0.01223317,
      "auxiliary_loss_mlp": 0.01025809,
      "balance_loss_clip": 1.04381323,
      "balance_loss_mlp": 1.01910615,
      "epoch": 0.9169722840137077,
      "flos": 30258151096320.0,
      "grad_norm": 2.760964531289636,
      "language_loss": 0.76240265,
      "learning_rate": 7.179034543888684e-08,
      "loss": 0.78489393,
      "num_input_tokens_seen": 164872065,
      "step": 7626,
      "time_per_iteration": 3.655902862548828
    },
    {
      "auxiliary_loss_clip": 0.01226556,
      "auxiliary_loss_mlp": 0.01025046,
      "balance_loss_clip": 1.04811311,
      "balance_loss_mlp": 1.01861513,
      "epoch": 0.9170925269043467,
      "flos": 22491643380480.0,
      "grad_norm": 2.1816700372421667,
      "language_loss": 0.77638471,
      "learning_rate": 7.158365766234808e-08,
      "loss": 0.79890072,
      "num_input_tokens_seen": 164890915,
      "step": 7627,
      "time_per_iteration": 2.6988086700439453
    },
    {
      "auxiliary_loss_clip": 0.01316525,
      "auxiliary_loss_mlp": 0.01027964,
      "balance_loss_clip": 1.04022062,
      "balance_loss_mlp": 1.02035594,
      "epoch": 0.9172127697949859,
      "flos": 22893914770560.0,
      "grad_norm": 2.247379273378707,
      "language_loss": 0.72328025,
      "learning_rate": 7.137726242503527e-08,
      "loss": 0.74672508,
      "num_input_tokens_seen": 164909835,
      "step": 7628,
      "time_per_iteration": 2.660611152648926
    },
    {
      "auxiliary_loss_clip": 0.01221647,
      "auxiliary_loss_mlp": 0.02567981,
      "balance_loss_clip": 1.0487113,
      "balance_loss_mlp": 0.99992299,
      "epoch": 0.917333012685625,
      "flos": 17451867882240.0,
      "grad_norm": 6.452056485032234,
      "language_loss": 0.78294754,
      "learning_rate": 7.11711597582585e-08,
      "loss": 0.82084382,
      "num_input_tokens_seen": 164927195,
      "step": 7629,
      "time_per_iteration": 2.649448871612549
    },
    {
      "auxiliary_loss_clip": 0.01320037,
      "auxiliary_loss_mlp": 0.01024766,
      "balance_loss_clip": 1.04012525,
      "balance_loss_mlp": 1.01815891,
      "epoch": 0.917453255576264,
      "flos": 14318949692160.0,
      "grad_norm": 2.9769075127472857,
      "language_loss": 0.80159128,
      "learning_rate": 7.096534969328271e-08,
      "loss": 0.82503933,
      "num_input_tokens_seen": 164944640,
      "step": 7630,
      "time_per_iteration": 2.6509616374969482
    },
    {
      "auxiliary_loss_clip": 0.0127458,
      "auxiliary_loss_mlp": 0.01022663,
      "balance_loss_clip": 1.04336071,
      "balance_loss_mlp": 1.01592445,
      "epoch": 0.9175734984669032,
      "flos": 20741177888640.0,
      "grad_norm": 1.9158430202860188,
      "language_loss": 0.8369109,
      "learning_rate": 7.075983226132987e-08,
      "loss": 0.85988331,
      "num_input_tokens_seen": 164963570,
      "step": 7631,
      "time_per_iteration": 2.7028024196624756
    },
    {
      "auxiliary_loss_clip": 0.01277672,
      "auxiliary_loss_mlp": 0.02569206,
      "balance_loss_clip": 1.04585576,
      "balance_loss_mlp": 0.99991721,
      "epoch": 0.9176937413575422,
      "flos": 14830497233280.0,
      "grad_norm": 3.233619369603297,
      "language_loss": 0.79308569,
      "learning_rate": 7.055460749357656e-08,
      "loss": 0.83155453,
      "num_input_tokens_seen": 164979850,
      "step": 7632,
      "time_per_iteration": 2.7047464847564697
    },
    {
      "auxiliary_loss_clip": 0.01268613,
      "auxiliary_loss_mlp": 0.01025657,
      "balance_loss_clip": 1.04753482,
      "balance_loss_mlp": 1.01814055,
      "epoch": 0.9178139842481813,
      "flos": 18474603828480.0,
      "grad_norm": 3.165785231860029,
      "language_loss": 0.7017293,
      "learning_rate": 7.034967542115521e-08,
      "loss": 0.72467196,
      "num_input_tokens_seen": 164998115,
      "step": 7633,
      "time_per_iteration": 2.616337537765503
    },
    {
      "auxiliary_loss_clip": 0.01218409,
      "auxiliary_loss_mlp": 0.02565981,
      "balance_loss_clip": 1.04640913,
      "balance_loss_mlp": 0.99993908,
      "epoch": 0.9179342271388204,
      "flos": 20047455544320.0,
      "grad_norm": 2.6501339691547976,
      "language_loss": 0.75613189,
      "learning_rate": 7.014503607515388e-08,
      "loss": 0.79397583,
      "num_input_tokens_seen": 165017420,
      "step": 7634,
      "time_per_iteration": 2.6475989818573
    },
    {
      "auxiliary_loss_clip": 0.01269648,
      "auxiliary_loss_mlp": 0.01027945,
      "balance_loss_clip": 1.04885173,
      "balance_loss_mlp": 1.02063775,
      "epoch": 0.9180544700294595,
      "flos": 24676232647680.0,
      "grad_norm": 3.1631165642956525,
      "language_loss": 0.68382359,
      "learning_rate": 6.994068948661592e-08,
      "loss": 0.70679951,
      "num_input_tokens_seen": 165035575,
      "step": 7635,
      "time_per_iteration": 2.6665782928466797
    },
    {
      "auxiliary_loss_clip": 0.01223024,
      "auxiliary_loss_mlp": 0.01030009,
      "balance_loss_clip": 1.04875612,
      "balance_loss_mlp": 1.02227867,
      "epoch": 0.9181747129200986,
      "flos": 16727478301440.0,
      "grad_norm": 1.985921368413001,
      "language_loss": 0.76900768,
      "learning_rate": 6.973663568654142e-08,
      "loss": 0.791538,
      "num_input_tokens_seen": 165053280,
      "step": 7636,
      "time_per_iteration": 2.709414005279541
    },
    {
      "auxiliary_loss_clip": 0.01174557,
      "auxiliary_loss_mlp": 0.01027371,
      "balance_loss_clip": 1.05110741,
      "balance_loss_mlp": 1.02047133,
      "epoch": 0.9182949558107377,
      "flos": 24271626873600.0,
      "grad_norm": 2.2495263401601835,
      "language_loss": 0.65652144,
      "learning_rate": 6.953287470588386e-08,
      "loss": 0.67854071,
      "num_input_tokens_seen": 165071235,
      "step": 7637,
      "time_per_iteration": 2.584444999694824
    },
    {
      "auxiliary_loss_clip": 0.01224121,
      "auxiliary_loss_mlp": 0.01025254,
      "balance_loss_clip": 1.04515314,
      "balance_loss_mlp": 1.01796174,
      "epoch": 0.9184151987013768,
      "flos": 22082117443200.0,
      "grad_norm": 2.292178211436983,
      "language_loss": 0.86096376,
      "learning_rate": 6.932940657555452e-08,
      "loss": 0.88345748,
      "num_input_tokens_seen": 165087365,
      "step": 7638,
      "time_per_iteration": 2.619983196258545
    },
    {
      "auxiliary_loss_clip": 0.01167845,
      "auxiliary_loss_mlp": 0.01021876,
      "balance_loss_clip": 1.0478313,
      "balance_loss_mlp": 1.01545024,
      "epoch": 0.9185354415920158,
      "flos": 32166732257280.0,
      "grad_norm": 1.4853165154290404,
      "language_loss": 0.76617765,
      "learning_rate": 6.912623132641938e-08,
      "loss": 0.78807485,
      "num_input_tokens_seen": 165112455,
      "step": 7639,
      "time_per_iteration": 2.720280408859253
    },
    {
      "auxiliary_loss_clip": 0.01275896,
      "auxiliary_loss_mlp": 0.01026907,
      "balance_loss_clip": 1.04598534,
      "balance_loss_mlp": 1.01965356,
      "epoch": 0.918655684482655,
      "flos": 20997831542400.0,
      "grad_norm": 2.0311636903400476,
      "language_loss": 0.76998258,
      "learning_rate": 6.892334898929952e-08,
      "loss": 0.79301065,
      "num_input_tokens_seen": 165132700,
      "step": 7640,
      "time_per_iteration": 2.728938579559326
    },
    {
      "auxiliary_loss_clip": 0.01217613,
      "auxiliary_loss_mlp": 0.01023944,
      "balance_loss_clip": 1.04508507,
      "balance_loss_mlp": 1.01737547,
      "epoch": 0.918775927373294,
      "flos": 15560704817280.0,
      "grad_norm": 2.0202243670884448,
      "language_loss": 0.84753656,
      "learning_rate": 6.872075959497236e-08,
      "loss": 0.86995208,
      "num_input_tokens_seen": 165151475,
      "step": 7641,
      "time_per_iteration": 2.5756380558013916
    },
    {
      "auxiliary_loss_clip": 0.01122694,
      "auxiliary_loss_mlp": 0.01024638,
      "balance_loss_clip": 1.04605556,
      "balance_loss_mlp": 1.01780105,
      "epoch": 0.9188961702639331,
      "flos": 29934057657600.0,
      "grad_norm": 4.817342337260383,
      "language_loss": 0.82826632,
      "learning_rate": 6.85184631741702e-08,
      "loss": 0.84973967,
      "num_input_tokens_seen": 165172040,
      "step": 7642,
      "time_per_iteration": 2.683485269546509
    },
    {
      "auxiliary_loss_clip": 0.01216449,
      "auxiliary_loss_mlp": 0.01027028,
      "balance_loss_clip": 1.04508853,
      "balance_loss_mlp": 1.0200218,
      "epoch": 0.9190164131545723,
      "flos": 20701244943360.0,
      "grad_norm": 2.0683013592185313,
      "language_loss": 0.77649605,
      "learning_rate": 6.831645975758161e-08,
      "loss": 0.79893082,
      "num_input_tokens_seen": 165189980,
      "step": 7643,
      "time_per_iteration": 2.5841498374938965
    },
    {
      "auxiliary_loss_clip": 0.01267739,
      "auxiliary_loss_mlp": 0.01028041,
      "balance_loss_clip": 1.04525185,
      "balance_loss_mlp": 1.02057886,
      "epoch": 0.9191366560452113,
      "flos": 25629912696960.0,
      "grad_norm": 5.9666980166046,
      "language_loss": 0.67322898,
      "learning_rate": 6.811474937585026e-08,
      "loss": 0.69618678,
      "num_input_tokens_seen": 165209770,
      "step": 7644,
      "time_per_iteration": 2.7142460346221924
    },
    {
      "auxiliary_loss_clip": 0.01318085,
      "auxiliary_loss_mlp": 0.01025401,
      "balance_loss_clip": 1.04414177,
      "balance_loss_mlp": 1.01923442,
      "epoch": 0.9192568989358504,
      "flos": 21434325615360.0,
      "grad_norm": 1.6448751413726141,
      "language_loss": 0.79166591,
      "learning_rate": 6.79133320595755e-08,
      "loss": 0.81510079,
      "num_input_tokens_seen": 165229690,
      "step": 7645,
      "time_per_iteration": 2.6831910610198975
    },
    {
      "auxiliary_loss_clip": 0.01273314,
      "auxiliary_loss_mlp": 0.0102307,
      "balance_loss_clip": 1.04812002,
      "balance_loss_mlp": 1.01647806,
      "epoch": 0.9193771418264896,
      "flos": 23185078416000.0,
      "grad_norm": 1.8958644704220184,
      "language_loss": 0.75231707,
      "learning_rate": 6.771220783931198e-08,
      "loss": 0.77528089,
      "num_input_tokens_seen": 165249850,
      "step": 7646,
      "time_per_iteration": 4.47619104385376
    },
    {
      "auxiliary_loss_clip": 0.01460073,
      "auxiliary_loss_mlp": 0.02506179,
      "balance_loss_clip": 1.03284943,
      "balance_loss_mlp": 0.99990672,
      "epoch": 0.9194973847171286,
      "flos": 70582963184640.0,
      "grad_norm": 0.8501237492570174,
      "language_loss": 0.64562428,
      "learning_rate": 6.751137674556994e-08,
      "loss": 0.68528676,
      "num_input_tokens_seen": 165310235,
      "step": 7647,
      "time_per_iteration": 3.8085343837738037
    },
    {
      "auxiliary_loss_clip": 0.01225331,
      "auxiliary_loss_mlp": 0.01026706,
      "balance_loss_clip": 1.04545951,
      "balance_loss_mlp": 1.01982141,
      "epoch": 0.9196176276077677,
      "flos": 14720682378240.0,
      "grad_norm": 2.14933106354972,
      "language_loss": 0.77659512,
      "learning_rate": 6.731083880881572e-08,
      "loss": 0.79911542,
      "num_input_tokens_seen": 165326455,
      "step": 7648,
      "time_per_iteration": 3.0257620811462402
    },
    {
      "auxiliary_loss_clip": 0.01272112,
      "auxiliary_loss_mlp": 0.01025692,
      "balance_loss_clip": 1.04872465,
      "balance_loss_mlp": 1.01901007,
      "epoch": 0.9197378704984068,
      "flos": 23294893271040.0,
      "grad_norm": 2.1587517348349694,
      "language_loss": 0.80774486,
      "learning_rate": 6.711059405947072e-08,
      "loss": 0.83072287,
      "num_input_tokens_seen": 165344645,
      "step": 7649,
      "time_per_iteration": 3.5893969535827637
    },
    {
      "auxiliary_loss_clip": 0.01312697,
      "auxiliary_loss_mlp": 0.0102229,
      "balance_loss_clip": 1.04435229,
      "balance_loss_mlp": 1.01560521,
      "epoch": 0.9198581133890459,
      "flos": 20302564913280.0,
      "grad_norm": 2.3922156479850587,
      "language_loss": 0.77253211,
      "learning_rate": 6.691064252791156e-08,
      "loss": 0.79588199,
      "num_input_tokens_seen": 165364120,
      "step": 7650,
      "time_per_iteration": 2.722959518432617
    },
    {
      "auxiliary_loss_clip": 0.01358275,
      "auxiliary_loss_mlp": 0.01026407,
      "balance_loss_clip": 1.04144871,
      "balance_loss_mlp": 1.01940358,
      "epoch": 0.9199783562796849,
      "flos": 17675663569920.0,
      "grad_norm": 1.6051689981763821,
      "language_loss": 0.77922815,
      "learning_rate": 6.67109842444713e-08,
      "loss": 0.80307496,
      "num_input_tokens_seen": 165383050,
      "step": 7651,
      "time_per_iteration": 2.776198148727417
    },
    {
      "auxiliary_loss_clip": 0.01227318,
      "auxiliary_loss_mlp": 0.02566657,
      "balance_loss_clip": 1.0515759,
      "balance_loss_mlp": 0.99993849,
      "epoch": 0.9200985991703241,
      "flos": 17676022705920.0,
      "grad_norm": 1.998464192675731,
      "language_loss": 0.76622158,
      "learning_rate": 6.651161923943704e-08,
      "loss": 0.80416131,
      "num_input_tokens_seen": 165400955,
      "step": 7652,
      "time_per_iteration": 4.059560060501099
    },
    {
      "auxiliary_loss_clip": 0.01219453,
      "auxiliary_loss_mlp": 0.01026428,
      "balance_loss_clip": 1.0447433,
      "balance_loss_mlp": 1.01881337,
      "epoch": 0.9202188420609632,
      "flos": 20996574566400.0,
      "grad_norm": 1.67228467511947,
      "language_loss": 0.76783705,
      "learning_rate": 6.631254754305326e-08,
      "loss": 0.79029584,
      "num_input_tokens_seen": 165420415,
      "step": 7653,
      "time_per_iteration": 2.6930644512176514
    },
    {
      "auxiliary_loss_clip": 0.01172676,
      "auxiliary_loss_mlp": 0.0102129,
      "balance_loss_clip": 1.0479728,
      "balance_loss_mlp": 1.01464367,
      "epoch": 0.9203390849516022,
      "flos": 13918222586880.0,
      "grad_norm": 2.26555383502116,
      "language_loss": 0.78172702,
      "learning_rate": 6.611376918551848e-08,
      "loss": 0.80366665,
      "num_input_tokens_seen": 165439200,
      "step": 7654,
      "time_per_iteration": 2.580793619155884
    },
    {
      "auxiliary_loss_clip": 0.01218444,
      "auxiliary_loss_mlp": 0.02567134,
      "balance_loss_clip": 1.0421176,
      "balance_loss_mlp": 0.99988937,
      "epoch": 0.9204593278422414,
      "flos": 21175912195200.0,
      "grad_norm": 3.0334438680721822,
      "language_loss": 0.79789042,
      "learning_rate": 6.591528419698744e-08,
      "loss": 0.83574617,
      "num_input_tokens_seen": 165458985,
      "step": 7655,
      "time_per_iteration": 2.6849617958068848
    },
    {
      "auxiliary_loss_clip": 0.01272939,
      "auxiliary_loss_mlp": 0.01025126,
      "balance_loss_clip": 1.04387188,
      "balance_loss_mlp": 1.01866531,
      "epoch": 0.9205795707328804,
      "flos": 14501375890560.0,
      "grad_norm": 2.2706713730159582,
      "language_loss": 0.83580172,
      "learning_rate": 6.571709260756986e-08,
      "loss": 0.85878241,
      "num_input_tokens_seen": 165475630,
      "step": 7656,
      "time_per_iteration": 2.634756326675415
    },
    {
      "auxiliary_loss_clip": 0.0122522,
      "auxiliary_loss_mlp": 0.01025291,
      "balance_loss_clip": 1.05240607,
      "balance_loss_mlp": 1.01831436,
      "epoch": 0.9206998136235195,
      "flos": 22417559579520.0,
      "grad_norm": 15.807396454302408,
      "language_loss": 0.76015913,
      "learning_rate": 6.551919444733122e-08,
      "loss": 0.7826643,
      "num_input_tokens_seen": 165493445,
      "step": 7657,
      "time_per_iteration": 2.733100414276123
    },
    {
      "auxiliary_loss_clip": 0.01271098,
      "auxiliary_loss_mlp": 0.01026349,
      "balance_loss_clip": 1.04711115,
      "balance_loss_mlp": 1.01949143,
      "epoch": 0.9208200565141585,
      "flos": 53358407544960.0,
      "grad_norm": 2.1112556293668585,
      "language_loss": 0.66032505,
      "learning_rate": 6.53215897462931e-08,
      "loss": 0.68329954,
      "num_input_tokens_seen": 165517200,
      "step": 7658,
      "time_per_iteration": 3.0016939640045166
    },
    {
      "auxiliary_loss_clip": 0.01220283,
      "auxiliary_loss_mlp": 0.0102846,
      "balance_loss_clip": 1.04574156,
      "balance_loss_mlp": 1.02156711,
      "epoch": 0.9209402994047977,
      "flos": 30589139946240.0,
      "grad_norm": 2.998306283332992,
      "language_loss": 0.74829656,
      "learning_rate": 6.512427853443103e-08,
      "loss": 0.77078402,
      "num_input_tokens_seen": 165539280,
      "step": 7659,
      "time_per_iteration": 2.6631088256835938
    },
    {
      "auxiliary_loss_clip": 0.01226243,
      "auxiliary_loss_mlp": 0.01022481,
      "balance_loss_clip": 1.04753995,
      "balance_loss_mlp": 1.01593387,
      "epoch": 0.9210605422954368,
      "flos": 29132711187840.0,
      "grad_norm": 1.6626931397643498,
      "language_loss": 0.7600981,
      "learning_rate": 6.492726084167799e-08,
      "loss": 0.78258538,
      "num_input_tokens_seen": 165561395,
      "step": 7660,
      "time_per_iteration": 2.7732231616973877
    },
    {
      "auxiliary_loss_clip": 0.01059711,
      "auxiliary_loss_mlp": 0.00999598,
      "balance_loss_clip": 1.00574422,
      "balance_loss_mlp": 0.99872828,
      "epoch": 0.9211807851860758,
      "flos": 54853838472960.0,
      "grad_norm": 0.7738463302921195,
      "language_loss": 0.57456428,
      "learning_rate": 6.473053669792072e-08,
      "loss": 0.59515738,
      "num_input_tokens_seen": 165616085,
      "step": 7661,
      "time_per_iteration": 3.025599241256714
    },
    {
      "auxiliary_loss_clip": 0.01223452,
      "auxiliary_loss_mlp": 0.01025222,
      "balance_loss_clip": 1.04711771,
      "balance_loss_mlp": 1.01848662,
      "epoch": 0.921301028076715,
      "flos": 19201974238080.0,
      "grad_norm": 3.470860934129358,
      "language_loss": 0.72978812,
      "learning_rate": 6.453410613300248e-08,
      "loss": 0.75227487,
      "num_input_tokens_seen": 165634015,
      "step": 7662,
      "time_per_iteration": 2.687568187713623
    },
    {
      "auxiliary_loss_clip": 0.01418205,
      "auxiliary_loss_mlp": 0.01022506,
      "balance_loss_clip": 1.04125524,
      "balance_loss_mlp": 1.01592016,
      "epoch": 0.921421270967354,
      "flos": 27526893765120.0,
      "grad_norm": 1.9068201663238014,
      "language_loss": 0.58195066,
      "learning_rate": 6.43379691767214e-08,
      "loss": 0.60635781,
      "num_input_tokens_seen": 165653220,
      "step": 7663,
      "time_per_iteration": 2.7942821979522705
    },
    {
      "auxiliary_loss_clip": 0.01274428,
      "auxiliary_loss_mlp": 0.00999458,
      "balance_loss_clip": 1.00766158,
      "balance_loss_mlp": 0.99861139,
      "epoch": 0.9215415138579931,
      "flos": 70209311955840.0,
      "grad_norm": 0.7202495577259429,
      "language_loss": 0.55046725,
      "learning_rate": 6.414212585883105e-08,
      "loss": 0.57320619,
      "num_input_tokens_seen": 165715850,
      "step": 7664,
      "time_per_iteration": 3.3678667545318604
    },
    {
      "auxiliary_loss_clip": 0.01274705,
      "auxiliary_loss_mlp": 0.01029629,
      "balance_loss_clip": 1.04661083,
      "balance_loss_mlp": 1.02262831,
      "epoch": 0.9216617567486323,
      "flos": 35553107790720.0,
      "grad_norm": 1.6195241331758947,
      "language_loss": 0.69930482,
      "learning_rate": 6.394657620904143e-08,
      "loss": 0.72234809,
      "num_input_tokens_seen": 165738960,
      "step": 7665,
      "time_per_iteration": 2.802504777908325
    },
    {
      "auxiliary_loss_clip": 0.01175245,
      "auxiliary_loss_mlp": 0.01025739,
      "balance_loss_clip": 1.0489397,
      "balance_loss_mlp": 1.01874495,
      "epoch": 0.9217819996392713,
      "flos": 29533330552320.0,
      "grad_norm": 1.8995984746531938,
      "language_loss": 0.71318138,
      "learning_rate": 6.375132025701657e-08,
      "loss": 0.73519117,
      "num_input_tokens_seen": 165761260,
      "step": 7666,
      "time_per_iteration": 2.6561758518218994
    },
    {
      "auxiliary_loss_clip": 0.01178049,
      "auxiliary_loss_mlp": 0.01028017,
      "balance_loss_clip": 1.05194199,
      "balance_loss_mlp": 1.0209713,
      "epoch": 0.9219022425299104,
      "flos": 14574669592320.0,
      "grad_norm": 4.254408088398737,
      "language_loss": 0.69506627,
      "learning_rate": 6.355635803237724e-08,
      "loss": 0.71712697,
      "num_input_tokens_seen": 165776960,
      "step": 7667,
      "time_per_iteration": 2.629960298538208
    },
    {
      "auxiliary_loss_clip": 0.01223089,
      "auxiliary_loss_mlp": 0.01026641,
      "balance_loss_clip": 1.04634476,
      "balance_loss_mlp": 1.01961088,
      "epoch": 0.9220224854205495,
      "flos": 18077503996800.0,
      "grad_norm": 1.9310618555927483,
      "language_loss": 0.799523,
      "learning_rate": 6.336168956469867e-08,
      "loss": 0.82202035,
      "num_input_tokens_seen": 165795435,
      "step": 7668,
      "time_per_iteration": 2.6667723655700684
    },
    {
      "auxiliary_loss_clip": 0.01265709,
      "auxiliary_loss_mlp": 0.01024111,
      "balance_loss_clip": 1.04516649,
      "balance_loss_mlp": 1.01749182,
      "epoch": 0.9221427283111886,
      "flos": 24790464875520.0,
      "grad_norm": 1.6411523776650996,
      "language_loss": 0.7191903,
      "learning_rate": 6.316731488351168e-08,
      "loss": 0.74208844,
      "num_input_tokens_seen": 165816625,
      "step": 7669,
      "time_per_iteration": 2.7147057056427
    },
    {
      "auxiliary_loss_clip": 0.0122288,
      "auxiliary_loss_mlp": 0.01025078,
      "balance_loss_clip": 1.0485636,
      "balance_loss_mlp": 1.01795268,
      "epoch": 0.9222629712018277,
      "flos": 13845036625920.0,
      "grad_norm": 1.7786618687253717,
      "language_loss": 0.63423848,
      "learning_rate": 6.297323401830334e-08,
      "loss": 0.65671808,
      "num_input_tokens_seen": 165835410,
      "step": 7670,
      "time_per_iteration": 2.619178056716919
    },
    {
      "auxiliary_loss_clip": 0.01223129,
      "auxiliary_loss_mlp": 0.01020799,
      "balance_loss_clip": 1.04679465,
      "balance_loss_mlp": 1.01378095,
      "epoch": 0.9223832140924668,
      "flos": 21616177196160.0,
      "grad_norm": 2.2254605605740645,
      "language_loss": 0.68947113,
      "learning_rate": 6.277944699851523e-08,
      "loss": 0.71191043,
      "num_input_tokens_seen": 165854930,
      "step": 7671,
      "time_per_iteration": 3.8282272815704346
    },
    {
      "auxiliary_loss_clip": 0.01173528,
      "auxiliary_loss_mlp": 0.01023101,
      "balance_loss_clip": 1.04925799,
      "balance_loss_mlp": 1.01594806,
      "epoch": 0.9225034569831059,
      "flos": 21142084561920.0,
      "grad_norm": 1.9463046491098237,
      "language_loss": 0.73851609,
      "learning_rate": 6.25859538535447e-08,
      "loss": 0.76048237,
      "num_input_tokens_seen": 165875725,
      "step": 7672,
      "time_per_iteration": 2.637486219406128
    },
    {
      "auxiliary_loss_clip": 0.01266246,
      "auxiliary_loss_mlp": 0.01025176,
      "balance_loss_clip": 1.04543066,
      "balance_loss_mlp": 1.01812458,
      "epoch": 0.9226236998737449,
      "flos": 12495046844160.0,
      "grad_norm": 14.046082153230564,
      "language_loss": 0.77396375,
      "learning_rate": 6.239275461274474e-08,
      "loss": 0.79687798,
      "num_input_tokens_seen": 165892100,
      "step": 7673,
      "time_per_iteration": 3.6801514625549316
    },
    {
      "auxiliary_loss_clip": 0.01221687,
      "auxiliary_loss_mlp": 0.01024508,
      "balance_loss_clip": 1.04797602,
      "balance_loss_mlp": 1.0176388,
      "epoch": 0.9227439427643841,
      "flos": 26214071581440.0,
      "grad_norm": 1.9502130959820356,
      "language_loss": 0.85965276,
      "learning_rate": 6.219984930542299e-08,
      "loss": 0.88211465,
      "num_input_tokens_seen": 165912840,
      "step": 7674,
      "time_per_iteration": 2.7060203552246094
    },
    {
      "auxiliary_loss_clip": 0.01225628,
      "auxiliary_loss_mlp": 0.01027493,
      "balance_loss_clip": 1.04725337,
      "balance_loss_mlp": 1.02113593,
      "epoch": 0.9228641856550232,
      "flos": 17967581400960.0,
      "grad_norm": 2.186827642523092,
      "language_loss": 0.75872844,
      "learning_rate": 6.200723796084383e-08,
      "loss": 0.78125966,
      "num_input_tokens_seen": 165930935,
      "step": 7675,
      "time_per_iteration": 3.6751155853271484
    },
    {
      "auxiliary_loss_clip": 0.01227925,
      "auxiliary_loss_mlp": 0.00998769,
      "balance_loss_clip": 1.00637376,
      "balance_loss_mlp": 0.99780953,
      "epoch": 0.9229844285456622,
      "flos": 70420609710720.0,
      "grad_norm": 0.7616374138038766,
      "language_loss": 0.62986958,
      "learning_rate": 6.181492060822546e-08,
      "loss": 0.65213656,
      "num_input_tokens_seen": 165991110,
      "step": 7676,
      "time_per_iteration": 3.1840269565582275
    },
    {
      "auxiliary_loss_clip": 0.01371811,
      "auxiliary_loss_mlp": 0.010305,
      "balance_loss_clip": 1.04186141,
      "balance_loss_mlp": 1.0233053,
      "epoch": 0.9231046714363014,
      "flos": 17967832796160.0,
      "grad_norm": 2.0612843727714836,
      "language_loss": 0.82026619,
      "learning_rate": 6.162289727674274e-08,
      "loss": 0.8442893,
      "num_input_tokens_seen": 166008790,
      "step": 7677,
      "time_per_iteration": 2.7368199825286865
    },
    {
      "auxiliary_loss_clip": 0.01320509,
      "auxiliary_loss_mlp": 0.01025488,
      "balance_loss_clip": 1.0452323,
      "balance_loss_mlp": 1.01947117,
      "epoch": 0.9232249143269404,
      "flos": 17858233422720.0,
      "grad_norm": 2.4349435396949644,
      "language_loss": 0.87797487,
      "learning_rate": 6.143116799552527e-08,
      "loss": 0.9014349,
      "num_input_tokens_seen": 166025035,
      "step": 7678,
      "time_per_iteration": 3.6829094886779785
    },
    {
      "auxiliary_loss_clip": 0.01226006,
      "auxiliary_loss_mlp": 0.01026091,
      "balance_loss_clip": 1.04879463,
      "balance_loss_mlp": 1.01928425,
      "epoch": 0.9233451572175795,
      "flos": 23404384903680.0,
      "grad_norm": 2.247363750575837,
      "language_loss": 0.56475681,
      "learning_rate": 6.123973279365802e-08,
      "loss": 0.58727777,
      "num_input_tokens_seen": 166044010,
      "step": 7679,
      "time_per_iteration": 2.716836929321289
    },
    {
      "auxiliary_loss_clip": 0.01225648,
      "auxiliary_loss_mlp": 0.01027205,
      "balance_loss_clip": 1.04998899,
      "balance_loss_mlp": 1.02071738,
      "epoch": 0.9234654001082186,
      "flos": 17999326045440.0,
      "grad_norm": 1.807401491478104,
      "language_loss": 0.77886069,
      "learning_rate": 6.10485917001824e-08,
      "loss": 0.80138922,
      "num_input_tokens_seen": 166061865,
      "step": 7680,
      "time_per_iteration": 2.6723430156707764
    },
    {
      "auxiliary_loss_clip": 0.01175114,
      "auxiliary_loss_mlp": 0.01027006,
      "balance_loss_clip": 1.04453993,
      "balance_loss_mlp": 1.02065539,
      "epoch": 0.9235856429988577,
      "flos": 24750747411840.0,
      "grad_norm": 1.6487303034629541,
      "language_loss": 0.81110525,
      "learning_rate": 6.085774474409322e-08,
      "loss": 0.83312643,
      "num_input_tokens_seen": 166082425,
      "step": 7681,
      "time_per_iteration": 2.6461377143859863
    },
    {
      "auxiliary_loss_clip": 0.01269146,
      "auxiliary_loss_mlp": 0.01022053,
      "balance_loss_clip": 1.04811537,
      "balance_loss_mlp": 1.01548481,
      "epoch": 0.9237058858894968,
      "flos": 14099894599680.0,
      "grad_norm": 2.1838280236195033,
      "language_loss": 0.70045346,
      "learning_rate": 6.066719195434267e-08,
      "loss": 0.72336543,
      "num_input_tokens_seen": 166100225,
      "step": 7682,
      "time_per_iteration": 2.7047085762023926
    },
    {
      "auxiliary_loss_clip": 0.01224989,
      "auxiliary_loss_mlp": 0.01022807,
      "balance_loss_clip": 1.04950988,
      "balance_loss_mlp": 1.0156306,
      "epoch": 0.9238261287801359,
      "flos": 28694529175680.0,
      "grad_norm": 2.0082459652859095,
      "language_loss": 0.66301,
      "learning_rate": 6.047693335983717e-08,
      "loss": 0.68548799,
      "num_input_tokens_seen": 166122570,
      "step": 7683,
      "time_per_iteration": 2.776366710662842
    },
    {
      "auxiliary_loss_clip": 0.01224373,
      "auxiliary_loss_mlp": 0.01027985,
      "balance_loss_clip": 1.04665995,
      "balance_loss_mlp": 1.02071297,
      "epoch": 0.923946371670775,
      "flos": 23111856541440.0,
      "grad_norm": 2.3833625705966797,
      "language_loss": 0.82276267,
      "learning_rate": 6.028696898943853e-08,
      "loss": 0.84528625,
      "num_input_tokens_seen": 166141630,
      "step": 7684,
      "time_per_iteration": 2.6630938053131104
    },
    {
      "auxiliary_loss_clip": 0.01272297,
      "auxiliary_loss_mlp": 0.0256893,
      "balance_loss_clip": 1.04495859,
      "balance_loss_mlp": 0.99992216,
      "epoch": 0.924066614561414,
      "flos": 21867120587520.0,
      "grad_norm": 2.6064861629315335,
      "language_loss": 0.70919275,
      "learning_rate": 6.00972988719648e-08,
      "loss": 0.74760503,
      "num_input_tokens_seen": 166159865,
      "step": 7685,
      "time_per_iteration": 2.6866466999053955
    },
    {
      "auxiliary_loss_clip": 0.01321234,
      "auxiliary_loss_mlp": 0.02568285,
      "balance_loss_clip": 1.04440653,
      "balance_loss_mlp": 0.99991548,
      "epoch": 0.9241868574520532,
      "flos": 28511887495680.0,
      "grad_norm": 2.6547624975412485,
      "language_loss": 0.70908141,
      "learning_rate": 5.990792303618807e-08,
      "loss": 0.74797654,
      "num_input_tokens_seen": 166179445,
      "step": 7686,
      "time_per_iteration": 2.7846834659576416
    },
    {
      "auxiliary_loss_clip": 0.01327692,
      "auxiliary_loss_mlp": 0.01025936,
      "balance_loss_clip": 1.04992151,
      "balance_loss_mlp": 1.01931143,
      "epoch": 0.9243071003426923,
      "flos": 30518324282880.0,
      "grad_norm": 1.8592493412049176,
      "language_loss": 0.69803941,
      "learning_rate": 5.971884151083695e-08,
      "loss": 0.72157568,
      "num_input_tokens_seen": 166201855,
      "step": 7687,
      "time_per_iteration": 2.7674665451049805
    },
    {
      "auxiliary_loss_clip": 0.0127535,
      "auxiliary_loss_mlp": 0.01023945,
      "balance_loss_clip": 1.04695463,
      "balance_loss_mlp": 1.01702785,
      "epoch": 0.9244273432333313,
      "flos": 28658331244800.0,
      "grad_norm": 1.8626296885726128,
      "language_loss": 0.74090058,
      "learning_rate": 5.9530054324595124e-08,
      "loss": 0.76389354,
      "num_input_tokens_seen": 166221970,
      "step": 7688,
      "time_per_iteration": 2.788607358932495
    },
    {
      "auxiliary_loss_clip": 0.01107862,
      "auxiliary_loss_mlp": 0.02504532,
      "balance_loss_clip": 1.00681317,
      "balance_loss_mlp": 0.99981117,
      "epoch": 0.9245475861239704,
      "flos": 66230589237120.0,
      "grad_norm": 0.7184408929410301,
      "language_loss": 0.57520521,
      "learning_rate": 5.934156150610103e-08,
      "loss": 0.61132908,
      "num_input_tokens_seen": 166279335,
      "step": 7689,
      "time_per_iteration": 3.239762544631958
    },
    {
      "auxiliary_loss_clip": 0.01271996,
      "auxiliary_loss_mlp": 0.01031694,
      "balance_loss_clip": 1.04581785,
      "balance_loss_mlp": 1.02492893,
      "epoch": 0.9246678290146095,
      "flos": 24239918142720.0,
      "grad_norm": 2.55743660460017,
      "language_loss": 0.79188991,
      "learning_rate": 5.915336308394914e-08,
      "loss": 0.81492686,
      "num_input_tokens_seen": 166298170,
      "step": 7690,
      "time_per_iteration": 2.6788175106048584
    },
    {
      "auxiliary_loss_clip": 0.01215798,
      "auxiliary_loss_mlp": 0.01023749,
      "balance_loss_clip": 1.04647422,
      "balance_loss_mlp": 1.01692438,
      "epoch": 0.9247880719052486,
      "flos": 18988808976000.0,
      "grad_norm": 1.796220889811145,
      "language_loss": 0.7702952,
      "learning_rate": 5.89654590866886e-08,
      "loss": 0.79269063,
      "num_input_tokens_seen": 166317670,
      "step": 7691,
      "time_per_iteration": 2.6483120918273926
    },
    {
      "auxiliary_loss_clip": 0.01427152,
      "auxiliary_loss_mlp": 0.010236,
      "balance_loss_clip": 1.04632497,
      "balance_loss_mlp": 1.01610219,
      "epoch": 0.9249083147958876,
      "flos": 24024095274240.0,
      "grad_norm": 2.161649314901735,
      "language_loss": 0.88450336,
      "learning_rate": 5.877784954282483e-08,
      "loss": 0.90901089,
      "num_input_tokens_seen": 166337010,
      "step": 7692,
      "time_per_iteration": 2.784393072128296
    },
    {
      "auxiliary_loss_clip": 0.01227543,
      "auxiliary_loss_mlp": 0.01027757,
      "balance_loss_clip": 1.04848897,
      "balance_loss_mlp": 1.02061307,
      "epoch": 0.9250285576865268,
      "flos": 30773972355840.0,
      "grad_norm": 2.7499012123479307,
      "language_loss": 0.72801268,
      "learning_rate": 5.8590534480817963e-08,
      "loss": 0.75056565,
      "num_input_tokens_seen": 166358735,
      "step": 7693,
      "time_per_iteration": 2.9204342365264893
    },
    {
      "auxiliary_loss_clip": 0.0117501,
      "auxiliary_loss_mlp": 0.01025117,
      "balance_loss_clip": 1.05112696,
      "balance_loss_mlp": 1.01834917,
      "epoch": 0.9251488005771659,
      "flos": 10633581348480.0,
      "grad_norm": 2.300030991527989,
      "language_loss": 0.72775251,
      "learning_rate": 5.840351392908349e-08,
      "loss": 0.74975377,
      "num_input_tokens_seen": 166374455,
      "step": 7694,
      "time_per_iteration": 2.528831958770752
    },
    {
      "auxiliary_loss_clip": 0.01279563,
      "auxiliary_loss_mlp": 0.02565513,
      "balance_loss_clip": 1.04625916,
      "balance_loss_mlp": 0.99990177,
      "epoch": 0.9252690434678049,
      "flos": 23586416052480.0,
      "grad_norm": 3.443298946856316,
      "language_loss": 0.71001655,
      "learning_rate": 5.821678791599205e-08,
      "loss": 0.74846733,
      "num_input_tokens_seen": 166393900,
      "step": 7695,
      "time_per_iteration": 2.7325503826141357
    },
    {
      "auxiliary_loss_clip": 0.01271301,
      "auxiliary_loss_mlp": 0.01026751,
      "balance_loss_clip": 1.0476346,
      "balance_loss_mlp": 1.0200243,
      "epoch": 0.9253892863584441,
      "flos": 21469158829440.0,
      "grad_norm": 1.8177289632942266,
      "language_loss": 0.80650729,
      "learning_rate": 5.803035646986965e-08,
      "loss": 0.8294878,
      "num_input_tokens_seen": 166413235,
      "step": 7696,
      "time_per_iteration": 2.675334930419922
    },
    {
      "auxiliary_loss_clip": 0.01175136,
      "auxiliary_loss_mlp": 0.01028389,
      "balance_loss_clip": 1.04827237,
      "balance_loss_mlp": 1.02123094,
      "epoch": 0.9255095292490831,
      "flos": 17456680304640.0,
      "grad_norm": 2.785506813784668,
      "language_loss": 0.67396724,
      "learning_rate": 5.7844219618998766e-08,
      "loss": 0.69600248,
      "num_input_tokens_seen": 166427560,
      "step": 7697,
      "time_per_iteration": 3.4817934036254883
    },
    {
      "auxiliary_loss_clip": 0.01313765,
      "auxiliary_loss_mlp": 0.0102423,
      "balance_loss_clip": 1.04071307,
      "balance_loss_mlp": 1.01760817,
      "epoch": 0.9256297721397222,
      "flos": 24750675584640.0,
      "grad_norm": 1.7721496804453651,
      "language_loss": 0.71958768,
      "learning_rate": 5.765837739161505e-08,
      "loss": 0.74296767,
      "num_input_tokens_seen": 166446680,
      "step": 7698,
      "time_per_iteration": 3.6690096855163574
    },
    {
      "auxiliary_loss_clip": 0.01322025,
      "auxiliary_loss_mlp": 0.01025278,
      "balance_loss_clip": 1.04599643,
      "balance_loss_mlp": 1.01873946,
      "epoch": 0.9257500150303614,
      "flos": 23112215677440.0,
      "grad_norm": 1.666850578159443,
      "language_loss": 0.74326801,
      "learning_rate": 5.7472829815911504e-08,
      "loss": 0.76674104,
      "num_input_tokens_seen": 166465505,
      "step": 7699,
      "time_per_iteration": 2.691148042678833
    },
    {
      "auxiliary_loss_clip": 0.01270068,
      "auxiliary_loss_mlp": 0.01021599,
      "balance_loss_clip": 1.04485393,
      "balance_loss_mlp": 1.01441991,
      "epoch": 0.9258702579210004,
      "flos": 22564685687040.0,
      "grad_norm": 1.8451428570727484,
      "language_loss": 0.82048154,
      "learning_rate": 5.7287576920035164e-08,
      "loss": 0.84339821,
      "num_input_tokens_seen": 166484520,
      "step": 7700,
      "time_per_iteration": 2.690112829208374
    },
    {
      "auxiliary_loss_clip": 0.01313652,
      "auxiliary_loss_mlp": 0.01023651,
      "balance_loss_clip": 1.04554152,
      "balance_loss_mlp": 1.01728487,
      "epoch": 0.9259905008116395,
      "flos": 30004298703360.0,
      "grad_norm": 1.7864136972950835,
      "language_loss": 0.76322836,
      "learning_rate": 5.7102618732088435e-08,
      "loss": 0.78660136,
      "num_input_tokens_seen": 166503850,
      "step": 7701,
      "time_per_iteration": 3.7211196422576904
    },
    {
      "auxiliary_loss_clip": 0.01177097,
      "auxiliary_loss_mlp": 0.01029833,
      "balance_loss_clip": 1.04733562,
      "balance_loss_mlp": 1.02318716,
      "epoch": 0.9261107437022786,
      "flos": 24572128055040.0,
      "grad_norm": 1.7089122409412214,
      "language_loss": 0.74632955,
      "learning_rate": 5.6917955280130216e-08,
      "loss": 0.76839888,
      "num_input_tokens_seen": 166525330,
      "step": 7702,
      "time_per_iteration": 2.6704318523406982
    },
    {
      "auxiliary_loss_clip": 0.01220168,
      "auxiliary_loss_mlp": 0.01024197,
      "balance_loss_clip": 1.04800177,
      "balance_loss_mlp": 1.01700854,
      "epoch": 0.9262309865929177,
      "flos": 22018448586240.0,
      "grad_norm": 2.3304280556542722,
      "language_loss": 0.72060859,
      "learning_rate": 5.6733586592172755e-08,
      "loss": 0.74305224,
      "num_input_tokens_seen": 166544825,
      "step": 7703,
      "time_per_iteration": 2.6581056118011475
    },
    {
      "auxiliary_loss_clip": 0.01264592,
      "auxiliary_loss_mlp": 0.0256079,
      "balance_loss_clip": 1.04283381,
      "balance_loss_mlp": 0.99991059,
      "epoch": 0.9263512294835567,
      "flos": 20339481116160.0,
      "grad_norm": 1.753744912981716,
      "language_loss": 0.80246782,
      "learning_rate": 5.6549512696185244e-08,
      "loss": 0.84072173,
      "num_input_tokens_seen": 166563325,
      "step": 7704,
      "time_per_iteration": 2.650956630706787
    },
    {
      "auxiliary_loss_clip": 0.01171871,
      "auxiliary_loss_mlp": 0.01026943,
      "balance_loss_clip": 1.05033946,
      "balance_loss_mlp": 1.02003217,
      "epoch": 0.9264714723741959,
      "flos": 21215378263680.0,
      "grad_norm": 1.625601258765649,
      "language_loss": 0.68217891,
      "learning_rate": 5.636573362009156e-08,
      "loss": 0.70416701,
      "num_input_tokens_seen": 166583385,
      "step": 7705,
      "time_per_iteration": 3.528310775756836
    },
    {
      "auxiliary_loss_clip": 0.01176181,
      "auxiliary_loss_mlp": 0.01028738,
      "balance_loss_clip": 1.04870892,
      "balance_loss_mlp": 1.02084672,
      "epoch": 0.926591715264835,
      "flos": 18004964480640.0,
      "grad_norm": 2.8554600777243797,
      "language_loss": 0.76712757,
      "learning_rate": 5.618224939177074e-08,
      "loss": 0.78917676,
      "num_input_tokens_seen": 166601290,
      "step": 7706,
      "time_per_iteration": 2.587123155593872
    },
    {
      "auxiliary_loss_clip": 0.01266142,
      "auxiliary_loss_mlp": 0.01021581,
      "balance_loss_clip": 1.04516387,
      "balance_loss_mlp": 1.01506579,
      "epoch": 0.926711958155474,
      "flos": 36167969825280.0,
      "grad_norm": 1.8130260201248998,
      "language_loss": 0.70227039,
      "learning_rate": 5.599906003905719e-08,
      "loss": 0.7251476,
      "num_input_tokens_seen": 166623835,
      "step": 7707,
      "time_per_iteration": 2.745382070541382
    },
    {
      "auxiliary_loss_clip": 0.01225014,
      "auxiliary_loss_mlp": 0.01028796,
      "balance_loss_clip": 1.05202866,
      "balance_loss_mlp": 1.02149189,
      "epoch": 0.9268322010461132,
      "flos": 21032736583680.0,
      "grad_norm": 2.6685322764308417,
      "language_loss": 0.81577253,
      "learning_rate": 5.581616558974023e-08,
      "loss": 0.8383106,
      "num_input_tokens_seen": 166642400,
      "step": 7708,
      "time_per_iteration": 2.6579818725585938
    },
    {
      "auxiliary_loss_clip": 0.01229982,
      "auxiliary_loss_mlp": 0.02567561,
      "balance_loss_clip": 1.04852152,
      "balance_loss_mlp": 0.99989283,
      "epoch": 0.9269524439367522,
      "flos": 22964838174720.0,
      "grad_norm": 1.7514163569503698,
      "language_loss": 0.79663223,
      "learning_rate": 5.5633566071565444e-08,
      "loss": 0.8346076,
      "num_input_tokens_seen": 166661640,
      "step": 7709,
      "time_per_iteration": 2.6408329010009766
    },
    {
      "auxiliary_loss_clip": 0.01423549,
      "auxiliary_loss_mlp": 0.01024015,
      "balance_loss_clip": 1.04381824,
      "balance_loss_mlp": 1.01750636,
      "epoch": 0.9270726868273913,
      "flos": 41975551468800.0,
      "grad_norm": 2.508073046011233,
      "language_loss": 0.70424289,
      "learning_rate": 5.5451261512232896e-08,
      "loss": 0.72871852,
      "num_input_tokens_seen": 166684320,
      "step": 7710,
      "time_per_iteration": 3.0209579467773438
    },
    {
      "auxiliary_loss_clip": 0.01224929,
      "auxiliary_loss_mlp": 0.01026881,
      "balance_loss_clip": 1.04506755,
      "balance_loss_mlp": 1.02006269,
      "epoch": 0.9271929297180305,
      "flos": 19791771557760.0,
      "grad_norm": 3.5500058418484226,
      "language_loss": 0.62538803,
      "learning_rate": 5.5269251939397576e-08,
      "loss": 0.64790612,
      "num_input_tokens_seen": 166703835,
      "step": 7711,
      "time_per_iteration": 3.066128730773926
    },
    {
      "auxiliary_loss_clip": 0.01321686,
      "auxiliary_loss_mlp": 0.01023266,
      "balance_loss_clip": 1.04141986,
      "balance_loss_mlp": 1.01628351,
      "epoch": 0.9273131726086695,
      "flos": 19968343839360.0,
      "grad_norm": 2.0488016804574882,
      "language_loss": 0.7671141,
      "learning_rate": 5.508753738067073e-08,
      "loss": 0.79056358,
      "num_input_tokens_seen": 166723375,
      "step": 7712,
      "time_per_iteration": 2.778892755508423
    },
    {
      "auxiliary_loss_clip": 0.01225083,
      "auxiliary_loss_mlp": 0.01019247,
      "balance_loss_clip": 1.04625845,
      "balance_loss_mlp": 1.01238036,
      "epoch": 0.9274334154993086,
      "flos": 23258587599360.0,
      "grad_norm": 1.9453749902112667,
      "language_loss": 0.79160124,
      "learning_rate": 5.4906117863617875e-08,
      "loss": 0.81404448,
      "num_input_tokens_seen": 166742760,
      "step": 7713,
      "time_per_iteration": 2.7032482624053955
    },
    {
      "auxiliary_loss_clip": 0.01317228,
      "auxiliary_loss_mlp": 0.01024759,
      "balance_loss_clip": 1.04257774,
      "balance_loss_mlp": 1.0183903,
      "epoch": 0.9275536583899477,
      "flos": 31795343585280.0,
      "grad_norm": 1.7803058715346476,
      "language_loss": 0.78010195,
      "learning_rate": 5.4724993415760533e-08,
      "loss": 0.80352187,
      "num_input_tokens_seen": 166761115,
      "step": 7714,
      "time_per_iteration": 2.7863426208496094
    },
    {
      "auxiliary_loss_clip": 0.01329858,
      "auxiliary_loss_mlp": 0.02565364,
      "balance_loss_clip": 1.04357886,
      "balance_loss_mlp": 0.99991763,
      "epoch": 0.9276739012805868,
      "flos": 18696998885760.0,
      "grad_norm": 2.1805709756597795,
      "language_loss": 0.74868041,
      "learning_rate": 5.454416406457496e-08,
      "loss": 0.78763264,
      "num_input_tokens_seen": 166780210,
      "step": 7715,
      "time_per_iteration": 2.702840566635132
    },
    {
      "auxiliary_loss_clip": 0.01221701,
      "auxiliary_loss_mlp": 0.01023098,
      "balance_loss_clip": 1.0466491,
      "balance_loss_mlp": 1.01637113,
      "epoch": 0.9277941441712259,
      "flos": 13879079740800.0,
      "grad_norm": 2.5938394828002616,
      "language_loss": 0.7428754,
      "learning_rate": 5.436362983749299e-08,
      "loss": 0.76532334,
      "num_input_tokens_seen": 166795380,
      "step": 7716,
      "time_per_iteration": 2.575099229812622
    },
    {
      "auxiliary_loss_clip": 0.01313794,
      "auxiliary_loss_mlp": 0.01018379,
      "balance_loss_clip": 1.0479629,
      "balance_loss_mlp": 1.01189685,
      "epoch": 0.927914387061865,
      "flos": 23258659426560.0,
      "grad_norm": 2.1566773996243525,
      "language_loss": 0.64545077,
      "learning_rate": 5.418339076190137e-08,
      "loss": 0.66877252,
      "num_input_tokens_seen": 166814890,
      "step": 7717,
      "time_per_iteration": 2.7410027980804443
    },
    {
      "auxiliary_loss_clip": 0.01268234,
      "auxiliary_loss_mlp": 0.01028123,
      "balance_loss_clip": 1.0455997,
      "balance_loss_mlp": 1.02144992,
      "epoch": 0.9280346299525041,
      "flos": 18073733068800.0,
      "grad_norm": 2.12654806262134,
      "language_loss": 0.88661242,
      "learning_rate": 5.400344686514202e-08,
      "loss": 0.90957606,
      "num_input_tokens_seen": 166832475,
      "step": 7718,
      "time_per_iteration": 2.678560495376587
    },
    {
      "auxiliary_loss_clip": 0.01219991,
      "auxiliary_loss_mlp": 0.01023248,
      "balance_loss_clip": 1.0477854,
      "balance_loss_mlp": 1.01670039,
      "epoch": 0.9281548728431431,
      "flos": 22342901160960.0,
      "grad_norm": 1.8221807449596406,
      "language_loss": 0.66893148,
      "learning_rate": 5.38237981745131e-08,
      "loss": 0.69136393,
      "num_input_tokens_seen": 166850590,
      "step": 7719,
      "time_per_iteration": 2.6776480674743652
    },
    {
      "auxiliary_loss_clip": 0.01224703,
      "auxiliary_loss_mlp": 0.02564967,
      "balance_loss_clip": 1.04648209,
      "balance_loss_mlp": 0.99989748,
      "epoch": 0.9282751157337822,
      "flos": 18843765857280.0,
      "grad_norm": 2.0051933085365383,
      "language_loss": 0.81323123,
      "learning_rate": 5.364444471726592e-08,
      "loss": 0.85112786,
      "num_input_tokens_seen": 166869795,
      "step": 7720,
      "time_per_iteration": 2.618046760559082
    },
    {
      "auxiliary_loss_clip": 0.01221729,
      "auxiliary_loss_mlp": 0.01026867,
      "balance_loss_clip": 1.04712105,
      "balance_loss_mlp": 1.01993823,
      "epoch": 0.9283953586244214,
      "flos": 25556834476800.0,
      "grad_norm": 3.4683001619516443,
      "language_loss": 0.80351269,
      "learning_rate": 5.346538652060939e-08,
      "loss": 0.8259986,
      "num_input_tokens_seen": 166891150,
      "step": 7721,
      "time_per_iteration": 2.670834541320801
    },
    {
      "auxiliary_loss_clip": 0.01269872,
      "auxiliary_loss_mlp": 0.01025931,
      "balance_loss_clip": 1.04778016,
      "balance_loss_mlp": 1.01957107,
      "epoch": 0.9285156015150604,
      "flos": 18223480869120.0,
      "grad_norm": 2.0176861073747885,
      "language_loss": 0.70098865,
      "learning_rate": 5.3286623611705994e-08,
      "loss": 0.72394663,
      "num_input_tokens_seen": 166909195,
      "step": 7722,
      "time_per_iteration": 2.6197986602783203
    },
    {
      "auxiliary_loss_clip": 0.01059431,
      "auxiliary_loss_mlp": 0.01002249,
      "balance_loss_clip": 1.00563264,
      "balance_loss_mlp": 1.00140905,
      "epoch": 0.9286358444056995,
      "flos": 66400017690240.0,
      "grad_norm": 0.8203169183894441,
      "language_loss": 0.60614705,
      "learning_rate": 5.3108156017673824e-08,
      "loss": 0.62676388,
      "num_input_tokens_seen": 166970955,
      "step": 7723,
      "time_per_iteration": 4.4203643798828125
    },
    {
      "auxiliary_loss_clip": 0.01177611,
      "auxiliary_loss_mlp": 0.01029189,
      "balance_loss_clip": 1.04798973,
      "balance_loss_mlp": 1.02184546,
      "epoch": 0.9287560872963386,
      "flos": 22345630594560.0,
      "grad_norm": 1.7878314897411958,
      "language_loss": 0.71651495,
      "learning_rate": 5.2929983765586775e-08,
      "loss": 0.73858297,
      "num_input_tokens_seen": 166989735,
      "step": 7724,
      "time_per_iteration": 3.6032068729400635
    },
    {
      "auxiliary_loss_clip": 0.01171192,
      "auxiliary_loss_mlp": 0.01022495,
      "balance_loss_clip": 1.04838634,
      "balance_loss_mlp": 1.01603115,
      "epoch": 0.9288763301869777,
      "flos": 25700225569920.0,
      "grad_norm": 1.7861405759556255,
      "language_loss": 0.6268549,
      "learning_rate": 5.275210688247278e-08,
      "loss": 0.64879179,
      "num_input_tokens_seen": 167010060,
      "step": 7725,
      "time_per_iteration": 2.586944341659546
    },
    {
      "auxiliary_loss_clip": 0.01374866,
      "auxiliary_loss_mlp": 0.01024251,
      "balance_loss_clip": 1.04728913,
      "balance_loss_mlp": 1.01756907,
      "epoch": 0.9289965730776167,
      "flos": 12312046028160.0,
      "grad_norm": 1.9533391621767997,
      "language_loss": 0.85105193,
      "learning_rate": 5.257452539531604e-08,
      "loss": 0.87504315,
      "num_input_tokens_seen": 167027130,
      "step": 7726,
      "time_per_iteration": 2.7508585453033447
    },
    {
      "auxiliary_loss_clip": 0.01222091,
      "auxiliary_loss_mlp": 0.01029814,
      "balance_loss_clip": 1.04517078,
      "balance_loss_mlp": 1.02309024,
      "epoch": 0.9291168159682559,
      "flos": 26685973486080.0,
      "grad_norm": 1.7414692614400629,
      "language_loss": 0.68328297,
      "learning_rate": 5.2397239331055445e-08,
      "loss": 0.70580202,
      "num_input_tokens_seen": 167049130,
      "step": 7727,
      "time_per_iteration": 3.636845588684082
    },
    {
      "auxiliary_loss_clip": 0.01264761,
      "auxiliary_loss_mlp": 0.01024482,
      "balance_loss_clip": 1.04727006,
      "balance_loss_mlp": 1.01733851,
      "epoch": 0.929237058858895,
      "flos": 14538256179840.0,
      "grad_norm": 2.1403804069176946,
      "language_loss": 0.80863726,
      "learning_rate": 5.2220248716585036e-08,
      "loss": 0.83152974,
      "num_input_tokens_seen": 167066810,
      "step": 7728,
      "time_per_iteration": 2.662332057952881
    },
    {
      "auxiliary_loss_clip": 0.01217327,
      "auxiliary_loss_mlp": 0.01025512,
      "balance_loss_clip": 1.04424238,
      "balance_loss_mlp": 1.01892579,
      "epoch": 0.929357301749534,
      "flos": 23835456023040.0,
      "grad_norm": 2.0937526968173863,
      "language_loss": 0.75596941,
      "learning_rate": 5.204355357875445e-08,
      "loss": 0.7783978,
      "num_input_tokens_seen": 167085155,
      "step": 7729,
      "time_per_iteration": 2.675457239151001
    },
    {
      "auxiliary_loss_clip": 0.01273035,
      "auxiliary_loss_mlp": 0.01030506,
      "balance_loss_clip": 1.04443431,
      "balance_loss_mlp": 1.02364004,
      "epoch": 0.9294775446401732,
      "flos": 12969319046400.0,
      "grad_norm": 2.2008735155387003,
      "language_loss": 0.70542848,
      "learning_rate": 5.1867153944367584e-08,
      "loss": 0.72846383,
      "num_input_tokens_seen": 167101545,
      "step": 7730,
      "time_per_iteration": 2.610868215560913
    },
    {
      "auxiliary_loss_clip": 0.01325017,
      "auxiliary_loss_mlp": 0.01025047,
      "balance_loss_clip": 1.04429531,
      "balance_loss_mlp": 1.01812053,
      "epoch": 0.9295977875308122,
      "flos": 26211809024640.0,
      "grad_norm": 1.7334318232814747,
      "language_loss": 0.7362836,
      "learning_rate": 5.16910498401848e-08,
      "loss": 0.75978422,
      "num_input_tokens_seen": 167120995,
      "step": 7731,
      "time_per_iteration": 2.771665334701538
    },
    {
      "auxiliary_loss_clip": 0.01172869,
      "auxiliary_loss_mlp": 0.01021556,
      "balance_loss_clip": 1.05067849,
      "balance_loss_mlp": 1.01549745,
      "epoch": 0.9297180304214513,
      "flos": 16472297105280.0,
      "grad_norm": 2.0434319159245296,
      "language_loss": 0.83635128,
      "learning_rate": 5.151524129292073e-08,
      "loss": 0.85829556,
      "num_input_tokens_seen": 167138890,
      "step": 7732,
      "time_per_iteration": 3.506770133972168
    },
    {
      "auxiliary_loss_clip": 0.01222977,
      "auxiliary_loss_mlp": 0.0102732,
      "balance_loss_clip": 1.04901648,
      "balance_loss_mlp": 1.02022147,
      "epoch": 0.9298382733120905,
      "flos": 24060436859520.0,
      "grad_norm": 2.0872207667755673,
      "language_loss": 0.66803205,
      "learning_rate": 5.1339728329245155e-08,
      "loss": 0.69053507,
      "num_input_tokens_seen": 167159455,
      "step": 7733,
      "time_per_iteration": 2.6235122680664062
    },
    {
      "auxiliary_loss_clip": 0.01177897,
      "auxiliary_loss_mlp": 0.01024902,
      "balance_loss_clip": 1.04924893,
      "balance_loss_mlp": 1.01811862,
      "epoch": 0.9299585162027295,
      "flos": 22127652910080.0,
      "grad_norm": 2.2172447950387744,
      "language_loss": 0.78790748,
      "learning_rate": 5.116451097578367e-08,
      "loss": 0.80993557,
      "num_input_tokens_seen": 167178495,
      "step": 7734,
      "time_per_iteration": 2.6340863704681396
    },
    {
      "auxiliary_loss_clip": 0.01321402,
      "auxiliary_loss_mlp": 0.01024865,
      "balance_loss_clip": 1.04301131,
      "balance_loss_mlp": 1.01831102,
      "epoch": 0.9300787590933686,
      "flos": 21471780522240.0,
      "grad_norm": 2.347249812006517,
      "language_loss": 0.74301648,
      "learning_rate": 5.0989589259115895e-08,
      "loss": 0.76647919,
      "num_input_tokens_seen": 167199380,
      "step": 7735,
      "time_per_iteration": 2.687514066696167
    },
    {
      "auxiliary_loss_clip": 0.01220767,
      "auxiliary_loss_mlp": 0.01032752,
      "balance_loss_clip": 1.04486895,
      "balance_loss_mlp": 1.02471697,
      "epoch": 0.9301990019840077,
      "flos": 17779588594560.0,
      "grad_norm": 2.208767087641274,
      "language_loss": 0.71731472,
      "learning_rate": 5.081496320577816e-08,
      "loss": 0.73984993,
      "num_input_tokens_seen": 167216500,
      "step": 7736,
      "time_per_iteration": 2.640324831008911
    },
    {
      "auxiliary_loss_clip": 0.01181964,
      "auxiliary_loss_mlp": 0.0099898,
      "balance_loss_clip": 1.02005267,
      "balance_loss_mlp": 0.99801409,
      "epoch": 0.9303192448746468,
      "flos": 58896122307840.0,
      "grad_norm": 0.9186404950393257,
      "language_loss": 0.6118896,
      "learning_rate": 5.0640632842260835e-08,
      "loss": 0.63369906,
      "num_input_tokens_seen": 167276760,
      "step": 7737,
      "time_per_iteration": 3.3043155670166016
    },
    {
      "auxiliary_loss_clip": 0.01316622,
      "auxiliary_loss_mlp": 0.02563536,
      "balance_loss_clip": 1.04582858,
      "balance_loss_mlp": 0.99990976,
      "epoch": 0.9304394877652858,
      "flos": 57663522172800.0,
      "grad_norm": 1.572472575035379,
      "language_loss": 0.72719574,
      "learning_rate": 5.0466598195009426e-08,
      "loss": 0.76599729,
      "num_input_tokens_seen": 167303630,
      "step": 7738,
      "time_per_iteration": 3.0005831718444824
    },
    {
      "auxiliary_loss_clip": 0.01220606,
      "auxiliary_loss_mlp": 0.01021906,
      "balance_loss_clip": 1.04490709,
      "balance_loss_mlp": 1.01566827,
      "epoch": 0.930559730655925,
      "flos": 20996143603200.0,
      "grad_norm": 1.9282547649349842,
      "language_loss": 0.70181489,
      "learning_rate": 5.0292859290425036e-08,
      "loss": 0.72424001,
      "num_input_tokens_seen": 167321500,
      "step": 7739,
      "time_per_iteration": 2.6587905883789062
    },
    {
      "auxiliary_loss_clip": 0.01172124,
      "auxiliary_loss_mlp": 0.01025277,
      "balance_loss_clip": 1.04951644,
      "balance_loss_mlp": 1.01901817,
      "epoch": 0.9306799735465641,
      "flos": 23258264376960.0,
      "grad_norm": 1.9860508867025712,
      "language_loss": 0.77873456,
      "learning_rate": 5.011941615486348e-08,
      "loss": 0.80070853,
      "num_input_tokens_seen": 167340615,
      "step": 7740,
      "time_per_iteration": 2.6158289909362793
    },
    {
      "auxiliary_loss_clip": 0.01171406,
      "auxiliary_loss_mlp": 0.01022432,
      "balance_loss_clip": 1.04799473,
      "balance_loss_mlp": 1.01613522,
      "epoch": 0.9308002164372031,
      "flos": 15231547560960.0,
      "grad_norm": 2.038649066424119,
      "language_loss": 0.84596819,
      "learning_rate": 4.994626881463659e-08,
      "loss": 0.86790657,
      "num_input_tokens_seen": 167356870,
      "step": 7741,
      "time_per_iteration": 2.552450656890869
    },
    {
      "auxiliary_loss_clip": 0.01360612,
      "auxiliary_loss_mlp": 0.01023171,
      "balance_loss_clip": 1.04058254,
      "balance_loss_mlp": 1.016361,
      "epoch": 0.9309204593278423,
      "flos": 30847481539200.0,
      "grad_norm": 4.693994822726858,
      "language_loss": 0.71296227,
      "learning_rate": 4.9773417296009814e-08,
      "loss": 0.73680013,
      "num_input_tokens_seen": 167378390,
      "step": 7742,
      "time_per_iteration": 2.858205556869507
    },
    {
      "auxiliary_loss_clip": 0.01228929,
      "auxiliary_loss_mlp": 0.01028108,
      "balance_loss_clip": 1.0490495,
      "balance_loss_mlp": 1.02082455,
      "epoch": 0.9310407022184813,
      "flos": 23037269950080.0,
      "grad_norm": 1.71489651736961,
      "language_loss": 0.65754515,
      "learning_rate": 4.960086162520527e-08,
      "loss": 0.68011552,
      "num_input_tokens_seen": 167398480,
      "step": 7743,
      "time_per_iteration": 2.693136215209961
    },
    {
      "auxiliary_loss_clip": 0.01274115,
      "auxiliary_loss_mlp": 0.01025491,
      "balance_loss_clip": 1.04295015,
      "balance_loss_mlp": 1.01898229,
      "epoch": 0.9311609451091204,
      "flos": 22127976132480.0,
      "grad_norm": 1.9201485035042742,
      "language_loss": 0.82350516,
      "learning_rate": 4.942860182839936e-08,
      "loss": 0.84650123,
      "num_input_tokens_seen": 167416825,
      "step": 7744,
      "time_per_iteration": 2.7034459114074707
    },
    {
      "auxiliary_loss_clip": 0.01271115,
      "auxiliary_loss_mlp": 0.01026163,
      "balance_loss_clip": 1.04569507,
      "balance_loss_mlp": 1.01926708,
      "epoch": 0.9312811879997596,
      "flos": 21099206701440.0,
      "grad_norm": 2.0036338772841704,
      "language_loss": 0.79605103,
      "learning_rate": 4.925663793172341e-08,
      "loss": 0.81902385,
      "num_input_tokens_seen": 167434785,
      "step": 7745,
      "time_per_iteration": 2.6840360164642334
    },
    {
      "auxiliary_loss_clip": 0.01161725,
      "auxiliary_loss_mlp": 0.02504905,
      "balance_loss_clip": 1.00712192,
      "balance_loss_mlp": 0.99985296,
      "epoch": 0.9314014308903986,
      "flos": 67148179096320.0,
      "grad_norm": 0.7842745074283981,
      "language_loss": 0.5652535,
      "learning_rate": 4.908496996126477e-08,
      "loss": 0.60191977,
      "num_input_tokens_seen": 167498245,
      "step": 7746,
      "time_per_iteration": 3.2649669647216797
    },
    {
      "auxiliary_loss_clip": 0.01232872,
      "auxiliary_loss_mlp": 0.01024979,
      "balance_loss_clip": 1.05675828,
      "balance_loss_mlp": 1.01802611,
      "epoch": 0.9315216737810377,
      "flos": 22565583527040.0,
      "grad_norm": 2.340446922808765,
      "language_loss": 0.76372683,
      "learning_rate": 4.89135979430646e-08,
      "loss": 0.78630531,
      "num_input_tokens_seen": 167518290,
      "step": 7747,
      "time_per_iteration": 2.6783008575439453
    },
    {
      "auxiliary_loss_clip": 0.01174334,
      "auxiliary_loss_mlp": 0.0103069,
      "balance_loss_clip": 1.05075479,
      "balance_loss_mlp": 1.02345133,
      "epoch": 0.9316419166716768,
      "flos": 23984054588160.0,
      "grad_norm": 4.125410637833837,
      "language_loss": 0.85510969,
      "learning_rate": 4.874252190312078e-08,
      "loss": 0.87715995,
      "num_input_tokens_seen": 167538675,
      "step": 7748,
      "time_per_iteration": 3.5444211959838867
    },
    {
      "auxiliary_loss_clip": 0.0112378,
      "auxiliary_loss_mlp": 0.01028255,
      "balance_loss_clip": 1.0460248,
      "balance_loss_mlp": 1.02099264,
      "epoch": 0.9317621595623159,
      "flos": 30230464688640.0,
      "grad_norm": 1.9517312534852662,
      "language_loss": 0.65004373,
      "learning_rate": 4.857174186738477e-08,
      "loss": 0.6715641,
      "num_input_tokens_seen": 167562025,
      "step": 7749,
      "time_per_iteration": 2.729975700378418
    },
    {
      "auxiliary_loss_clip": 0.01175479,
      "auxiliary_loss_mlp": 0.01020914,
      "balance_loss_clip": 1.05149364,
      "balance_loss_mlp": 1.01403892,
      "epoch": 0.931882402452955,
      "flos": 15742735966080.0,
      "grad_norm": 4.700425355625792,
      "language_loss": 0.73036432,
      "learning_rate": 4.840125786176408e-08,
      "loss": 0.75232828,
      "num_input_tokens_seen": 167578230,
      "step": 7750,
      "time_per_iteration": 3.4785571098327637
    },
    {
      "auxiliary_loss_clip": 0.01272199,
      "auxiliary_loss_mlp": 0.01029593,
      "balance_loss_clip": 1.04703259,
      "balance_loss_mlp": 1.02240777,
      "epoch": 0.932002645343594,
      "flos": 28366521154560.0,
      "grad_norm": 1.8225008428527856,
      "language_loss": 0.77340567,
      "learning_rate": 4.823106991212067e-08,
      "loss": 0.79642361,
      "num_input_tokens_seen": 167597470,
      "step": 7751,
      "time_per_iteration": 2.6980156898498535
    },
    {
      "auxiliary_loss_clip": 0.012234,
      "auxiliary_loss_mlp": 0.01021336,
      "balance_loss_clip": 1.04703736,
      "balance_loss_mlp": 1.0146544,
      "epoch": 0.9321228882342332,
      "flos": 15341146934400.0,
      "grad_norm": 2.120786092795345,
      "language_loss": 0.83541811,
      "learning_rate": 4.806117804427212e-08,
      "loss": 0.85786545,
      "num_input_tokens_seen": 167615405,
      "step": 7752,
      "time_per_iteration": 2.644279718399048
    },
    {
      "auxiliary_loss_clip": 0.01220847,
      "auxiliary_loss_mlp": 0.01021093,
      "balance_loss_clip": 1.04563701,
      "balance_loss_mlp": 1.01469469,
      "epoch": 0.9322431311248722,
      "flos": 17895365107200.0,
      "grad_norm": 1.903726425000934,
      "language_loss": 0.64602953,
      "learning_rate": 4.7891582283990926e-08,
      "loss": 0.66844893,
      "num_input_tokens_seen": 167634130,
      "step": 7753,
      "time_per_iteration": 3.539407968521118
    },
    {
      "auxiliary_loss_clip": 0.01218693,
      "auxiliary_loss_mlp": 0.01027567,
      "balance_loss_clip": 1.04266763,
      "balance_loss_mlp": 1.02107942,
      "epoch": 0.9323633740155113,
      "flos": 24169713010560.0,
      "grad_norm": 1.6476462940931886,
      "language_loss": 0.72739983,
      "learning_rate": 4.772228265700473e-08,
      "loss": 0.74986249,
      "num_input_tokens_seen": 167654990,
      "step": 7754,
      "time_per_iteration": 2.7077977657318115
    },
    {
      "auxiliary_loss_clip": 0.01227755,
      "auxiliary_loss_mlp": 0.01021379,
      "balance_loss_clip": 1.04868281,
      "balance_loss_mlp": 1.01457834,
      "epoch": 0.9324836169061504,
      "flos": 15043482927360.0,
      "grad_norm": 2.317424078870103,
      "language_loss": 0.75901079,
      "learning_rate": 4.75532791889961e-08,
      "loss": 0.78150213,
      "num_input_tokens_seen": 167671690,
      "step": 7755,
      "time_per_iteration": 2.591948986053467
    },
    {
      "auxiliary_loss_clip": 0.01221653,
      "auxiliary_loss_mlp": 0.01024032,
      "balance_loss_clip": 1.04488659,
      "balance_loss_mlp": 1.0169661,
      "epoch": 0.9326038597967895,
      "flos": 18624890332800.0,
      "grad_norm": 1.8911021387323868,
      "language_loss": 0.65782601,
      "learning_rate": 4.738457190560252e-08,
      "loss": 0.68028289,
      "num_input_tokens_seen": 167690800,
      "step": 7756,
      "time_per_iteration": 2.7287280559539795
    },
    {
      "auxiliary_loss_clip": 0.01367111,
      "auxiliary_loss_mlp": 0.01023799,
      "balance_loss_clip": 1.0470264,
      "balance_loss_mlp": 1.01704264,
      "epoch": 0.9327241026874286,
      "flos": 18952646958720.0,
      "grad_norm": 2.362074826983453,
      "language_loss": 0.79086751,
      "learning_rate": 4.721616083241664e-08,
      "loss": 0.8147766,
      "num_input_tokens_seen": 167709055,
      "step": 7757,
      "time_per_iteration": 3.462134838104248
    },
    {
      "auxiliary_loss_clip": 0.0122002,
      "auxiliary_loss_mlp": 0.01027038,
      "balance_loss_clip": 1.04602396,
      "balance_loss_mlp": 1.02006745,
      "epoch": 0.9328443455780677,
      "flos": 29570282668800.0,
      "grad_norm": 3.0052127375951327,
      "language_loss": 0.77693874,
      "learning_rate": 4.7048045994986684e-08,
      "loss": 0.79940933,
      "num_input_tokens_seen": 167729915,
      "step": 7758,
      "time_per_iteration": 2.593822956085205
    },
    {
      "auxiliary_loss_clip": 0.0112722,
      "auxiliary_loss_mlp": 0.01024469,
      "balance_loss_clip": 1.04851532,
      "balance_loss_mlp": 1.01761734,
      "epoch": 0.9329645884687068,
      "flos": 30081722469120.0,
      "grad_norm": 2.0386099466548244,
      "language_loss": 0.91214001,
      "learning_rate": 4.688022741881559e-08,
      "loss": 0.93365693,
      "num_input_tokens_seen": 167750440,
      "step": 7759,
      "time_per_iteration": 2.599095582962036
    },
    {
      "auxiliary_loss_clip": 0.01218986,
      "auxiliary_loss_mlp": 0.01022354,
      "balance_loss_clip": 1.0469023,
      "balance_loss_mlp": 1.01578259,
      "epoch": 0.9330848313593458,
      "flos": 21867982513920.0,
      "grad_norm": 3.355303747181579,
      "language_loss": 0.7510674,
      "learning_rate": 4.671270512936076e-08,
      "loss": 0.77348077,
      "num_input_tokens_seen": 167769600,
      "step": 7760,
      "time_per_iteration": 2.4895782470703125
    },
    {
      "auxiliary_loss_clip": 0.01317739,
      "auxiliary_loss_mlp": 0.01021269,
      "balance_loss_clip": 1.04397714,
      "balance_loss_mlp": 1.01508474,
      "epoch": 0.933205074249985,
      "flos": 22127221946880.0,
      "grad_norm": 1.8623749035013923,
      "language_loss": 0.82931304,
      "learning_rate": 4.6545479152035884e-08,
      "loss": 0.85270309,
      "num_input_tokens_seen": 167788770,
      "step": 7761,
      "time_per_iteration": 2.703263521194458
    },
    {
      "auxiliary_loss_clip": 0.01224012,
      "auxiliary_loss_mlp": 0.01023426,
      "balance_loss_clip": 1.04825139,
      "balance_loss_mlp": 1.01714015,
      "epoch": 0.9333253171406241,
      "flos": 15341254675200.0,
      "grad_norm": 1.956444068680892,
      "language_loss": 0.76125681,
      "learning_rate": 4.637854951220821e-08,
      "loss": 0.78373116,
      "num_input_tokens_seen": 167805555,
      "step": 7762,
      "time_per_iteration": 2.5428106784820557
    },
    {
      "auxiliary_loss_clip": 0.0131546,
      "auxiliary_loss_mlp": 0.01028576,
      "balance_loss_clip": 1.04364395,
      "balance_loss_mlp": 1.02182293,
      "epoch": 0.9334455600312631,
      "flos": 15706142985600.0,
      "grad_norm": 1.9025363427225546,
      "language_loss": 0.74947953,
      "learning_rate": 4.621191623520171e-08,
      "loss": 0.77291989,
      "num_input_tokens_seen": 167823985,
      "step": 7763,
      "time_per_iteration": 2.696183204650879
    },
    {
      "auxiliary_loss_clip": 0.01331765,
      "auxiliary_loss_mlp": 0.01027887,
      "balance_loss_clip": 1.04385591,
      "balance_loss_mlp": 1.02062094,
      "epoch": 0.9335658029219023,
      "flos": 22163563532160.0,
      "grad_norm": 2.3644031001433743,
      "language_loss": 0.85183918,
      "learning_rate": 4.604557934629372e-08,
      "loss": 0.87543571,
      "num_input_tokens_seen": 167843060,
      "step": 7764,
      "time_per_iteration": 2.9611563682556152
    },
    {
      "auxiliary_loss_clip": 0.01265356,
      "auxiliary_loss_mlp": 0.01024153,
      "balance_loss_clip": 1.0466764,
      "balance_loss_mlp": 1.01765001,
      "epoch": 0.9336860458125413,
      "flos": 20266833859200.0,
      "grad_norm": 1.8282747930680068,
      "language_loss": 0.80487716,
      "learning_rate": 4.587953887071805e-08,
      "loss": 0.82777226,
      "num_input_tokens_seen": 167862880,
      "step": 7765,
      "time_per_iteration": 2.7727267742156982
    },
    {
      "auxiliary_loss_clip": 0.01264855,
      "auxiliary_loss_mlp": 0.01027334,
      "balance_loss_clip": 1.04190874,
      "balance_loss_mlp": 1.02044058,
      "epoch": 0.9338062887031804,
      "flos": 20919689504640.0,
      "grad_norm": 9.51625757219278,
      "language_loss": 0.86189389,
      "learning_rate": 4.5713794833662554e-08,
      "loss": 0.88481581,
      "num_input_tokens_seen": 167882095,
      "step": 7766,
      "time_per_iteration": 2.6742160320281982
    },
    {
      "auxiliary_loss_clip": 0.01173065,
      "auxiliary_loss_mlp": 0.01025121,
      "balance_loss_clip": 1.04780936,
      "balance_loss_mlp": 1.01788461,
      "epoch": 0.9339265315938196,
      "flos": 23221635482880.0,
      "grad_norm": 1.7856242478746225,
      "language_loss": 0.63259321,
      "learning_rate": 4.5548347260270236e-08,
      "loss": 0.65457511,
      "num_input_tokens_seen": 167901385,
      "step": 7767,
      "time_per_iteration": 2.5892162322998047
    },
    {
      "auxiliary_loss_clip": 0.01315702,
      "auxiliary_loss_mlp": 0.0102623,
      "balance_loss_clip": 1.04260027,
      "balance_loss_mlp": 1.01901531,
      "epoch": 0.9340467744844586,
      "flos": 22820261932800.0,
      "grad_norm": 2.5837787685161073,
      "language_loss": 0.69617593,
      "learning_rate": 4.538319617564012e-08,
      "loss": 0.71959531,
      "num_input_tokens_seen": 167920405,
      "step": 7768,
      "time_per_iteration": 2.7438905239105225
    },
    {
      "auxiliary_loss_clip": 0.01272277,
      "auxiliary_loss_mlp": 0.01025153,
      "balance_loss_clip": 1.04426765,
      "balance_loss_mlp": 1.01836097,
      "epoch": 0.9341670173750977,
      "flos": 23660428026240.0,
      "grad_norm": 2.1079067109164455,
      "language_loss": 0.74963152,
      "learning_rate": 4.521834160482485e-08,
      "loss": 0.77260584,
      "num_input_tokens_seen": 167939145,
      "step": 7769,
      "time_per_iteration": 2.6656606197357178
    },
    {
      "auxiliary_loss_clip": 0.01225618,
      "auxiliary_loss_mlp": 0.01025201,
      "balance_loss_clip": 1.04831147,
      "balance_loss_mlp": 1.01847172,
      "epoch": 0.9342872602657368,
      "flos": 24824256595200.0,
      "grad_norm": 2.5302708216246845,
      "language_loss": 0.82187676,
      "learning_rate": 4.5053783572832846e-08,
      "loss": 0.84438497,
      "num_input_tokens_seen": 167959325,
      "step": 7770,
      "time_per_iteration": 2.726696491241455
    },
    {
      "auxiliary_loss_clip": 0.01218898,
      "auxiliary_loss_mlp": 0.01024147,
      "balance_loss_clip": 1.0473696,
      "balance_loss_mlp": 1.01740551,
      "epoch": 0.9344075031563759,
      "flos": 25771831332480.0,
      "grad_norm": 1.782350678596645,
      "language_loss": 0.76269865,
      "learning_rate": 4.488952210462771e-08,
      "loss": 0.78512907,
      "num_input_tokens_seen": 167979530,
      "step": 7771,
      "time_per_iteration": 2.6394431591033936
    },
    {
      "auxiliary_loss_clip": 0.01171833,
      "auxiliary_loss_mlp": 0.0102589,
      "balance_loss_clip": 1.04935408,
      "balance_loss_mlp": 1.01884174,
      "epoch": 0.9345277460470149,
      "flos": 25551303782400.0,
      "grad_norm": 2.0924657086736986,
      "language_loss": 0.85650635,
      "learning_rate": 4.4725557225127495e-08,
      "loss": 0.87848365,
      "num_input_tokens_seen": 167997870,
      "step": 7772,
      "time_per_iteration": 2.6760668754577637
    },
    {
      "auxiliary_loss_clip": 0.01221689,
      "auxiliary_loss_mlp": 0.01028934,
      "balance_loss_clip": 1.04760158,
      "balance_loss_mlp": 1.0225563,
      "epoch": 0.9346479889376541,
      "flos": 34313112432000.0,
      "grad_norm": 1.7150973678870693,
      "language_loss": 0.79342103,
      "learning_rate": 4.456188895920565e-08,
      "loss": 0.81592727,
      "num_input_tokens_seen": 168019625,
      "step": 7773,
      "time_per_iteration": 2.69814395904541
    },
    {
      "auxiliary_loss_clip": 0.01173158,
      "auxiliary_loss_mlp": 0.01026089,
      "balance_loss_clip": 1.04935205,
      "balance_loss_mlp": 1.01902556,
      "epoch": 0.9347682318282932,
      "flos": 19093739581440.0,
      "grad_norm": 2.3605814368180456,
      "language_loss": 0.85379821,
      "learning_rate": 4.439851733169031e-08,
      "loss": 0.8757906,
      "num_input_tokens_seen": 168037415,
      "step": 7774,
      "time_per_iteration": 3.4479074478149414
    },
    {
      "auxiliary_loss_clip": 0.01318068,
      "auxiliary_loss_mlp": 0.01021314,
      "balance_loss_clip": 1.04437232,
      "balance_loss_mlp": 1.01476932,
      "epoch": 0.9348884747189322,
      "flos": 26249587153920.0,
      "grad_norm": 2.4016968483513974,
      "language_loss": 0.69679999,
      "learning_rate": 4.4235442367365204e-08,
      "loss": 0.7201938,
      "num_input_tokens_seen": 168057725,
      "step": 7775,
      "time_per_iteration": 3.7420363426208496
    },
    {
      "auxiliary_loss_clip": 0.01271751,
      "auxiliary_loss_mlp": 0.01023846,
      "balance_loss_clip": 1.04252326,
      "balance_loss_mlp": 1.01707458,
      "epoch": 0.9350087176095714,
      "flos": 18333080242560.0,
      "grad_norm": 2.5820126061632265,
      "language_loss": 0.79612839,
      "learning_rate": 4.4072664090968545e-08,
      "loss": 0.81908435,
      "num_input_tokens_seen": 168076110,
      "step": 7776,
      "time_per_iteration": 2.667287826538086
    },
    {
      "auxiliary_loss_clip": 0.01273926,
      "auxiliary_loss_mlp": 0.01023685,
      "balance_loss_clip": 1.04359221,
      "balance_loss_mlp": 1.01661313,
      "epoch": 0.9351289605002104,
      "flos": 19318253541120.0,
      "grad_norm": 1.8720929691893313,
      "language_loss": 0.84638,
      "learning_rate": 4.391018252719347e-08,
      "loss": 0.8693561,
      "num_input_tokens_seen": 168095905,
      "step": 7777,
      "time_per_iteration": 2.6974427700042725
    },
    {
      "auxiliary_loss_clip": 0.01275633,
      "auxiliary_loss_mlp": 0.01022656,
      "balance_loss_clip": 1.04490352,
      "balance_loss_mlp": 1.01567292,
      "epoch": 0.9352492033908495,
      "flos": 18799990156800.0,
      "grad_norm": 1.6876943433302813,
      "language_loss": 0.69254941,
      "learning_rate": 4.374799770068849e-08,
      "loss": 0.7155323,
      "num_input_tokens_seen": 168112580,
      "step": 7778,
      "time_per_iteration": 2.618621826171875
    },
    {
      "auxiliary_loss_clip": 0.01218463,
      "auxiliary_loss_mlp": 0.0102284,
      "balance_loss_clip": 1.04731607,
      "balance_loss_mlp": 1.0155952,
      "epoch": 0.9353694462814887,
      "flos": 29530134241920.0,
      "grad_norm": 2.5105608486700133,
      "language_loss": 0.74953413,
      "learning_rate": 4.358610963605658e-08,
      "loss": 0.77194715,
      "num_input_tokens_seen": 168133030,
      "step": 7779,
      "time_per_iteration": 3.5370163917541504
    },
    {
      "auxiliary_loss_clip": 0.0117651,
      "auxiliary_loss_mlp": 0.0102555,
      "balance_loss_clip": 1.04999924,
      "balance_loss_mlp": 1.01869512,
      "epoch": 0.9354896891721277,
      "flos": 30665450390400.0,
      "grad_norm": 4.187774208764383,
      "language_loss": 0.68856889,
      "learning_rate": 4.342451835785677e-08,
      "loss": 0.71058953,
      "num_input_tokens_seen": 168153940,
      "step": 7780,
      "time_per_iteration": 2.686305046081543
    },
    {
      "auxiliary_loss_clip": 0.01270228,
      "auxiliary_loss_mlp": 0.01027973,
      "balance_loss_clip": 1.04636955,
      "balance_loss_mlp": 1.02177715,
      "epoch": 0.9356099320627668,
      "flos": 19463907191040.0,
      "grad_norm": 3.0600037164696916,
      "language_loss": 0.75352693,
      "learning_rate": 4.3263223890601665e-08,
      "loss": 0.77650893,
      "num_input_tokens_seen": 168172650,
      "step": 7781,
      "time_per_iteration": 2.6157405376434326
    },
    {
      "auxiliary_loss_clip": 0.01219177,
      "auxiliary_loss_mlp": 0.02562922,
      "balance_loss_clip": 1.0495261,
      "balance_loss_mlp": 0.99989647,
      "epoch": 0.9357301749534058,
      "flos": 19098156954240.0,
      "grad_norm": 1.83192166448584,
      "language_loss": 0.7974298,
      "learning_rate": 4.31022262587597e-08,
      "loss": 0.83525085,
      "num_input_tokens_seen": 168191325,
      "step": 7782,
      "time_per_iteration": 2.709728479385376
    },
    {
      "auxiliary_loss_clip": 0.01224674,
      "auxiliary_loss_mlp": 0.01026235,
      "balance_loss_clip": 1.05092525,
      "balance_loss_mlp": 1.01882946,
      "epoch": 0.935850417844045,
      "flos": 23550361776000.0,
      "grad_norm": 1.6746965565345462,
      "language_loss": 0.6603688,
      "learning_rate": 4.2941525486754225e-08,
      "loss": 0.6828779,
      "num_input_tokens_seen": 168211645,
      "step": 7783,
      "time_per_iteration": 3.5759010314941406
    },
    {
      "auxiliary_loss_clip": 0.01316348,
      "auxiliary_loss_mlp": 0.01020384,
      "balance_loss_clip": 1.0461247,
      "balance_loss_mlp": 1.01427436,
      "epoch": 0.935970660734684,
      "flos": 18588333265920.0,
      "grad_norm": 2.250218890259956,
      "language_loss": 0.7964431,
      "learning_rate": 4.278112159896286e-08,
      "loss": 0.81981039,
      "num_input_tokens_seen": 168229485,
      "step": 7784,
      "time_per_iteration": 2.674241542816162
    },
    {
      "auxiliary_loss_clip": 0.01265212,
      "auxiliary_loss_mlp": 0.01022713,
      "balance_loss_clip": 1.04024446,
      "balance_loss_mlp": 1.01632059,
      "epoch": 0.9360909036253231,
      "flos": 20631255292800.0,
      "grad_norm": 1.6973182146331591,
      "language_loss": 0.67477763,
      "learning_rate": 4.2621014619719896e-08,
      "loss": 0.69765687,
      "num_input_tokens_seen": 168247250,
      "step": 7785,
      "time_per_iteration": 2.725257635116577
    },
    {
      "auxiliary_loss_clip": 0.01168919,
      "auxiliary_loss_mlp": 0.00999843,
      "balance_loss_clip": 1.00548077,
      "balance_loss_mlp": 0.99890101,
      "epoch": 0.9362111465159623,
      "flos": 61791421052160.0,
      "grad_norm": 0.720324846069768,
      "language_loss": 0.58613384,
      "learning_rate": 4.246120457331215e-08,
      "loss": 0.60782146,
      "num_input_tokens_seen": 168309425,
      "step": 7786,
      "time_per_iteration": 3.257882833480835
    },
    {
      "auxiliary_loss_clip": 0.01270294,
      "auxiliary_loss_mlp": 0.0102828,
      "balance_loss_clip": 1.04698217,
      "balance_loss_mlp": 1.02130651,
      "epoch": 0.9363313894066013,
      "flos": 24170395368960.0,
      "grad_norm": 1.9288290804178072,
      "language_loss": 0.72089636,
      "learning_rate": 4.2301691483983325e-08,
      "loss": 0.743882,
      "num_input_tokens_seen": 168329545,
      "step": 7787,
      "time_per_iteration": 2.695051670074463
    },
    {
      "auxiliary_loss_clip": 0.01226197,
      "auxiliary_loss_mlp": 0.01030472,
      "balance_loss_clip": 1.04839015,
      "balance_loss_mlp": 1.02353668,
      "epoch": 0.9364516322972404,
      "flos": 20120354196480.0,
      "grad_norm": 1.9632626053935698,
      "language_loss": 0.76130056,
      "learning_rate": 4.214247537593163e-08,
      "loss": 0.78386724,
      "num_input_tokens_seen": 168348795,
      "step": 7788,
      "time_per_iteration": 2.6557021141052246
    },
    {
      "auxiliary_loss_clip": 0.01272414,
      "auxiliary_loss_mlp": 0.01026128,
      "balance_loss_clip": 1.04378617,
      "balance_loss_mlp": 1.01988447,
      "epoch": 0.9365718751878795,
      "flos": 20703758895360.0,
      "grad_norm": 1.9265344570949134,
      "language_loss": 0.80387914,
      "learning_rate": 4.1983556273309293e-08,
      "loss": 0.82686454,
      "num_input_tokens_seen": 168367545,
      "step": 7789,
      "time_per_iteration": 2.6794092655181885
    },
    {
      "auxiliary_loss_clip": 0.01175808,
      "auxiliary_loss_mlp": 0.0102742,
      "balance_loss_clip": 1.0497725,
      "balance_loss_mlp": 1.02011228,
      "epoch": 0.9366921180785186,
      "flos": 18655270260480.0,
      "grad_norm": 2.6843477379406973,
      "language_loss": 0.68990958,
      "learning_rate": 4.182493420022526e-08,
      "loss": 0.71194178,
      "num_input_tokens_seen": 168383215,
      "step": 7790,
      "time_per_iteration": 2.5598158836364746
    },
    {
      "auxiliary_loss_clip": 0.01225401,
      "auxiliary_loss_mlp": 0.0102473,
      "balance_loss_clip": 1.04600644,
      "balance_loss_mlp": 1.01828051,
      "epoch": 0.9368123609691577,
      "flos": 25774955815680.0,
      "grad_norm": 3.7622765839679975,
      "language_loss": 0.78405237,
      "learning_rate": 4.166660918074139e-08,
      "loss": 0.80655366,
      "num_input_tokens_seen": 168403120,
      "step": 7791,
      "time_per_iteration": 2.7598819732666016
    },
    {
      "auxiliary_loss_clip": 0.01317939,
      "auxiliary_loss_mlp": 0.01021718,
      "balance_loss_clip": 1.04486144,
      "balance_loss_mlp": 1.01510441,
      "epoch": 0.9369326038597968,
      "flos": 25553386771200.0,
      "grad_norm": 1.4095714046721284,
      "language_loss": 0.73268402,
      "learning_rate": 4.15085812388758e-08,
      "loss": 0.75608063,
      "num_input_tokens_seen": 168425340,
      "step": 7792,
      "time_per_iteration": 2.709152936935425
    },
    {
      "auxiliary_loss_clip": 0.01269265,
      "auxiliary_loss_mlp": 0.01027831,
      "balance_loss_clip": 1.0458827,
      "balance_loss_mlp": 1.02075267,
      "epoch": 0.9370528467504359,
      "flos": 23220019370880.0,
      "grad_norm": 2.030050301832805,
      "language_loss": 0.78865612,
      "learning_rate": 4.135085039860153e-08,
      "loss": 0.81162709,
      "num_input_tokens_seen": 168444740,
      "step": 7793,
      "time_per_iteration": 2.705562114715576
    },
    {
      "auxiliary_loss_clip": 0.01266614,
      "auxiliary_loss_mlp": 0.01023462,
      "balance_loss_clip": 1.04692197,
      "balance_loss_mlp": 1.01596951,
      "epoch": 0.9371730896410749,
      "flos": 24967468120320.0,
      "grad_norm": 2.1559037928504803,
      "language_loss": 0.78582972,
      "learning_rate": 4.1193416683845906e-08,
      "loss": 0.80873042,
      "num_input_tokens_seen": 168463670,
      "step": 7794,
      "time_per_iteration": 2.7035701274871826
    },
    {
      "auxiliary_loss_clip": 0.01319772,
      "auxiliary_loss_mlp": 0.01024646,
      "balance_loss_clip": 1.04557753,
      "balance_loss_mlp": 1.0184145,
      "epoch": 0.9372933325317141,
      "flos": 15553091134080.0,
      "grad_norm": 4.1954482502577,
      "language_loss": 0.83661544,
      "learning_rate": 4.103628011849136e-08,
      "loss": 0.86005968,
      "num_input_tokens_seen": 168479030,
      "step": 7795,
      "time_per_iteration": 2.6687703132629395
    },
    {
      "auxiliary_loss_clip": 0.01274392,
      "auxiliary_loss_mlp": 0.01024992,
      "balance_loss_clip": 1.04718637,
      "balance_loss_mlp": 1.01754773,
      "epoch": 0.9374135754223532,
      "flos": 21871861182720.0,
      "grad_norm": 1.9418738709170267,
      "language_loss": 0.75851369,
      "learning_rate": 4.0879440726375506e-08,
      "loss": 0.78150755,
      "num_input_tokens_seen": 168496815,
      "step": 7796,
      "time_per_iteration": 2.727123737335205
    },
    {
      "auxiliary_loss_clip": 0.01271665,
      "auxiliary_loss_mlp": 0.01025332,
      "balance_loss_clip": 1.04327416,
      "balance_loss_mlp": 1.01830173,
      "epoch": 0.9375338183129922,
      "flos": 22631048064000.0,
      "grad_norm": 2.6464163581679965,
      "language_loss": 0.56367534,
      "learning_rate": 4.0722898531291074e-08,
      "loss": 0.58664531,
      "num_input_tokens_seen": 168514055,
      "step": 7797,
      "time_per_iteration": 2.7609939575195312
    },
    {
      "auxiliary_loss_clip": 0.01276988,
      "auxiliary_loss_mlp": 0.0102868,
      "balance_loss_clip": 1.0477314,
      "balance_loss_mlp": 1.02160478,
      "epoch": 0.9376540612036314,
      "flos": 26104292640000.0,
      "grad_norm": 1.8614383585557865,
      "language_loss": 0.7693063,
      "learning_rate": 4.0566653556985295e-08,
      "loss": 0.79236293,
      "num_input_tokens_seen": 168534600,
      "step": 7798,
      "time_per_iteration": 2.9020912647247314
    },
    {
      "auxiliary_loss_clip": 0.01470398,
      "auxiliary_loss_mlp": 0.01025951,
      "balance_loss_clip": 1.04192376,
      "balance_loss_mlp": 1.01869702,
      "epoch": 0.9377743040942704,
      "flos": 19717580016000.0,
      "grad_norm": 3.5236179013408884,
      "language_loss": 0.81386608,
      "learning_rate": 4.0410705827159886e-08,
      "loss": 0.83882958,
      "num_input_tokens_seen": 168551895,
      "step": 7799,
      "time_per_iteration": 3.9069552421569824
    },
    {
      "auxiliary_loss_clip": 0.0127221,
      "auxiliary_loss_mlp": 0.01023203,
      "balance_loss_clip": 1.04274273,
      "balance_loss_mlp": 1.01598477,
      "epoch": 0.9378945469849095,
      "flos": 15267530010240.0,
      "grad_norm": 3.0783342080545824,
      "language_loss": 0.71168971,
      "learning_rate": 4.0255055365472356e-08,
      "loss": 0.73464382,
      "num_input_tokens_seen": 168569990,
      "step": 7800,
      "time_per_iteration": 3.121744394302368
    },
    {
      "auxiliary_loss_clip": 0.01420758,
      "auxiliary_loss_mlp": 0.01030277,
      "balance_loss_clip": 1.03855026,
      "balance_loss_mlp": 1.02357149,
      "epoch": 0.9380147898755486,
      "flos": 20591394174720.0,
      "grad_norm": 3.03388606472478,
      "language_loss": 0.75337321,
      "learning_rate": 4.009970219553471e-08,
      "loss": 0.77788359,
      "num_input_tokens_seen": 168586940,
      "step": 7801,
      "time_per_iteration": 3.9288625717163086
    },
    {
      "auxiliary_loss_clip": 0.01226171,
      "auxiliary_loss_mlp": 0.01021736,
      "balance_loss_clip": 1.04858053,
      "balance_loss_mlp": 1.01460111,
      "epoch": 0.9381350327661877,
      "flos": 26281116316800.0,
      "grad_norm": 3.2015040629572686,
      "language_loss": 0.76808023,
      "learning_rate": 3.99446463409141e-08,
      "loss": 0.79055929,
      "num_input_tokens_seen": 168604795,
      "step": 7802,
      "time_per_iteration": 2.6398537158966064
    },
    {
      "auxiliary_loss_clip": 0.01229548,
      "auxiliary_loss_mlp": 0.01028964,
      "balance_loss_clip": 1.04737353,
      "balance_loss_mlp": 1.02187085,
      "epoch": 0.9382552756568268,
      "flos": 23586344225280.0,
      "grad_norm": 2.8047452198016125,
      "language_loss": 0.6903255,
      "learning_rate": 3.978988782513215e-08,
      "loss": 0.71291065,
      "num_input_tokens_seen": 168622290,
      "step": 7803,
      "time_per_iteration": 2.652686834335327
    },
    {
      "auxiliary_loss_clip": 0.01226201,
      "auxiliary_loss_mlp": 0.01028079,
      "balance_loss_clip": 1.04719257,
      "balance_loss_mlp": 1.02131987,
      "epoch": 0.9383755185474659,
      "flos": 28438809275520.0,
      "grad_norm": 1.799681124036644,
      "language_loss": 0.76447511,
      "learning_rate": 3.963542667166586e-08,
      "loss": 0.78701788,
      "num_input_tokens_seen": 168642395,
      "step": 7804,
      "time_per_iteration": 3.621711492538452
    },
    {
      "auxiliary_loss_clip": 0.01219703,
      "auxiliary_loss_mlp": 0.01024717,
      "balance_loss_clip": 1.04652643,
      "balance_loss_mlp": 1.01805639,
      "epoch": 0.938495761438105,
      "flos": 20449583280000.0,
      "grad_norm": 6.104763460693421,
      "language_loss": 0.68092996,
      "learning_rate": 3.9481262903946486e-08,
      "loss": 0.70337415,
      "num_input_tokens_seen": 168661840,
      "step": 7805,
      "time_per_iteration": 2.6802186965942383
    },
    {
      "auxiliary_loss_clip": 0.01280349,
      "auxiliary_loss_mlp": 0.00999109,
      "balance_loss_clip": 1.00625229,
      "balance_loss_mlp": 0.99820852,
      "epoch": 0.938616004328744,
      "flos": 69302711658240.0,
      "grad_norm": 0.7788715424616803,
      "language_loss": 0.54444706,
      "learning_rate": 3.932739654536066e-08,
      "loss": 0.56724161,
      "num_input_tokens_seen": 168724540,
      "step": 7806,
      "time_per_iteration": 3.2498064041137695
    },
    {
      "auxiliary_loss_clip": 0.01219561,
      "auxiliary_loss_mlp": 0.01022451,
      "balance_loss_clip": 1.04741836,
      "balance_loss_mlp": 1.01575398,
      "epoch": 0.9387362472193832,
      "flos": 18911636605440.0,
      "grad_norm": 2.265981059557103,
      "language_loss": 0.74418378,
      "learning_rate": 3.917382761925014e-08,
      "loss": 0.76660395,
      "num_input_tokens_seen": 168740375,
      "step": 7807,
      "time_per_iteration": 2.5700900554656982
    },
    {
      "auxiliary_loss_clip": 0.01216816,
      "auxiliary_loss_mlp": 0.01024341,
      "balance_loss_clip": 1.04724598,
      "balance_loss_mlp": 1.01764405,
      "epoch": 0.9388564901100223,
      "flos": 26501967089280.0,
      "grad_norm": 1.7986561298312898,
      "language_loss": 0.78901815,
      "learning_rate": 3.9020556148910754e-08,
      "loss": 0.81142974,
      "num_input_tokens_seen": 168759730,
      "step": 7808,
      "time_per_iteration": 2.6741905212402344
    },
    {
      "auxiliary_loss_clip": 0.01169776,
      "auxiliary_loss_mlp": 0.00997922,
      "balance_loss_clip": 1.00610459,
      "balance_loss_mlp": 0.99708128,
      "epoch": 0.9389767330006613,
      "flos": 58941083157120.0,
      "grad_norm": 0.7063008486423541,
      "language_loss": 0.56624949,
      "learning_rate": 3.8867582157593895e-08,
      "loss": 0.58792651,
      "num_input_tokens_seen": 168813935,
      "step": 7809,
      "time_per_iteration": 4.010875701904297
    },
    {
      "auxiliary_loss_clip": 0.01219182,
      "auxiliary_loss_mlp": 0.01022203,
      "balance_loss_clip": 1.0497309,
      "balance_loss_mlp": 1.0156523,
      "epoch": 0.9390969758913005,
      "flos": 31102554994560.0,
      "grad_norm": 2.4822216211766635,
      "language_loss": 0.76758981,
      "learning_rate": 3.871490566850544e-08,
      "loss": 0.79000378,
      "num_input_tokens_seen": 168838145,
      "step": 7810,
      "time_per_iteration": 2.728760004043579
    },
    {
      "auxiliary_loss_clip": 0.01269127,
      "auxiliary_loss_mlp": 0.01020167,
      "balance_loss_clip": 1.04534805,
      "balance_loss_mlp": 1.01340497,
      "epoch": 0.9392172187819395,
      "flos": 22419391173120.0,
      "grad_norm": 1.7776844108076166,
      "language_loss": 0.70651203,
      "learning_rate": 3.856252670480642e-08,
      "loss": 0.72940499,
      "num_input_tokens_seen": 168856805,
      "step": 7811,
      "time_per_iteration": 2.726381778717041
    },
    {
      "auxiliary_loss_clip": 0.01271443,
      "auxiliary_loss_mlp": 0.01022836,
      "balance_loss_clip": 1.04223049,
      "balance_loss_mlp": 1.01585054,
      "epoch": 0.9393374616725786,
      "flos": 19719483436800.0,
      "grad_norm": 2.1429204283715806,
      "language_loss": 0.81527936,
      "learning_rate": 3.841044528961279e-08,
      "loss": 0.83822221,
      "num_input_tokens_seen": 168874600,
      "step": 7812,
      "time_per_iteration": 2.626502752304077
    },
    {
      "auxiliary_loss_clip": 0.01172788,
      "auxiliary_loss_mlp": 0.0102625,
      "balance_loss_clip": 1.04717743,
      "balance_loss_mlp": 1.01870716,
      "epoch": 0.9394577045632178,
      "flos": 24170215800960.0,
      "grad_norm": 1.8067567426507496,
      "language_loss": 0.78877449,
      "learning_rate": 3.825866144599477e-08,
      "loss": 0.81076485,
      "num_input_tokens_seen": 168893655,
      "step": 7813,
      "time_per_iteration": 2.618201494216919
    },
    {
      "auxiliary_loss_clip": 0.01270754,
      "auxiliary_loss_mlp": 0.01025046,
      "balance_loss_clip": 1.04552758,
      "balance_loss_mlp": 1.01790559,
      "epoch": 0.9395779474538568,
      "flos": 19023929498880.0,
      "grad_norm": 2.367181984966003,
      "language_loss": 0.7547012,
      "learning_rate": 3.8107175196978145e-08,
      "loss": 0.77765918,
      "num_input_tokens_seen": 168909960,
      "step": 7814,
      "time_per_iteration": 2.6070165634155273
    },
    {
      "auxiliary_loss_clip": 0.01328002,
      "auxiliary_loss_mlp": 0.01025329,
      "balance_loss_clip": 1.04826546,
      "balance_loss_mlp": 1.01895475,
      "epoch": 0.9396981903444959,
      "flos": 14319129260160.0,
      "grad_norm": 2.3042926569760755,
      "language_loss": 0.77119911,
      "learning_rate": 3.7955986565542996e-08,
      "loss": 0.79473239,
      "num_input_tokens_seen": 168928040,
      "step": 7815,
      "time_per_iteration": 2.6926443576812744
    },
    {
      "auxiliary_loss_clip": 0.01319879,
      "auxiliary_loss_mlp": 0.01020181,
      "balance_loss_clip": 1.04310584,
      "balance_loss_mlp": 1.01343131,
      "epoch": 0.9398184332351349,
      "flos": 34787564202240.0,
      "grad_norm": 2.2395194249864976,
      "language_loss": 0.68358821,
      "learning_rate": 3.780509557462497e-08,
      "loss": 0.70698881,
      "num_input_tokens_seen": 168948240,
      "step": 7816,
      "time_per_iteration": 2.8270039558410645
    },
    {
      "auxiliary_loss_clip": 0.01270014,
      "auxiliary_loss_mlp": 0.01028195,
      "balance_loss_clip": 1.04338765,
      "balance_loss_mlp": 1.02107251,
      "epoch": 0.9399386761257741,
      "flos": 25372253462400.0,
      "grad_norm": 1.589681271638446,
      "language_loss": 0.75704533,
      "learning_rate": 3.765450224711375e-08,
      "loss": 0.78002739,
      "num_input_tokens_seen": 168968745,
      "step": 7817,
      "time_per_iteration": 2.634904623031616
    },
    {
      "auxiliary_loss_clip": 0.01272823,
      "auxiliary_loss_mlp": 0.0102,
      "balance_loss_clip": 1.04936361,
      "balance_loss_mlp": 1.01334786,
      "epoch": 0.9400589190164131,
      "flos": 27304965584640.0,
      "grad_norm": 2.1592043760598947,
      "language_loss": 0.80082047,
      "learning_rate": 3.750420660585396e-08,
      "loss": 0.82374871,
      "num_input_tokens_seen": 168990685,
      "step": 7818,
      "time_per_iteration": 2.65928316116333
    },
    {
      "auxiliary_loss_clip": 0.01171334,
      "auxiliary_loss_mlp": 0.01024482,
      "balance_loss_clip": 1.04975748,
      "balance_loss_mlp": 1.01773524,
      "epoch": 0.9401791619070522,
      "flos": 23399859790080.0,
      "grad_norm": 1.958116767396989,
      "language_loss": 0.80092776,
      "learning_rate": 3.735420867364603e-08,
      "loss": 0.82288587,
      "num_input_tokens_seen": 169011665,
      "step": 7819,
      "time_per_iteration": 2.587510347366333
    },
    {
      "auxiliary_loss_clip": 0.01415847,
      "auxiliary_loss_mlp": 0.01021489,
      "balance_loss_clip": 1.03825879,
      "balance_loss_mlp": 1.01539469,
      "epoch": 0.9402994047976914,
      "flos": 35881403120640.0,
      "grad_norm": 1.748219621314816,
      "language_loss": 0.61773652,
      "learning_rate": 3.7204508473244186e-08,
      "loss": 0.64210987,
      "num_input_tokens_seen": 169035290,
      "step": 7820,
      "time_per_iteration": 2.907794952392578
    },
    {
      "auxiliary_loss_clip": 0.01459272,
      "auxiliary_loss_mlp": 0.01028242,
      "balance_loss_clip": 1.03992414,
      "balance_loss_mlp": 1.02145338,
      "epoch": 0.9404196476883304,
      "flos": 22236821320320.0,
      "grad_norm": 1.6804941775891205,
      "language_loss": 0.69470435,
      "learning_rate": 3.7055106027357395e-08,
      "loss": 0.71957952,
      "num_input_tokens_seen": 169055155,
      "step": 7821,
      "time_per_iteration": 2.8467531204223633
    },
    {
      "auxiliary_loss_clip": 0.01215148,
      "auxiliary_loss_mlp": 0.01019839,
      "balance_loss_clip": 1.04449487,
      "balance_loss_mlp": 1.01313639,
      "epoch": 0.9405398905789695,
      "flos": 18915802583040.0,
      "grad_norm": 2.1026785217184663,
      "language_loss": 0.71671307,
      "learning_rate": 3.690600135865063e-08,
      "loss": 0.73906296,
      "num_input_tokens_seen": 169072080,
      "step": 7822,
      "time_per_iteration": 2.805140495300293
    },
    {
      "auxiliary_loss_clip": 0.01169929,
      "auxiliary_loss_mlp": 0.00997939,
      "balance_loss_clip": 1.00693977,
      "balance_loss_mlp": 0.99706328,
      "epoch": 0.9406601334696086,
      "flos": 70274130048000.0,
      "grad_norm": 0.78395613323091,
      "language_loss": 0.58023262,
      "learning_rate": 3.675719448974246e-08,
      "loss": 0.60191131,
      "num_input_tokens_seen": 169137170,
      "step": 7823,
      "time_per_iteration": 3.3798115253448486
    },
    {
      "auxiliary_loss_clip": 0.0137335,
      "auxiliary_loss_mlp": 0.02565756,
      "balance_loss_clip": 1.04512489,
      "balance_loss_mlp": 0.99993646,
      "epoch": 0.9407803763602477,
      "flos": 22165071903360.0,
      "grad_norm": 2.10487629180476,
      "language_loss": 0.60542142,
      "learning_rate": 3.6608685443207054e-08,
      "loss": 0.64481246,
      "num_input_tokens_seen": 169156320,
      "step": 7824,
      "time_per_iteration": 2.754261016845703
    },
    {
      "auxiliary_loss_clip": 0.01320631,
      "auxiliary_loss_mlp": 0.01022922,
      "balance_loss_clip": 1.04309607,
      "balance_loss_mlp": 1.0162853,
      "epoch": 0.9409006192508867,
      "flos": 18879496911360.0,
      "grad_norm": 2.326311077982442,
      "language_loss": 0.66801107,
      "learning_rate": 3.646047424157306e-08,
      "loss": 0.69144654,
      "num_input_tokens_seen": 169173295,
      "step": 7825,
      "time_per_iteration": 3.8016135692596436
    },
    {
      "auxiliary_loss_clip": 0.01273035,
      "auxiliary_loss_mlp": 0.0102479,
      "balance_loss_clip": 1.0490278,
      "balance_loss_mlp": 1.01715446,
      "epoch": 0.9410208621415259,
      "flos": 23368258800000.0,
      "grad_norm": 3.960701598183781,
      "language_loss": 0.68926591,
      "learning_rate": 3.631256090732382e-08,
      "loss": 0.71224415,
      "num_input_tokens_seen": 169193755,
      "step": 7826,
      "time_per_iteration": 2.668837308883667
    },
    {
      "auxiliary_loss_clip": 0.01322823,
      "auxiliary_loss_mlp": 0.01026256,
      "balance_loss_clip": 1.04630899,
      "balance_loss_mlp": 1.01962245,
      "epoch": 0.941141105032165,
      "flos": 22742227635840.0,
      "grad_norm": 1.6440466404120444,
      "language_loss": 0.83175302,
      "learning_rate": 3.6164945462897833e-08,
      "loss": 0.8552438,
      "num_input_tokens_seen": 169213045,
      "step": 7827,
      "time_per_iteration": 3.650991201400757
    },
    {
      "auxiliary_loss_clip": 0.01220038,
      "auxiliary_loss_mlp": 0.02562405,
      "balance_loss_clip": 1.0487771,
      "balance_loss_mlp": 0.99990612,
      "epoch": 0.941261347922804,
      "flos": 20704908130560.0,
      "grad_norm": 2.3523004645171706,
      "language_loss": 0.7577306,
      "learning_rate": 3.6017627930687856e-08,
      "loss": 0.795555,
      "num_input_tokens_seen": 169232870,
      "step": 7828,
      "time_per_iteration": 2.6554887294769287
    },
    {
      "auxiliary_loss_clip": 0.01367304,
      "auxiliary_loss_mlp": 0.01021456,
      "balance_loss_clip": 1.04015541,
      "balance_loss_mlp": 1.0144614,
      "epoch": 0.9413815908134432,
      "flos": 19421998997760.0,
      "grad_norm": 3.243715009580539,
      "language_loss": 0.773794,
      "learning_rate": 3.587060833304267e-08,
      "loss": 0.79768157,
      "num_input_tokens_seen": 169251060,
      "step": 7829,
      "time_per_iteration": 2.6814963817596436
    },
    {
      "auxiliary_loss_clip": 0.01225168,
      "auxiliary_loss_mlp": 0.01030603,
      "balance_loss_clip": 1.04892421,
      "balance_loss_mlp": 1.02353358,
      "epoch": 0.9415018337040822,
      "flos": 17493452853120.0,
      "grad_norm": 2.301280753462474,
      "language_loss": 0.64281952,
      "learning_rate": 3.5723886692264225e-08,
      "loss": 0.6653772,
      "num_input_tokens_seen": 169268600,
      "step": 7830,
      "time_per_iteration": 3.5092315673828125
    },
    {
      "auxiliary_loss_clip": 0.01268346,
      "auxiliary_loss_mlp": 0.01024634,
      "balance_loss_clip": 1.04392695,
      "balance_loss_mlp": 1.01852441,
      "epoch": 0.9416220765947213,
      "flos": 31831613343360.0,
      "grad_norm": 3.7575859321750946,
      "language_loss": 0.61906224,
      "learning_rate": 3.557746303061071e-08,
      "loss": 0.64199209,
      "num_input_tokens_seen": 169290355,
      "step": 7831,
      "time_per_iteration": 2.70263934135437
    },
    {
      "auxiliary_loss_clip": 0.01269309,
      "auxiliary_loss_mlp": 0.01025065,
      "balance_loss_clip": 1.04495919,
      "balance_loss_mlp": 1.01900661,
      "epoch": 0.9417423194853605,
      "flos": 23511973115520.0,
      "grad_norm": 1.6795847693451915,
      "language_loss": 0.72507656,
      "learning_rate": 3.543133737029391e-08,
      "loss": 0.74802029,
      "num_input_tokens_seen": 169310865,
      "step": 7832,
      "time_per_iteration": 2.7003724575042725
    },
    {
      "auxiliary_loss_clip": 0.01227407,
      "auxiliary_loss_mlp": 0.01022803,
      "balance_loss_clip": 1.04822087,
      "balance_loss_mlp": 1.01554942,
      "epoch": 0.9418625623759995,
      "flos": 23915106432000.0,
      "grad_norm": 2.060320809966886,
      "language_loss": 0.69341028,
      "learning_rate": 3.5285509733481214e-08,
      "loss": 0.71591234,
      "num_input_tokens_seen": 169330590,
      "step": 7833,
      "time_per_iteration": 2.66176700592041
    },
    {
      "auxiliary_loss_clip": 0.01224413,
      "auxiliary_loss_mlp": 0.01027567,
      "balance_loss_clip": 1.04712319,
      "balance_loss_mlp": 1.02020931,
      "epoch": 0.9419828052666386,
      "flos": 18076965292800.0,
      "grad_norm": 1.6300509161129637,
      "language_loss": 0.76369846,
      "learning_rate": 3.513998014229469e-08,
      "loss": 0.78621823,
      "num_input_tokens_seen": 169349540,
      "step": 7834,
      "time_per_iteration": 2.6615893840789795
    },
    {
      "auxiliary_loss_clip": 0.01169182,
      "auxiliary_loss_mlp": 0.01024285,
      "balance_loss_clip": 1.04572344,
      "balance_loss_mlp": 1.01744795,
      "epoch": 0.9421030481572777,
      "flos": 17712328377600.0,
      "grad_norm": 8.719509777245694,
      "language_loss": 0.86428106,
      "learning_rate": 3.499474861881069e-08,
      "loss": 0.88621581,
      "num_input_tokens_seen": 169366765,
      "step": 7835,
      "time_per_iteration": 3.608135938644409
    },
    {
      "auxiliary_loss_clip": 0.01419382,
      "auxiliary_loss_mlp": 0.01026209,
      "balance_loss_clip": 1.04301155,
      "balance_loss_mlp": 1.0189786,
      "epoch": 0.9422232910479168,
      "flos": 20194114775040.0,
      "grad_norm": 1.8393513058277469,
      "language_loss": 0.67939782,
      "learning_rate": 3.4849815185061136e-08,
      "loss": 0.70385367,
      "num_input_tokens_seen": 169386655,
      "step": 7836,
      "time_per_iteration": 2.727985382080078
    },
    {
      "auxiliary_loss_clip": 0.0122124,
      "auxiliary_loss_mlp": 0.01023385,
      "balance_loss_clip": 1.04568779,
      "balance_loss_mlp": 1.0168258,
      "epoch": 0.9423435339385559,
      "flos": 18442571875200.0,
      "grad_norm": 2.0789199579757143,
      "language_loss": 0.76339692,
      "learning_rate": 3.470517986303223e-08,
      "loss": 0.78584313,
      "num_input_tokens_seen": 169405640,
      "step": 7837,
      "time_per_iteration": 2.6368422508239746
    },
    {
      "auxiliary_loss_clip": 0.01319784,
      "auxiliary_loss_mlp": 0.01024361,
      "balance_loss_clip": 1.04650593,
      "balance_loss_mlp": 1.01752758,
      "epoch": 0.942463776829195,
      "flos": 20080636732800.0,
      "grad_norm": 1.6079317276356988,
      "language_loss": 0.79236639,
      "learning_rate": 3.4560842674664856e-08,
      "loss": 0.81580782,
      "num_input_tokens_seen": 169424155,
      "step": 7838,
      "time_per_iteration": 2.6587727069854736
    },
    {
      "auxiliary_loss_clip": 0.01221683,
      "auxiliary_loss_mlp": 0.01022222,
      "balance_loss_clip": 1.04411829,
      "balance_loss_mlp": 1.01571298,
      "epoch": 0.9425840197198341,
      "flos": 22636255536000.0,
      "grad_norm": 1.9438233937997522,
      "language_loss": 0.75275016,
      "learning_rate": 3.441680364185506e-08,
      "loss": 0.77518922,
      "num_input_tokens_seen": 169444025,
      "step": 7839,
      "time_per_iteration": 2.670668125152588
    },
    {
      "auxiliary_loss_clip": 0.01274583,
      "auxiliary_loss_mlp": 0.01028136,
      "balance_loss_clip": 1.0475347,
      "balance_loss_mlp": 1.02091789,
      "epoch": 0.9427042626104731,
      "flos": 19937892084480.0,
      "grad_norm": 3.0839895707673954,
      "language_loss": 0.74857801,
      "learning_rate": 3.427306278645314e-08,
      "loss": 0.77160525,
      "num_input_tokens_seen": 169462480,
      "step": 7840,
      "time_per_iteration": 2.639906406402588
    },
    {
      "auxiliary_loss_clip": 0.01365411,
      "auxiliary_loss_mlp": 0.01024738,
      "balance_loss_clip": 1.0434103,
      "balance_loss_mlp": 1.01793408,
      "epoch": 0.9428245055011123,
      "flos": 22856998567680.0,
      "grad_norm": 1.7125817416605325,
      "language_loss": 0.73373443,
      "learning_rate": 3.4129620130264767e-08,
      "loss": 0.75763589,
      "num_input_tokens_seen": 169480840,
      "step": 7841,
      "time_per_iteration": 2.7568180561065674
    },
    {
      "auxiliary_loss_clip": 0.01274751,
      "auxiliary_loss_mlp": 0.02564972,
      "balance_loss_clip": 1.04660392,
      "balance_loss_mlp": 0.99990416,
      "epoch": 0.9429447483917514,
      "flos": 20951757371520.0,
      "grad_norm": 2.3163433994749445,
      "language_loss": 0.77774084,
      "learning_rate": 3.398647569505009e-08,
      "loss": 0.81613809,
      "num_input_tokens_seen": 169498265,
      "step": 7842,
      "time_per_iteration": 2.611524820327759
    },
    {
      "auxiliary_loss_clip": 0.01321634,
      "auxiliary_loss_mlp": 0.01023634,
      "balance_loss_clip": 1.04377842,
      "balance_loss_mlp": 1.01679754,
      "epoch": 0.9430649912823904,
      "flos": 18843658116480.0,
      "grad_norm": 2.7487274902195398,
      "language_loss": 0.75006974,
      "learning_rate": 3.384362950252373e-08,
      "loss": 0.77352238,
      "num_input_tokens_seen": 169515235,
      "step": 7843,
      "time_per_iteration": 2.678293228149414
    },
    {
      "auxiliary_loss_clip": 0.01268676,
      "auxiliary_loss_mlp": 0.01023088,
      "balance_loss_clip": 1.04428387,
      "balance_loss_mlp": 1.01657665,
      "epoch": 0.9431852341730296,
      "flos": 32556038837760.0,
      "grad_norm": 1.8978414734320195,
      "language_loss": 0.57322311,
      "learning_rate": 3.3701081574355473e-08,
      "loss": 0.59614068,
      "num_input_tokens_seen": 169537195,
      "step": 7844,
      "time_per_iteration": 2.7819316387176514
    },
    {
      "auxiliary_loss_clip": 0.01067173,
      "auxiliary_loss_mlp": 0.00999418,
      "balance_loss_clip": 1.00690317,
      "balance_loss_mlp": 0.9985652,
      "epoch": 0.9433054770636686,
      "flos": 66904490252160.0,
      "grad_norm": 0.6405429864430534,
      "language_loss": 0.51653421,
      "learning_rate": 3.3558831932169796e-08,
      "loss": 0.53720009,
      "num_input_tokens_seen": 169605865,
      "step": 7845,
      "time_per_iteration": 3.317901372909546
    },
    {
      "auxiliary_loss_clip": 0.01219562,
      "auxiliary_loss_mlp": 0.01026687,
      "balance_loss_clip": 1.04838777,
      "balance_loss_mlp": 1.01979423,
      "epoch": 0.9434257199543077,
      "flos": 26140346916480.0,
      "grad_norm": 1.871089389278832,
      "language_loss": 0.88656461,
      "learning_rate": 3.341688059754588e-08,
      "loss": 0.9090271,
      "num_input_tokens_seen": 169621520,
      "step": 7846,
      "time_per_iteration": 2.6857049465179443
    },
    {
      "auxiliary_loss_clip": 0.01329309,
      "auxiliary_loss_mlp": 0.02565364,
      "balance_loss_clip": 1.04420447,
      "balance_loss_mlp": 0.99993145,
      "epoch": 0.9435459628449467,
      "flos": 25003486483200.0,
      "grad_norm": 2.4618544063761805,
      "language_loss": 0.7784276,
      "learning_rate": 3.327522759201762e-08,
      "loss": 0.81737435,
      "num_input_tokens_seen": 169641390,
      "step": 7847,
      "time_per_iteration": 2.673247814178467
    },
    {
      "auxiliary_loss_clip": 0.01321898,
      "auxiliary_loss_mlp": 0.01025852,
      "balance_loss_clip": 1.04608512,
      "balance_loss_mlp": 1.01886392,
      "epoch": 0.9436662057355859,
      "flos": 22163240309760.0,
      "grad_norm": 2.3302821373845473,
      "language_loss": 0.67467129,
      "learning_rate": 3.313387293707359e-08,
      "loss": 0.69814879,
      "num_input_tokens_seen": 169660095,
      "step": 7848,
      "time_per_iteration": 2.7527167797088623
    },
    {
      "auxiliary_loss_clip": 0.01323081,
      "auxiliary_loss_mlp": 0.01024219,
      "balance_loss_clip": 1.04766846,
      "balance_loss_mlp": 1.01698899,
      "epoch": 0.943786448626225,
      "flos": 20118522602880.0,
      "grad_norm": 2.1800263765041996,
      "language_loss": 0.68424505,
      "learning_rate": 3.29928166541571e-08,
      "loss": 0.70771801,
      "num_input_tokens_seen": 169679050,
      "step": 7849,
      "time_per_iteration": 2.657149314880371
    },
    {
      "auxiliary_loss_clip": 0.01265899,
      "auxiliary_loss_mlp": 0.01026194,
      "balance_loss_clip": 1.04416239,
      "balance_loss_mlp": 1.01892185,
      "epoch": 0.943906691516864,
      "flos": 22090808534400.0,
      "grad_norm": 2.1590195324876826,
      "language_loss": 0.80777359,
      "learning_rate": 3.2852058764666346e-08,
      "loss": 0.83069444,
      "num_input_tokens_seen": 169698150,
      "step": 7850,
      "time_per_iteration": 2.6995131969451904
    },
    {
      "auxiliary_loss_clip": 0.01308793,
      "auxiliary_loss_mlp": 0.01022033,
      "balance_loss_clip": 1.04658437,
      "balance_loss_mlp": 1.01579285,
      "epoch": 0.9440269344075032,
      "flos": 35298501212160.0,
      "grad_norm": 2.6663616432996124,
      "language_loss": 0.68594193,
      "learning_rate": 3.2711599289954264e-08,
      "loss": 0.70925021,
      "num_input_tokens_seen": 169722185,
      "step": 7851,
      "time_per_iteration": 3.7326853275299072
    },
    {
      "auxiliary_loss_clip": 0.01418399,
      "auxiliary_loss_mlp": 0.01024866,
      "balance_loss_clip": 1.04174972,
      "balance_loss_mlp": 1.01798177,
      "epoch": 0.9441471772981422,
      "flos": 19238136255360.0,
      "grad_norm": 2.055785699784817,
      "language_loss": 0.78081656,
      "learning_rate": 3.257143825132847e-08,
      "loss": 0.80524921,
      "num_input_tokens_seen": 169740355,
      "step": 7852,
      "time_per_iteration": 2.7102925777435303
    },
    {
      "auxiliary_loss_clip": 0.0127095,
      "auxiliary_loss_mlp": 0.01022759,
      "balance_loss_clip": 1.0450865,
      "balance_loss_mlp": 1.01676571,
      "epoch": 0.9442674201887813,
      "flos": 25739799379200.0,
      "grad_norm": 1.9179132295994927,
      "language_loss": 0.76189929,
      "learning_rate": 3.243157567005106e-08,
      "loss": 0.78483641,
      "num_input_tokens_seen": 169758535,
      "step": 7853,
      "time_per_iteration": 3.569077968597412
    },
    {
      "auxiliary_loss_clip": 0.01179717,
      "auxiliary_loss_mlp": 0.01030882,
      "balance_loss_clip": 1.05430436,
      "balance_loss_mlp": 1.02340734,
      "epoch": 0.9443876630794205,
      "flos": 15523321737600.0,
      "grad_norm": 2.4052174926809893,
      "language_loss": 0.64054978,
      "learning_rate": 3.2292011567339296e-08,
      "loss": 0.66265583,
      "num_input_tokens_seen": 169776340,
      "step": 7854,
      "time_per_iteration": 2.5195908546447754
    },
    {
      "auxiliary_loss_clip": 0.01223622,
      "auxiliary_loss_mlp": 0.02563056,
      "balance_loss_clip": 1.04680824,
      "balance_loss_mlp": 0.99991298,
      "epoch": 0.9445079059700595,
      "flos": 13400821128960.0,
      "grad_norm": 2.0958728095011443,
      "language_loss": 0.56007719,
      "learning_rate": 3.21527459643649e-08,
      "loss": 0.59794402,
      "num_input_tokens_seen": 169793225,
      "step": 7855,
      "time_per_iteration": 3.5367391109466553
    },
    {
      "auxiliary_loss_clip": 0.01226882,
      "auxiliary_loss_mlp": 0.01023286,
      "balance_loss_clip": 1.04971647,
      "balance_loss_mlp": 1.01647902,
      "epoch": 0.9446281488606986,
      "flos": 23659242877440.0,
      "grad_norm": 2.4058220870223255,
      "language_loss": 0.74361205,
      "learning_rate": 3.2013778882254536e-08,
      "loss": 0.7661137,
      "num_input_tokens_seen": 169812020,
      "step": 7856,
      "time_per_iteration": 2.6752138137817383
    },
    {
      "auxiliary_loss_clip": 0.01219509,
      "auxiliary_loss_mlp": 0.01022638,
      "balance_loss_clip": 1.04730034,
      "balance_loss_mlp": 1.01594472,
      "epoch": 0.9447483917513377,
      "flos": 25557337267200.0,
      "grad_norm": 1.6800690916219647,
      "language_loss": 0.75645727,
      "learning_rate": 3.1875110342088676e-08,
      "loss": 0.77887875,
      "num_input_tokens_seen": 169833470,
      "step": 7857,
      "time_per_iteration": 2.6489193439483643
    },
    {
      "auxiliary_loss_clip": 0.01265194,
      "auxiliary_loss_mlp": 0.01027365,
      "balance_loss_clip": 1.04606771,
      "balance_loss_mlp": 1.02033794,
      "epoch": 0.9448686346419768,
      "flos": 24535463247360.0,
      "grad_norm": 1.8506240206849007,
      "language_loss": 0.6555438,
      "learning_rate": 3.1736740364904035e-08,
      "loss": 0.67846942,
      "num_input_tokens_seen": 169854000,
      "step": 7858,
      "time_per_iteration": 2.6844778060913086
    },
    {
      "auxiliary_loss_clip": 0.01368981,
      "auxiliary_loss_mlp": 0.02563974,
      "balance_loss_clip": 1.04160547,
      "balance_loss_mlp": 0.99993008,
      "epoch": 0.9449888775326158,
      "flos": 14721256995840.0,
      "grad_norm": 2.296228228810353,
      "language_loss": 0.77255499,
      "learning_rate": 3.159866897169094e-08,
      "loss": 0.81188452,
      "num_input_tokens_seen": 169872200,
      "step": 7859,
      "time_per_iteration": 2.693107843399048
    },
    {
      "auxiliary_loss_clip": 0.01225529,
      "auxiliary_loss_mlp": 0.01026047,
      "balance_loss_clip": 1.0456233,
      "balance_loss_mlp": 1.01910281,
      "epoch": 0.945109120423255,
      "flos": 15447873219840.0,
      "grad_norm": 2.0775856967077826,
      "language_loss": 0.76278675,
      "learning_rate": 3.146089618339487e-08,
      "loss": 0.78530246,
      "num_input_tokens_seen": 169889055,
      "step": 7860,
      "time_per_iteration": 2.6582260131835938
    },
    {
      "auxiliary_loss_clip": 0.01325601,
      "auxiliary_loss_mlp": 0.01026294,
      "balance_loss_clip": 1.04599559,
      "balance_loss_mlp": 1.01881397,
      "epoch": 0.9452293633138941,
      "flos": 25448097029760.0,
      "grad_norm": 1.8952901052977706,
      "language_loss": 0.6800763,
      "learning_rate": 3.132342202091554e-08,
      "loss": 0.70359528,
      "num_input_tokens_seen": 169909280,
      "step": 7861,
      "time_per_iteration": 3.662045478820801
    },
    {
      "auxiliary_loss_clip": 0.0117579,
      "auxiliary_loss_mlp": 0.01027808,
      "balance_loss_clip": 1.04891908,
      "balance_loss_mlp": 1.02076912,
      "epoch": 0.9453496062045331,
      "flos": 21215342350080.0,
      "grad_norm": 2.1051376048129407,
      "language_loss": 0.67919689,
      "learning_rate": 3.1186246505107595e-08,
      "loss": 0.70123291,
      "num_input_tokens_seen": 169928420,
      "step": 7862,
      "time_per_iteration": 2.5810420513153076
    },
    {
      "auxiliary_loss_clip": 0.01219906,
      "auxiliary_loss_mlp": 0.01031612,
      "balance_loss_clip": 1.04835796,
      "balance_loss_mlp": 1.02459991,
      "epoch": 0.9454698490951723,
      "flos": 20010898477440.0,
      "grad_norm": 1.9332771472019947,
      "language_loss": 0.83665621,
      "learning_rate": 3.104936965678084e-08,
      "loss": 0.85917139,
      "num_input_tokens_seen": 169946750,
      "step": 7863,
      "time_per_iteration": 2.650239944458008
    },
    {
      "auxiliary_loss_clip": 0.01221377,
      "auxiliary_loss_mlp": 0.01026727,
      "balance_loss_clip": 1.04652071,
      "balance_loss_mlp": 1.02007556,
      "epoch": 0.9455900919858113,
      "flos": 21069652786560.0,
      "grad_norm": 2.0257643891388684,
      "language_loss": 0.81814027,
      "learning_rate": 3.091279149669956e-08,
      "loss": 0.84062123,
      "num_input_tokens_seen": 169965540,
      "step": 7864,
      "time_per_iteration": 2.6025078296661377
    },
    {
      "auxiliary_loss_clip": 0.01221968,
      "auxiliary_loss_mlp": 0.02564481,
      "balance_loss_clip": 1.04851389,
      "balance_loss_mlp": 0.99992371,
      "epoch": 0.9457103348764504,
      "flos": 20740854666240.0,
      "grad_norm": 1.9408365011303668,
      "language_loss": 0.73954463,
      "learning_rate": 3.0776512045581624e-08,
      "loss": 0.77740908,
      "num_input_tokens_seen": 169984330,
      "step": 7865,
      "time_per_iteration": 2.6383557319641113
    },
    {
      "auxiliary_loss_clip": 0.01268194,
      "auxiliary_loss_mlp": 0.01028721,
      "balance_loss_clip": 1.04575849,
      "balance_loss_mlp": 1.02084732,
      "epoch": 0.9458305777670896,
      "flos": 21428363957760.0,
      "grad_norm": 1.9249477057142563,
      "language_loss": 0.77646059,
      "learning_rate": 3.0640531324101384e-08,
      "loss": 0.79942977,
      "num_input_tokens_seen": 170002095,
      "step": 7866,
      "time_per_iteration": 2.6450350284576416
    },
    {
      "auxiliary_loss_clip": 0.01222164,
      "auxiliary_loss_mlp": 0.01030046,
      "balance_loss_clip": 1.05060375,
      "balance_loss_mlp": 1.02277124,
      "epoch": 0.9459508206577286,
      "flos": 20011185786240.0,
      "grad_norm": 1.818548218700609,
      "language_loss": 0.75988615,
      "learning_rate": 3.0504849352886554e-08,
      "loss": 0.78240824,
      "num_input_tokens_seen": 170020240,
      "step": 7867,
      "time_per_iteration": 2.662421226501465
    },
    {
      "auxiliary_loss_clip": 0.01223206,
      "auxiliary_loss_mlp": 0.01025007,
      "balance_loss_clip": 1.04812562,
      "balance_loss_mlp": 1.01808429,
      "epoch": 0.9460710635483677,
      "flos": 12166428291840.0,
      "grad_norm": 2.8537368757391532,
      "language_loss": 0.72059405,
      "learning_rate": 3.036946615252023e-08,
      "loss": 0.74307621,
      "num_input_tokens_seen": 170035770,
      "step": 7868,
      "time_per_iteration": 2.5725772380828857
    },
    {
      "auxiliary_loss_clip": 0.01177434,
      "auxiliary_loss_mlp": 0.01022352,
      "balance_loss_clip": 1.0472343,
      "balance_loss_mlp": 1.01544726,
      "epoch": 0.9461913064390068,
      "flos": 34276196229120.0,
      "grad_norm": 2.1846210336437215,
      "language_loss": 0.67007077,
      "learning_rate": 3.0234381743539984e-08,
      "loss": 0.69206864,
      "num_input_tokens_seen": 170053385,
      "step": 7869,
      "time_per_iteration": 2.758647918701172
    },
    {
      "auxiliary_loss_clip": 0.01277295,
      "auxiliary_loss_mlp": 0.01026253,
      "balance_loss_clip": 1.04565883,
      "balance_loss_mlp": 1.01904082,
      "epoch": 0.9463115493296459,
      "flos": 19463763536640.0,
      "grad_norm": 1.9934296615922622,
      "language_loss": 0.80276543,
      "learning_rate": 3.0099596146437863e-08,
      "loss": 0.8258009,
      "num_input_tokens_seen": 170070490,
      "step": 7870,
      "time_per_iteration": 2.586824655532837
    },
    {
      "auxiliary_loss_clip": 0.010596,
      "auxiliary_loss_mlp": 0.01001575,
      "balance_loss_clip": 1.00574529,
      "balance_loss_mlp": 1.00071716,
      "epoch": 0.946431792220285,
      "flos": 70570824387840.0,
      "grad_norm": 0.7795799464668808,
      "language_loss": 0.59991312,
      "learning_rate": 2.996510938166086e-08,
      "loss": 0.62052482,
      "num_input_tokens_seen": 170133465,
      "step": 7871,
      "time_per_iteration": 3.206479787826538
    },
    {
      "auxiliary_loss_clip": 0.0122036,
      "auxiliary_loss_mlp": 0.01024905,
      "balance_loss_clip": 1.04927099,
      "balance_loss_mlp": 1.01827717,
      "epoch": 0.9465520351109241,
      "flos": 18947906363520.0,
      "grad_norm": 3.36107724233894,
      "language_loss": 0.74096745,
      "learning_rate": 2.983092146960997e-08,
      "loss": 0.7634201,
      "num_input_tokens_seen": 170150810,
      "step": 7872,
      "time_per_iteration": 2.5966572761535645
    },
    {
      "auxiliary_loss_clip": 0.01274833,
      "auxiliary_loss_mlp": 0.01025068,
      "balance_loss_clip": 1.04297829,
      "balance_loss_mlp": 1.0180347,
      "epoch": 0.9466722780015632,
      "flos": 19135647774720.0,
      "grad_norm": 1.9938572455152703,
      "language_loss": 0.80164742,
      "learning_rate": 2.9697032430642256e-08,
      "loss": 0.82464635,
      "num_input_tokens_seen": 170169025,
      "step": 7873,
      "time_per_iteration": 2.6297965049743652
    },
    {
      "auxiliary_loss_clip": 0.01167886,
      "auxiliary_loss_mlp": 0.01022138,
      "balance_loss_clip": 1.04689872,
      "balance_loss_mlp": 1.01518822,
      "epoch": 0.9467925208922022,
      "flos": 17237912520960.0,
      "grad_norm": 3.267085958774322,
      "language_loss": 0.7383495,
      "learning_rate": 2.9563442285067906e-08,
      "loss": 0.76024973,
      "num_input_tokens_seen": 170186070,
      "step": 7874,
      "time_per_iteration": 2.6666948795318604
    },
    {
      "auxiliary_loss_clip": 0.01222639,
      "auxiliary_loss_mlp": 0.01035629,
      "balance_loss_clip": 1.04811096,
      "balance_loss_mlp": 1.02852404,
      "epoch": 0.9469127637828414,
      "flos": 29169016859520.0,
      "grad_norm": 1.7933177635014923,
      "language_loss": 0.80022699,
      "learning_rate": 2.943015105315294e-08,
      "loss": 0.8228097,
      "num_input_tokens_seen": 170206265,
      "step": 7875,
      "time_per_iteration": 2.6828339099884033
    },
    {
      "auxiliary_loss_clip": 0.01372469,
      "auxiliary_loss_mlp": 0.01027828,
      "balance_loss_clip": 1.04265201,
      "balance_loss_mlp": 1.02061582,
      "epoch": 0.9470330066734804,
      "flos": 26030460234240.0,
      "grad_norm": 2.725456587571462,
      "language_loss": 0.66334653,
      "learning_rate": 2.929715875511718e-08,
      "loss": 0.68734944,
      "num_input_tokens_seen": 170225300,
      "step": 7876,
      "time_per_iteration": 2.741497039794922
    },
    {
      "auxiliary_loss_clip": 0.01222608,
      "auxiliary_loss_mlp": 0.01023897,
      "balance_loss_clip": 1.04499495,
      "balance_loss_mlp": 1.01696455,
      "epoch": 0.9471532495641195,
      "flos": 23440906056960.0,
      "grad_norm": 2.110084012498291,
      "language_loss": 0.70110178,
      "learning_rate": 2.9164465411135375e-08,
      "loss": 0.72356689,
      "num_input_tokens_seen": 170245070,
      "step": 7877,
      "time_per_iteration": 3.585805654525757
    },
    {
      "auxiliary_loss_clip": 0.01221275,
      "auxiliary_loss_mlp": 0.01032661,
      "balance_loss_clip": 1.05041337,
      "balance_loss_mlp": 1.0254097,
      "epoch": 0.9472734924547586,
      "flos": 15815850099840.0,
      "grad_norm": 2.224166685672258,
      "language_loss": 0.81181228,
      "learning_rate": 2.9032071041337426e-08,
      "loss": 0.83435166,
      "num_input_tokens_seen": 170263305,
      "step": 7878,
      "time_per_iteration": 2.6328377723693848
    },
    {
      "auxiliary_loss_clip": 0.01264718,
      "auxiliary_loss_mlp": 0.01023312,
      "balance_loss_clip": 1.04361176,
      "balance_loss_mlp": 1.01685965,
      "epoch": 0.9473937353453977,
      "flos": 11181793697280.0,
      "grad_norm": 2.2409133339159846,
      "language_loss": 0.72939217,
      "learning_rate": 2.889997566580704e-08,
      "loss": 0.75227249,
      "num_input_tokens_seen": 170281460,
      "step": 7879,
      "time_per_iteration": 3.561384439468384
    },
    {
      "auxiliary_loss_clip": 0.01175068,
      "auxiliary_loss_mlp": 0.01028103,
      "balance_loss_clip": 1.04833913,
      "balance_loss_mlp": 1.0204761,
      "epoch": 0.9475139782360368,
      "flos": 25775530433280.0,
      "grad_norm": 1.8065299252391038,
      "language_loss": 0.70417809,
      "learning_rate": 2.8768179304583086e-08,
      "loss": 0.72620988,
      "num_input_tokens_seen": 170303515,
      "step": 7880,
      "time_per_iteration": 2.647545337677002
    },
    {
      "auxiliary_loss_clip": 0.01318601,
      "auxiliary_loss_mlp": 0.0102394,
      "balance_loss_clip": 1.04560256,
      "balance_loss_mlp": 1.01745796,
      "epoch": 0.9476342211266758,
      "flos": 22820046451200.0,
      "grad_norm": 1.8350941180393248,
      "language_loss": 0.73656905,
      "learning_rate": 2.8636681977659117e-08,
      "loss": 0.75999451,
      "num_input_tokens_seen": 170323165,
      "step": 7881,
      "time_per_iteration": 2.724052667617798
    },
    {
      "auxiliary_loss_clip": 0.01369096,
      "auxiliary_loss_mlp": 0.01028184,
      "balance_loss_clip": 1.04587412,
      "balance_loss_mlp": 1.02092671,
      "epoch": 0.947754464017315,
      "flos": 20193611984640.0,
      "grad_norm": 2.4209632629188653,
      "language_loss": 0.78264689,
      "learning_rate": 2.850548370498318e-08,
      "loss": 0.8066197,
      "num_input_tokens_seen": 170341005,
      "step": 7882,
      "time_per_iteration": 3.594080686569214
    },
    {
      "auxiliary_loss_clip": 0.01220236,
      "auxiliary_loss_mlp": 0.01022199,
      "balance_loss_clip": 1.04519057,
      "balance_loss_mlp": 1.01566362,
      "epoch": 0.9478747069079541,
      "flos": 24717925359360.0,
      "grad_norm": 1.71743451038497,
      "language_loss": 0.71584147,
      "learning_rate": 2.8374584506457798e-08,
      "loss": 0.73826575,
      "num_input_tokens_seen": 170362280,
      "step": 7883,
      "time_per_iteration": 2.675935745239258
    },
    {
      "auxiliary_loss_clip": 0.01270744,
      "auxiliary_loss_mlp": 0.01021027,
      "balance_loss_clip": 1.04866648,
      "balance_loss_mlp": 1.01424062,
      "epoch": 0.9479949497985931,
      "flos": 21361355136000.0,
      "grad_norm": 3.917139735036052,
      "language_loss": 0.67144263,
      "learning_rate": 2.824398440193998e-08,
      "loss": 0.69436026,
      "num_input_tokens_seen": 170381080,
      "step": 7884,
      "time_per_iteration": 2.614469051361084
    },
    {
      "auxiliary_loss_clip": 0.01371271,
      "auxiliary_loss_mlp": 0.01025114,
      "balance_loss_clip": 1.04217911,
      "balance_loss_mlp": 1.01728213,
      "epoch": 0.9481151926892323,
      "flos": 18148606968960.0,
      "grad_norm": 1.9554909252378858,
      "language_loss": 0.71505862,
      "learning_rate": 2.811368341124232e-08,
      "loss": 0.73902249,
      "num_input_tokens_seen": 170400150,
      "step": 7885,
      "time_per_iteration": 2.723264694213867
    },
    {
      "auxiliary_loss_clip": 0.01273199,
      "auxiliary_loss_mlp": 0.01025215,
      "balance_loss_clip": 1.04563808,
      "balance_loss_mlp": 1.01862597,
      "epoch": 0.9482354355798713,
      "flos": 22128012046080.0,
      "grad_norm": 5.803291532243222,
      "language_loss": 0.68115336,
      "learning_rate": 2.7983681554131222e-08,
      "loss": 0.70413744,
      "num_input_tokens_seen": 170420410,
      "step": 7886,
      "time_per_iteration": 2.6504855155944824
    },
    {
      "auxiliary_loss_clip": 0.01274697,
      "auxiliary_loss_mlp": 0.01022602,
      "balance_loss_clip": 1.0451535,
      "balance_loss_mlp": 1.01576829,
      "epoch": 0.9483556784705104,
      "flos": 19063072344960.0,
      "grad_norm": 2.3108124162063635,
      "language_loss": 0.70401692,
      "learning_rate": 2.7853978850327365e-08,
      "loss": 0.72698998,
      "num_input_tokens_seen": 170439580,
      "step": 7887,
      "time_per_iteration": 3.6407430171966553
    },
    {
      "auxiliary_loss_clip": 0.01319097,
      "auxiliary_loss_mlp": 0.01024139,
      "balance_loss_clip": 1.04932737,
      "balance_loss_mlp": 1.01741552,
      "epoch": 0.9484759213611496,
      "flos": 25777110631680.0,
      "grad_norm": 2.155835176809916,
      "language_loss": 0.87402797,
      "learning_rate": 2.7724575319507225e-08,
      "loss": 0.89746028,
      "num_input_tokens_seen": 170459290,
      "step": 7888,
      "time_per_iteration": 2.8227415084838867
    },
    {
      "auxiliary_loss_clip": 0.01220653,
      "auxiliary_loss_mlp": 0.010182,
      "balance_loss_clip": 1.04421604,
      "balance_loss_mlp": 1.011235,
      "epoch": 0.9485961642517886,
      "flos": 20667740532480.0,
      "grad_norm": 1.9814881923355818,
      "language_loss": 0.77128732,
      "learning_rate": 2.759547098130044e-08,
      "loss": 0.79367578,
      "num_input_tokens_seen": 170478020,
      "step": 7889,
      "time_per_iteration": 2.700204610824585
    },
    {
      "auxiliary_loss_clip": 0.0117132,
      "auxiliary_loss_mlp": 0.01023704,
      "balance_loss_clip": 1.04841793,
      "balance_loss_mlp": 1.01715398,
      "epoch": 0.9487164071424277,
      "flos": 22674069578880.0,
      "grad_norm": 2.1136480140116727,
      "language_loss": 0.7673136,
      "learning_rate": 2.746666585529267e-08,
      "loss": 0.78926384,
      "num_input_tokens_seen": 170498295,
      "step": 7890,
      "time_per_iteration": 2.636631965637207
    },
    {
      "auxiliary_loss_clip": 0.01218355,
      "auxiliary_loss_mlp": 0.01029091,
      "balance_loss_clip": 1.04580283,
      "balance_loss_mlp": 1.02177143,
      "epoch": 0.9488366500330668,
      "flos": 38726461716480.0,
      "grad_norm": 2.7798965759248175,
      "language_loss": 0.74650878,
      "learning_rate": 2.73381599610234e-08,
      "loss": 0.76898324,
      "num_input_tokens_seen": 170518695,
      "step": 7891,
      "time_per_iteration": 2.721801519393921
    },
    {
      "auxiliary_loss_clip": 0.01217826,
      "auxiliary_loss_mlp": 0.01025311,
      "balance_loss_clip": 1.04457283,
      "balance_loss_mlp": 1.01848078,
      "epoch": 0.9489568929237059,
      "flos": 27890920149120.0,
      "grad_norm": 2.035437326630357,
      "language_loss": 0.7185967,
      "learning_rate": 2.7209953317987033e-08,
      "loss": 0.74102813,
      "num_input_tokens_seen": 170539735,
      "step": 7892,
      "time_per_iteration": 2.682130813598633
    },
    {
      "auxiliary_loss_clip": 0.01224587,
      "auxiliary_loss_mlp": 0.01023583,
      "balance_loss_clip": 1.04892945,
      "balance_loss_mlp": 1.016729,
      "epoch": 0.9490771358143449,
      "flos": 33580642291200.0,
      "grad_norm": 2.3903012948024975,
      "language_loss": 0.78453505,
      "learning_rate": 2.7082045945631793e-08,
      "loss": 0.80701679,
      "num_input_tokens_seen": 170561950,
      "step": 7893,
      "time_per_iteration": 2.7091119289398193
    },
    {
      "auxiliary_loss_clip": 0.01311631,
      "auxiliary_loss_mlp": 0.01025593,
      "balance_loss_clip": 1.04422057,
      "balance_loss_mlp": 1.01870584,
      "epoch": 0.9491973787049841,
      "flos": 14793796512000.0,
      "grad_norm": 2.10284918361491,
      "language_loss": 0.69594646,
      "learning_rate": 2.6954437863361712e-08,
      "loss": 0.71931869,
      "num_input_tokens_seen": 170579865,
      "step": 7894,
      "time_per_iteration": 2.674400806427002
    },
    {
      "auxiliary_loss_clip": 0.01421787,
      "auxiliary_loss_mlp": 0.01025581,
      "balance_loss_clip": 1.04365492,
      "balance_loss_mlp": 1.01903903,
      "epoch": 0.9493176215956232,
      "flos": 25332535998720.0,
      "grad_norm": 2.164929149028275,
      "language_loss": 0.70719212,
      "learning_rate": 2.6827129090534862e-08,
      "loss": 0.73166579,
      "num_input_tokens_seen": 170600165,
      "step": 7895,
      "time_per_iteration": 2.807718276977539
    },
    {
      "auxiliary_loss_clip": 0.01275286,
      "auxiliary_loss_mlp": 0.01023531,
      "balance_loss_clip": 1.04964709,
      "balance_loss_mlp": 1.0167743,
      "epoch": 0.9494378644862622,
      "flos": 21029971236480.0,
      "grad_norm": 1.9203535620428298,
      "language_loss": 0.77826798,
      "learning_rate": 2.670011964646335e-08,
      "loss": 0.80125618,
      "num_input_tokens_seen": 170618845,
      "step": 7896,
      "time_per_iteration": 2.679957389831543
    },
    {
      "auxiliary_loss_clip": 0.01477654,
      "auxiliary_loss_mlp": 0.01030384,
      "balance_loss_clip": 1.0352515,
      "balance_loss_mlp": 1.02293372,
      "epoch": 0.9495581073769014,
      "flos": 15195134148480.0,
      "grad_norm": 2.5846020874473465,
      "language_loss": 0.68311596,
      "learning_rate": 2.657340955041487e-08,
      "loss": 0.7081964,
      "num_input_tokens_seen": 170637620,
      "step": 7897,
      "time_per_iteration": 2.8025341033935547
    },
    {
      "auxiliary_loss_clip": 0.01277532,
      "auxiliary_loss_mlp": 0.01026753,
      "balance_loss_clip": 1.05245245,
      "balance_loss_mlp": 1.01939142,
      "epoch": 0.9496783502675404,
      "flos": 28616566705920.0,
      "grad_norm": 1.9623904334750848,
      "language_loss": 0.71867549,
      "learning_rate": 2.6446998821611167e-08,
      "loss": 0.74171841,
      "num_input_tokens_seen": 170657815,
      "step": 7898,
      "time_per_iteration": 2.7200379371643066
    },
    {
      "auxiliary_loss_clip": 0.01372171,
      "auxiliary_loss_mlp": 0.01029088,
      "balance_loss_clip": 1.04316378,
      "balance_loss_mlp": 1.02160764,
      "epoch": 0.9497985931581795,
      "flos": 14866874732160.0,
      "grad_norm": 18.73521421515372,
      "language_loss": 0.71612591,
      "learning_rate": 2.6320887479228228e-08,
      "loss": 0.74013853,
      "num_input_tokens_seen": 170674415,
      "step": 7899,
      "time_per_iteration": 2.73938250541687
    },
    {
      "auxiliary_loss_clip": 0.01273674,
      "auxiliary_loss_mlp": 0.01024803,
      "balance_loss_clip": 1.04567158,
      "balance_loss_mlp": 1.01813293,
      "epoch": 0.9499188360488187,
      "flos": 27193319136000.0,
      "grad_norm": 2.293531199432286,
      "language_loss": 0.72567511,
      "learning_rate": 2.619507554239786e-08,
      "loss": 0.74865985,
      "num_input_tokens_seen": 170692975,
      "step": 7900,
      "time_per_iteration": 2.693652391433716
    },
    {
      "auxiliary_loss_clip": 0.01275535,
      "auxiliary_loss_mlp": 0.0102618,
      "balance_loss_clip": 1.04799962,
      "balance_loss_mlp": 1.01905131,
      "epoch": 0.9500390789394577,
      "flos": 24316479982080.0,
      "grad_norm": 1.6069591778442895,
      "language_loss": 0.69780928,
      "learning_rate": 2.606956303020502e-08,
      "loss": 0.72082651,
      "num_input_tokens_seen": 170713780,
      "step": 7901,
      "time_per_iteration": 2.698528528213501
    },
    {
      "auxiliary_loss_clip": 0.01222702,
      "auxiliary_loss_mlp": 0.01027249,
      "balance_loss_clip": 1.04958248,
      "balance_loss_mlp": 1.02072251,
      "epoch": 0.9501593218300968,
      "flos": 14354752573440.0,
      "grad_norm": 4.0684661698936075,
      "language_loss": 0.84074163,
      "learning_rate": 2.5944349961690036e-08,
      "loss": 0.86324114,
      "num_input_tokens_seen": 170730800,
      "step": 7902,
      "time_per_iteration": 2.7622976303100586
    },
    {
      "auxiliary_loss_clip": 0.0131949,
      "auxiliary_loss_mlp": 0.01024244,
      "balance_loss_clip": 1.04549432,
      "balance_loss_mlp": 1.01766336,
      "epoch": 0.9502795647207359,
      "flos": 38728113742080.0,
      "grad_norm": 1.7480365927912171,
      "language_loss": 0.73392051,
      "learning_rate": 2.581943635584749e-08,
      "loss": 0.75735784,
      "num_input_tokens_seen": 170753630,
      "step": 7903,
      "time_per_iteration": 3.7062292098999023
    },
    {
      "auxiliary_loss_clip": 0.01262777,
      "auxiliary_loss_mlp": 0.01022271,
      "balance_loss_clip": 1.04347229,
      "balance_loss_mlp": 1.01587248,
      "epoch": 0.950399807611375,
      "flos": 40808023799040.0,
      "grad_norm": 2.9015903742344076,
      "language_loss": 0.65552723,
      "learning_rate": 2.569482223162689e-08,
      "loss": 0.67837763,
      "num_input_tokens_seen": 170777605,
      "step": 7904,
      "time_per_iteration": 2.828725814819336
    },
    {
      "auxiliary_loss_clip": 0.01222535,
      "auxiliary_loss_mlp": 0.0102241,
      "balance_loss_clip": 1.04670858,
      "balance_loss_mlp": 1.01538837,
      "epoch": 0.950520050502014,
      "flos": 23440403266560.0,
      "grad_norm": 1.856456565987711,
      "language_loss": 0.72384095,
      "learning_rate": 2.5570507607932e-08,
      "loss": 0.74629045,
      "num_input_tokens_seen": 170797520,
      "step": 7905,
      "time_per_iteration": 3.6076018810272217
    },
    {
      "auxiliary_loss_clip": 0.01227632,
      "auxiliary_loss_mlp": 0.01026788,
      "balance_loss_clip": 1.04766762,
      "balance_loss_mlp": 1.01945925,
      "epoch": 0.9506402933926532,
      "flos": 17783718658560.0,
      "grad_norm": 3.4274385944112695,
      "language_loss": 0.63471371,
      "learning_rate": 2.54464925036213e-08,
      "loss": 0.65725791,
      "num_input_tokens_seen": 170814810,
      "step": 7906,
      "time_per_iteration": 2.607809066772461
    },
    {
      "auxiliary_loss_clip": 0.01219937,
      "auxiliary_loss_mlp": 0.0102263,
      "balance_loss_clip": 1.04682517,
      "balance_loss_mlp": 1.01628482,
      "epoch": 0.9507605362832923,
      "flos": 32561928668160.0,
      "grad_norm": 1.8277544952463367,
      "language_loss": 0.60820353,
      "learning_rate": 2.532277693750773e-08,
      "loss": 0.63062912,
      "num_input_tokens_seen": 170835735,
      "step": 7907,
      "time_per_iteration": 3.628143787384033
    },
    {
      "auxiliary_loss_clip": 0.01374904,
      "auxiliary_loss_mlp": 0.01027085,
      "balance_loss_clip": 1.04717755,
      "balance_loss_mlp": 1.02037978,
      "epoch": 0.9508807791739313,
      "flos": 19602054898560.0,
      "grad_norm": 2.71923818864093,
      "language_loss": 0.7572515,
      "learning_rate": 2.5199360928358948e-08,
      "loss": 0.78127134,
      "num_input_tokens_seen": 170852970,
      "step": 7908,
      "time_per_iteration": 2.6756675243377686
    },
    {
      "auxiliary_loss_clip": 0.01215721,
      "auxiliary_loss_mlp": 0.02561633,
      "balance_loss_clip": 1.04400086,
      "balance_loss_mlp": 0.99991447,
      "epoch": 0.9510010220645704,
      "flos": 21471852349440.0,
      "grad_norm": 1.7698988133398943,
      "language_loss": 0.87494451,
      "learning_rate": 2.507624449489665e-08,
      "loss": 0.91271806,
      "num_input_tokens_seen": 170871600,
      "step": 7909,
      "time_per_iteration": 2.6500160694122314
    },
    {
      "auxiliary_loss_clip": 0.0126941,
      "auxiliary_loss_mlp": 0.01025027,
      "balance_loss_clip": 1.04702377,
      "balance_loss_mlp": 1.01730216,
      "epoch": 0.9511212649552095,
      "flos": 18879999701760.0,
      "grad_norm": 1.7743467113676687,
      "language_loss": 0.65046,
      "learning_rate": 2.495342765579811e-08,
      "loss": 0.67340446,
      "num_input_tokens_seen": 170890260,
      "step": 7910,
      "time_per_iteration": 2.6095948219299316
    },
    {
      "auxiliary_loss_clip": 0.01371001,
      "auxiliary_loss_mlp": 0.01029789,
      "balance_loss_clip": 1.04603624,
      "balance_loss_mlp": 1.02281773,
      "epoch": 0.9512415078458486,
      "flos": 20810521094400.0,
      "grad_norm": 7.308141005900582,
      "language_loss": 0.71351987,
      "learning_rate": 2.4830910429693984e-08,
      "loss": 0.73752773,
      "num_input_tokens_seen": 170910220,
      "step": 7911,
      "time_per_iteration": 2.7559518814086914
    },
    {
      "auxiliary_loss_clip": 0.01170913,
      "auxiliary_loss_mlp": 0.01021878,
      "balance_loss_clip": 1.04618669,
      "balance_loss_mlp": 1.01497269,
      "epoch": 0.9513617507364877,
      "flos": 18369565482240.0,
      "grad_norm": 1.9262877184407075,
      "language_loss": 0.79396713,
      "learning_rate": 2.470869283517052e-08,
      "loss": 0.81589508,
      "num_input_tokens_seen": 170928255,
      "step": 7912,
      "time_per_iteration": 3.444391965866089
    },
    {
      "auxiliary_loss_clip": 0.01215329,
      "auxiliary_loss_mlp": 0.01027571,
      "balance_loss_clip": 1.04349864,
      "balance_loss_mlp": 1.02096641,
      "epoch": 0.9514819936271268,
      "flos": 25010166412800.0,
      "grad_norm": 1.6213556531067619,
      "language_loss": 0.77215266,
      "learning_rate": 2.458677489076777e-08,
      "loss": 0.79458165,
      "num_input_tokens_seen": 170949265,
      "step": 7913,
      "time_per_iteration": 2.641322612762451
    },
    {
      "auxiliary_loss_clip": 0.01216199,
      "auxiliary_loss_mlp": 0.01020546,
      "balance_loss_clip": 1.04479682,
      "balance_loss_mlp": 1.01373959,
      "epoch": 0.9516022365177659,
      "flos": 18662129758080.0,
      "grad_norm": 1.6554421180715795,
      "language_loss": 0.82773453,
      "learning_rate": 2.446515661498072e-08,
      "loss": 0.85010195,
      "num_input_tokens_seen": 170968595,
      "step": 7914,
      "time_per_iteration": 2.5824618339538574
    },
    {
      "auxiliary_loss_clip": 0.01425367,
      "auxiliary_loss_mlp": 0.01023413,
      "balance_loss_clip": 1.04477775,
      "balance_loss_mlp": 1.01684737,
      "epoch": 0.9517224794084049,
      "flos": 25372109808000.0,
      "grad_norm": 2.1672663900505538,
      "language_loss": 0.74549663,
      "learning_rate": 2.434383802625861e-08,
      "loss": 0.76998436,
      "num_input_tokens_seen": 170987550,
      "step": 7915,
      "time_per_iteration": 2.825965404510498
    },
    {
      "auxiliary_loss_clip": 0.01321283,
      "auxiliary_loss_mlp": 0.01027456,
      "balance_loss_clip": 1.04230082,
      "balance_loss_mlp": 1.02080762,
      "epoch": 0.9518427222990441,
      "flos": 21470918595840.0,
      "grad_norm": 15.44718959070033,
      "language_loss": 0.74151647,
      "learning_rate": 2.4222819143005168e-08,
      "loss": 0.76500386,
      "num_input_tokens_seen": 171007145,
      "step": 7916,
      "time_per_iteration": 2.6974971294403076
    },
    {
      "auxiliary_loss_clip": 0.01171261,
      "auxiliary_loss_mlp": 0.01026512,
      "balance_loss_clip": 1.05067003,
      "balance_loss_mlp": 1.01958013,
      "epoch": 0.9519629651896832,
      "flos": 21033634423680.0,
      "grad_norm": 1.966315916075314,
      "language_loss": 0.80973279,
      "learning_rate": 2.4102099983579706e-08,
      "loss": 0.83171052,
      "num_input_tokens_seen": 171026295,
      "step": 7917,
      "time_per_iteration": 2.552006244659424
    },
    {
      "auxiliary_loss_clip": 0.01223016,
      "auxiliary_loss_mlp": 0.01022263,
      "balance_loss_clip": 1.04622722,
      "balance_loss_mlp": 1.01490748,
      "epoch": 0.9520832080803222,
      "flos": 21689219502720.0,
      "grad_norm": 1.6309664772230155,
      "language_loss": 0.77325058,
      "learning_rate": 2.3981680566294236e-08,
      "loss": 0.79570335,
      "num_input_tokens_seen": 171045895,
      "step": 7918,
      "time_per_iteration": 2.630385398864746
    },
    {
      "auxiliary_loss_clip": 0.01170223,
      "auxiliary_loss_mlp": 0.01024471,
      "balance_loss_clip": 1.04863572,
      "balance_loss_mlp": 1.01794159,
      "epoch": 0.9522034509709614,
      "flos": 23145289125120.0,
      "grad_norm": 1.7914071044664213,
      "language_loss": 0.73269314,
      "learning_rate": 2.3861560909416822e-08,
      "loss": 0.7546401,
      "num_input_tokens_seen": 171065445,
      "step": 7919,
      "time_per_iteration": 2.59612774848938
    },
    {
      "auxiliary_loss_clip": 0.01272861,
      "auxiliary_loss_mlp": 0.01024652,
      "balance_loss_clip": 1.04447818,
      "balance_loss_mlp": 1.01809502,
      "epoch": 0.9523236938616004,
      "flos": 24679428958080.0,
      "grad_norm": 3.1084448742888875,
      "language_loss": 0.82521987,
      "learning_rate": 2.3741741031169325e-08,
      "loss": 0.84819496,
      "num_input_tokens_seen": 171085015,
      "step": 7920,
      "time_per_iteration": 2.7746214866638184
    },
    {
      "auxiliary_loss_clip": 0.01362921,
      "auxiliary_loss_mlp": 0.0102537,
      "balance_loss_clip": 1.04054141,
      "balance_loss_mlp": 1.01882195,
      "epoch": 0.9524439367522395,
      "flos": 22672309812480.0,
      "grad_norm": 1.858111912452729,
      "language_loss": 0.71091688,
      "learning_rate": 2.3622220949728544e-08,
      "loss": 0.73479974,
      "num_input_tokens_seen": 171103900,
      "step": 7921,
      "time_per_iteration": 2.7585113048553467
    },
    {
      "auxiliary_loss_clip": 0.01218101,
      "auxiliary_loss_mlp": 0.01025225,
      "balance_loss_clip": 1.04448175,
      "balance_loss_mlp": 1.01817942,
      "epoch": 0.9525641796428787,
      "flos": 34055525024640.0,
      "grad_norm": 3.229541052412379,
      "language_loss": 0.61558485,
      "learning_rate": 2.3503000683225526e-08,
      "loss": 0.63801813,
      "num_input_tokens_seen": 171121615,
      "step": 7922,
      "time_per_iteration": 2.7931573390960693
    },
    {
      "auxiliary_loss_clip": 0.01170669,
      "auxiliary_loss_mlp": 0.01025463,
      "balance_loss_clip": 1.04665995,
      "balance_loss_mlp": 1.01813149,
      "epoch": 0.9526844225335177,
      "flos": 16727083251840.0,
      "grad_norm": 2.3188323094512064,
      "language_loss": 0.84627229,
      "learning_rate": 2.3384080249745585e-08,
      "loss": 0.86823356,
      "num_input_tokens_seen": 171139505,
      "step": 7923,
      "time_per_iteration": 2.5449109077453613
    },
    {
      "auxiliary_loss_clip": 0.01375141,
      "auxiliary_loss_mlp": 0.01026528,
      "balance_loss_clip": 1.04254711,
      "balance_loss_mlp": 1.0195868,
      "epoch": 0.9528046654241568,
      "flos": 36939367330560.0,
      "grad_norm": 3.8208314481889767,
      "language_loss": 0.83132696,
      "learning_rate": 2.3265459667329178e-08,
      "loss": 0.85534358,
      "num_input_tokens_seen": 171158995,
      "step": 7924,
      "time_per_iteration": 2.8333001136779785
    },
    {
      "auxiliary_loss_clip": 0.01269711,
      "auxiliary_loss_mlp": 0.01024728,
      "balance_loss_clip": 1.04621911,
      "balance_loss_mlp": 1.01821661,
      "epoch": 0.9529249083147959,
      "flos": 18255010032000.0,
      "grad_norm": 2.070232717613571,
      "language_loss": 0.86406279,
      "learning_rate": 2.31471389539708e-08,
      "loss": 0.88700724,
      "num_input_tokens_seen": 171176120,
      "step": 7925,
      "time_per_iteration": 2.6046979427337646
    },
    {
      "auxiliary_loss_clip": 0.0122482,
      "auxiliary_loss_mlp": 0.02561092,
      "balance_loss_clip": 1.04939413,
      "balance_loss_mlp": 0.99989629,
      "epoch": 0.953045151205435,
      "flos": 28658438985600.0,
      "grad_norm": 2.2810645191404544,
      "language_loss": 0.72885001,
      "learning_rate": 2.3029118127619872e-08,
      "loss": 0.76670909,
      "num_input_tokens_seen": 171195835,
      "step": 7926,
      "time_per_iteration": 2.717371940612793
    },
    {
      "auxiliary_loss_clip": 0.01267306,
      "auxiliary_loss_mlp": 0.01024566,
      "balance_loss_clip": 1.04609084,
      "balance_loss_mlp": 1.01751482,
      "epoch": 0.953165394096074,
      "flos": 21835232288640.0,
      "grad_norm": 2.328240085517227,
      "language_loss": 0.86628079,
      "learning_rate": 2.2911397206179628e-08,
      "loss": 0.8891995,
      "num_input_tokens_seen": 171212585,
      "step": 7927,
      "time_per_iteration": 2.67322039604187
    },
    {
      "auxiliary_loss_clip": 0.01172927,
      "auxiliary_loss_mlp": 0.01024356,
      "balance_loss_clip": 1.0502454,
      "balance_loss_mlp": 1.01686084,
      "epoch": 0.9532856369867132,
      "flos": 19975059682560.0,
      "grad_norm": 2.6586155641863285,
      "language_loss": 0.62984848,
      "learning_rate": 2.279397620750845e-08,
      "loss": 0.65182132,
      "num_input_tokens_seen": 171231630,
      "step": 7928,
      "time_per_iteration": 3.5344791412353516
    },
    {
      "auxiliary_loss_clip": 0.01263898,
      "auxiliary_loss_mlp": 0.01023161,
      "balance_loss_clip": 1.04301143,
      "balance_loss_mlp": 1.01653314,
      "epoch": 0.9534058798773523,
      "flos": 15049588239360.0,
      "grad_norm": 2.7181276444157847,
      "language_loss": 0.78817028,
      "learning_rate": 2.2676855149419195e-08,
      "loss": 0.81104088,
      "num_input_tokens_seen": 171248800,
      "step": 7929,
      "time_per_iteration": 2.615699052810669
    },
    {
      "auxiliary_loss_clip": 0.01271609,
      "auxiliary_loss_mlp": 0.01026099,
      "balance_loss_clip": 1.05065012,
      "balance_loss_mlp": 1.0193578,
      "epoch": 0.9535261227679913,
      "flos": 17602800831360.0,
      "grad_norm": 2.4145679829898268,
      "language_loss": 0.7563861,
      "learning_rate": 2.2560034049678988e-08,
      "loss": 0.77936327,
      "num_input_tokens_seen": 171263150,
      "step": 7930,
      "time_per_iteration": 2.580643892288208
    },
    {
      "auxiliary_loss_clip": 0.01176472,
      "auxiliary_loss_mlp": 0.01024285,
      "balance_loss_clip": 1.05094743,
      "balance_loss_mlp": 1.01699781,
      "epoch": 0.9536463656586305,
      "flos": 23142954741120.0,
      "grad_norm": 1.8183434837482446,
      "language_loss": 0.75371516,
      "learning_rate": 2.2443512926008988e-08,
      "loss": 0.77572274,
      "num_input_tokens_seen": 171282480,
      "step": 7931,
      "time_per_iteration": 3.539654493331909
    },
    {
      "auxiliary_loss_clip": 0.01324487,
      "auxiliary_loss_mlp": 0.01022698,
      "balance_loss_clip": 1.04362142,
      "balance_loss_mlp": 1.01606715,
      "epoch": 0.9537666085492695,
      "flos": 18625033987200.0,
      "grad_norm": 2.7826062644086433,
      "language_loss": 0.69903219,
      "learning_rate": 2.2327291796085946e-08,
      "loss": 0.72250408,
      "num_input_tokens_seen": 171300840,
      "step": 7932,
      "time_per_iteration": 2.6827640533447266
    },
    {
      "auxiliary_loss_clip": 0.0117537,
      "auxiliary_loss_mlp": 0.01025378,
      "balance_loss_clip": 1.04964125,
      "balance_loss_mlp": 1.01810634,
      "epoch": 0.9538868514399086,
      "flos": 18989347680000.0,
      "grad_norm": 3.078576226361324,
      "language_loss": 0.7685715,
      "learning_rate": 2.2211370677540197e-08,
      "loss": 0.79057896,
      "num_input_tokens_seen": 171317365,
      "step": 7933,
      "time_per_iteration": 3.539621114730835
    },
    {
      "auxiliary_loss_clip": 0.01173992,
      "auxiliary_loss_mlp": 0.01027507,
      "balance_loss_clip": 1.04940724,
      "balance_loss_mlp": 1.02063775,
      "epoch": 0.9540070943305478,
      "flos": 16800556521600.0,
      "grad_norm": 2.3741341954905617,
      "language_loss": 0.78162867,
      "learning_rate": 2.2095749587957012e-08,
      "loss": 0.8036437,
      "num_input_tokens_seen": 171335270,
      "step": 7934,
      "time_per_iteration": 2.579008102416992
    },
    {
      "auxiliary_loss_clip": 0.01270751,
      "auxiliary_loss_mlp": 0.01024802,
      "balance_loss_clip": 1.04317546,
      "balance_loss_mlp": 1.01757789,
      "epoch": 0.9541273372211868,
      "flos": 20156911263360.0,
      "grad_norm": 2.004230762694471,
      "language_loss": 0.69621235,
      "learning_rate": 2.1980428544876138e-08,
      "loss": 0.71916795,
      "num_input_tokens_seen": 171353910,
      "step": 7935,
      "time_per_iteration": 2.618901252746582
    },
    {
      "auxiliary_loss_clip": 0.01370908,
      "auxiliary_loss_mlp": 0.01023657,
      "balance_loss_clip": 1.03809333,
      "balance_loss_mlp": 1.01668322,
      "epoch": 0.9542475801118259,
      "flos": 26725511381760.0,
      "grad_norm": 1.8025924355702376,
      "language_loss": 0.74391669,
      "learning_rate": 2.1865407565791584e-08,
      "loss": 0.76786232,
      "num_input_tokens_seen": 171375480,
      "step": 7936,
      "time_per_iteration": 2.7850942611694336
    },
    {
      "auxiliary_loss_clip": 0.0127219,
      "auxiliary_loss_mlp": 0.01022185,
      "balance_loss_clip": 1.04483354,
      "balance_loss_mlp": 1.01479101,
      "epoch": 0.954367823002465,
      "flos": 23330911633920.0,
      "grad_norm": 3.3679451855893126,
      "language_loss": 0.77284825,
      "learning_rate": 2.175068666815183e-08,
      "loss": 0.79579198,
      "num_input_tokens_seen": 171396320,
      "step": 7937,
      "time_per_iteration": 2.660050868988037
    },
    {
      "auxiliary_loss_clip": 0.01323569,
      "auxiliary_loss_mlp": 0.01030816,
      "balance_loss_clip": 1.04349697,
      "balance_loss_mlp": 1.02395594,
      "epoch": 0.9544880658931041,
      "flos": 14902713527040.0,
      "grad_norm": 2.828183917223483,
      "language_loss": 0.79406023,
      "learning_rate": 2.163626586935985e-08,
      "loss": 0.81760412,
      "num_input_tokens_seen": 171412860,
      "step": 7938,
      "time_per_iteration": 2.6824791431427
    },
    {
      "auxiliary_loss_clip": 0.01223201,
      "auxiliary_loss_mlp": 0.01024861,
      "balance_loss_clip": 1.04666686,
      "balance_loss_mlp": 1.0178659,
      "epoch": 0.9546083087837431,
      "flos": 29095902725760.0,
      "grad_norm": 3.2781531495277343,
      "language_loss": 0.62692094,
      "learning_rate": 2.1522145186773755e-08,
      "loss": 0.64940155,
      "num_input_tokens_seen": 171431780,
      "step": 7939,
      "time_per_iteration": 3.578819751739502
    },
    {
      "auxiliary_loss_clip": 0.0126757,
      "auxiliary_loss_mlp": 0.01028958,
      "balance_loss_clip": 1.04460263,
      "balance_loss_mlp": 1.02252078,
      "epoch": 0.9547285516743822,
      "flos": 21142335957120.0,
      "grad_norm": 1.8803747063053404,
      "language_loss": 0.85850632,
      "learning_rate": 2.140832463770481e-08,
      "loss": 0.88147157,
      "num_input_tokens_seen": 171450975,
      "step": 7940,
      "time_per_iteration": 2.6643025875091553
    },
    {
      "auxiliary_loss_clip": 0.01276131,
      "auxiliary_loss_mlp": 0.01023158,
      "balance_loss_clip": 1.04561913,
      "balance_loss_mlp": 1.01596689,
      "epoch": 0.9548487945650214,
      "flos": 27490157130240.0,
      "grad_norm": 2.4508610854659976,
      "language_loss": 0.75941831,
      "learning_rate": 2.129480423941987e-08,
      "loss": 0.78241122,
      "num_input_tokens_seen": 171467645,
      "step": 7941,
      "time_per_iteration": 2.754242181777954
    },
    {
      "auxiliary_loss_clip": 0.01172257,
      "auxiliary_loss_mlp": 0.0102419,
      "balance_loss_clip": 1.04590189,
      "balance_loss_mlp": 1.01758003,
      "epoch": 0.9549690374556604,
      "flos": 22273198819200.0,
      "grad_norm": 2.4006556428520827,
      "language_loss": 0.80193555,
      "learning_rate": 2.1181584009140052e-08,
      "loss": 0.8239001,
      "num_input_tokens_seen": 171487185,
      "step": 7942,
      "time_per_iteration": 2.6505181789398193
    },
    {
      "auxiliary_loss_clip": 0.01329253,
      "auxiliary_loss_mlp": 0.01025038,
      "balance_loss_clip": 1.04531503,
      "balance_loss_mlp": 1.01796031,
      "epoch": 0.9550892803462995,
      "flos": 17595294888960.0,
      "grad_norm": 2.1299278025103843,
      "language_loss": 0.83743197,
      "learning_rate": 2.10686639640405e-08,
      "loss": 0.86097491,
      "num_input_tokens_seen": 171501275,
      "step": 7943,
      "time_per_iteration": 2.7308623790740967
    },
    {
      "auxiliary_loss_clip": 0.01124114,
      "auxiliary_loss_mlp": 0.0102094,
      "balance_loss_clip": 1.047683,
      "balance_loss_mlp": 1.01443172,
      "epoch": 0.9552095232369386,
      "flos": 24353144789760.0,
      "grad_norm": 1.8452083359676565,
      "language_loss": 0.81117976,
      "learning_rate": 2.0956044121251294e-08,
      "loss": 0.83263028,
      "num_input_tokens_seen": 171520060,
      "step": 7944,
      "time_per_iteration": 2.6003708839416504
    },
    {
      "auxiliary_loss_clip": 0.01323037,
      "auxiliary_loss_mlp": 0.01024578,
      "balance_loss_clip": 1.04966831,
      "balance_loss_mlp": 1.01725304,
      "epoch": 0.9553297661275777,
      "flos": 22746860490240.0,
      "grad_norm": 2.0094512220159535,
      "language_loss": 0.80923295,
      "learning_rate": 2.084372449785654e-08,
      "loss": 0.83270907,
      "num_input_tokens_seen": 171539895,
      "step": 7945,
      "time_per_iteration": 2.7282297611236572
    },
    {
      "auxiliary_loss_clip": 0.01271871,
      "auxiliary_loss_mlp": 0.01023351,
      "balance_loss_clip": 1.04399586,
      "balance_loss_mlp": 1.0163058,
      "epoch": 0.9554500090182168,
      "flos": 15413866018560.0,
      "grad_norm": 1.7116244930347377,
      "language_loss": 0.68821818,
      "learning_rate": 2.0731705110895282e-08,
      "loss": 0.71117032,
      "num_input_tokens_seen": 171557385,
      "step": 7946,
      "time_per_iteration": 2.618002414703369
    },
    {
      "auxiliary_loss_clip": 0.01225884,
      "auxiliary_loss_mlp": 0.01029316,
      "balance_loss_clip": 1.04941797,
      "balance_loss_mlp": 1.02158809,
      "epoch": 0.9555702519088559,
      "flos": 23513517400320.0,
      "grad_norm": 1.9829492131995907,
      "language_loss": 0.86783206,
      "learning_rate": 2.0619985977360587e-08,
      "loss": 0.89038408,
      "num_input_tokens_seen": 171575705,
      "step": 7947,
      "time_per_iteration": 2.612980842590332
    },
    {
      "auxiliary_loss_clip": 0.0136928,
      "auxiliary_loss_mlp": 0.01023749,
      "balance_loss_clip": 1.03939402,
      "balance_loss_mlp": 1.01715946,
      "epoch": 0.955690494799495,
      "flos": 22962072827520.0,
      "grad_norm": 1.8278700797881262,
      "language_loss": 0.76770216,
      "learning_rate": 2.0508567114200237e-08,
      "loss": 0.79163241,
      "num_input_tokens_seen": 171595620,
      "step": 7948,
      "time_per_iteration": 2.8509252071380615
    },
    {
      "auxiliary_loss_clip": 0.01274131,
      "auxiliary_loss_mlp": 0.01021024,
      "balance_loss_clip": 1.0457741,
      "balance_loss_mlp": 1.01446486,
      "epoch": 0.955810737690134,
      "flos": 26031250333440.0,
      "grad_norm": 2.5841543967902316,
      "language_loss": 0.79247302,
      "learning_rate": 2.0397448538316485e-08,
      "loss": 0.81542456,
      "num_input_tokens_seen": 171616660,
      "step": 7949,
      "time_per_iteration": 2.7303056716918945
    },
    {
      "auxiliary_loss_clip": 0.01320368,
      "auxiliary_loss_mlp": 0.01027084,
      "balance_loss_clip": 1.04492891,
      "balance_loss_mlp": 1.02008617,
      "epoch": 0.9559309805807732,
      "flos": 20849951249280.0,
      "grad_norm": 2.0232244568218745,
      "language_loss": 0.66766703,
      "learning_rate": 2.028663026656563e-08,
      "loss": 0.69114155,
      "num_input_tokens_seen": 171635515,
      "step": 7950,
      "time_per_iteration": 2.6849939823150635
    },
    {
      "auxiliary_loss_clip": 0.01172579,
      "auxiliary_loss_mlp": 0.02565301,
      "balance_loss_clip": 1.04995632,
      "balance_loss_mlp": 0.9999162,
      "epoch": 0.9560512234714122,
      "flos": 21578219498880.0,
      "grad_norm": 1.9189074658067655,
      "language_loss": 0.71874201,
      "learning_rate": 2.0176112315758885e-08,
      "loss": 0.7561208,
      "num_input_tokens_seen": 171653305,
      "step": 7951,
      "time_per_iteration": 2.574220895767212
    },
    {
      "auxiliary_loss_clip": 0.01375177,
      "auxiliary_loss_mlp": 0.01025254,
      "balance_loss_clip": 1.04576087,
      "balance_loss_mlp": 1.01793408,
      "epoch": 0.9561714663620513,
      "flos": 17450144029440.0,
      "grad_norm": 4.5189985138413284,
      "language_loss": 0.69003671,
      "learning_rate": 2.0065894702661957e-08,
      "loss": 0.71404099,
      "num_input_tokens_seen": 171669980,
      "step": 7952,
      "time_per_iteration": 2.7664084434509277
    },
    {
      "auxiliary_loss_clip": 0.01317206,
      "auxiliary_loss_mlp": 0.02567385,
      "balance_loss_clip": 1.04134786,
      "balance_loss_mlp": 0.99991977,
      "epoch": 0.9562917092526905,
      "flos": 26098510550400.0,
      "grad_norm": 1.7798972250302554,
      "language_loss": 0.78303099,
      "learning_rate": 1.9955977443994577e-08,
      "loss": 0.82187694,
      "num_input_tokens_seen": 171689970,
      "step": 7953,
      "time_per_iteration": 2.7336721420288086
    },
    {
      "auxiliary_loss_clip": 0.01277228,
      "auxiliary_loss_mlp": 0.01029587,
      "balance_loss_clip": 1.04718661,
      "balance_loss_mlp": 1.02148032,
      "epoch": 0.9564119521433295,
      "flos": 24096742531200.0,
      "grad_norm": 2.5739729363265242,
      "language_loss": 0.62341881,
      "learning_rate": 1.9846360556430965e-08,
      "loss": 0.64648694,
      "num_input_tokens_seen": 171708270,
      "step": 7954,
      "time_per_iteration": 2.8120670318603516
    },
    {
      "auxiliary_loss_clip": 0.01172157,
      "auxiliary_loss_mlp": 0.0102351,
      "balance_loss_clip": 1.04915905,
      "balance_loss_mlp": 1.0168283,
      "epoch": 0.9565321950339686,
      "flos": 32008903896960.0,
      "grad_norm": 2.2150691759150583,
      "language_loss": 0.61562878,
      "learning_rate": 1.973704405660004e-08,
      "loss": 0.63758546,
      "num_input_tokens_seen": 171729385,
      "step": 7955,
      "time_per_iteration": 3.568995475769043
    },
    {
      "auxiliary_loss_clip": 0.01416375,
      "auxiliary_loss_mlp": 0.01027409,
      "balance_loss_clip": 1.04120994,
      "balance_loss_mlp": 1.02073884,
      "epoch": 0.9566524379246077,
      "flos": 23588642695680.0,
      "grad_norm": 1.9936397015207787,
      "language_loss": 0.78086323,
      "learning_rate": 1.9628027961085203e-08,
      "loss": 0.80530107,
      "num_input_tokens_seen": 171752615,
      "step": 7956,
      "time_per_iteration": 3.741464138031006
    },
    {
      "auxiliary_loss_clip": 0.01313068,
      "auxiliary_loss_mlp": 0.01025221,
      "balance_loss_clip": 1.03936601,
      "balance_loss_mlp": 1.01819658,
      "epoch": 0.9567726808152468,
      "flos": 38067716240640.0,
      "grad_norm": 1.9503820050021772,
      "language_loss": 0.8365128,
      "learning_rate": 1.9519312286423894e-08,
      "loss": 0.85989571,
      "num_input_tokens_seen": 171775810,
      "step": 7957,
      "time_per_iteration": 2.838115692138672
    },
    {
      "auxiliary_loss_clip": 0.01216621,
      "auxiliary_loss_mlp": 0.01023559,
      "balance_loss_clip": 1.04634511,
      "balance_loss_mlp": 1.01689875,
      "epoch": 0.9568929237058859,
      "flos": 22744059229440.0,
      "grad_norm": 1.6024066104172787,
      "language_loss": 0.77930462,
      "learning_rate": 1.9410897049108255e-08,
      "loss": 0.80170643,
      "num_input_tokens_seen": 171795090,
      "step": 7958,
      "time_per_iteration": 2.583789825439453
    },
    {
      "auxiliary_loss_clip": 0.01178101,
      "auxiliary_loss_mlp": 0.01029882,
      "balance_loss_clip": 1.0518465,
      "balance_loss_mlp": 1.02236545,
      "epoch": 0.957013166596525,
      "flos": 23841633162240.0,
      "grad_norm": 1.8004896964857606,
      "language_loss": 0.91047442,
      "learning_rate": 1.9302782265584905e-08,
      "loss": 0.93255419,
      "num_input_tokens_seen": 171815755,
      "step": 7959,
      "time_per_iteration": 3.5316457748413086
    },
    {
      "auxiliary_loss_clip": 0.01365997,
      "auxiliary_loss_mlp": 0.01027599,
      "balance_loss_clip": 1.04448295,
      "balance_loss_mlp": 1.02085435,
      "epoch": 0.9571334094871641,
      "flos": 17639286071040.0,
      "grad_norm": 2.776038993437981,
      "language_loss": 0.87115633,
      "learning_rate": 1.9194967952254282e-08,
      "loss": 0.89509225,
      "num_input_tokens_seen": 171834330,
      "step": 7960,
      "time_per_iteration": 2.677776336669922
    },
    {
      "auxiliary_loss_clip": 0.01220232,
      "auxiliary_loss_mlp": 0.01030935,
      "balance_loss_clip": 1.04889345,
      "balance_loss_mlp": 1.02444363,
      "epoch": 0.9572536523778031,
      "flos": 15369623441280.0,
      "grad_norm": 3.3833993752261358,
      "language_loss": 0.80833483,
      "learning_rate": 1.9087454125472635e-08,
      "loss": 0.83084649,
      "num_input_tokens_seen": 171848805,
      "step": 7961,
      "time_per_iteration": 2.5659666061401367
    },
    {
      "auxiliary_loss_clip": 0.01172724,
      "auxiliary_loss_mlp": 0.01027047,
      "balance_loss_clip": 1.0486362,
      "balance_loss_mlp": 1.02012646,
      "epoch": 0.9573738952684423,
      "flos": 24969838417920.0,
      "grad_norm": 1.8408963451253924,
      "language_loss": 0.78196877,
      "learning_rate": 1.8980240801548696e-08,
      "loss": 0.80396652,
      "num_input_tokens_seen": 171867995,
      "step": 7962,
      "time_per_iteration": 2.660219430923462
    },
    {
      "auxiliary_loss_clip": 0.01266735,
      "auxiliary_loss_mlp": 0.01030635,
      "balance_loss_clip": 1.04788876,
      "balance_loss_mlp": 1.02432585,
      "epoch": 0.9574941381590814,
      "flos": 25769461034880.0,
      "grad_norm": 1.963082046522566,
      "language_loss": 0.74171257,
      "learning_rate": 1.8873327996747458e-08,
      "loss": 0.76468635,
      "num_input_tokens_seen": 171886495,
      "step": 7963,
      "time_per_iteration": 2.686218023300171
    },
    {
      "auxiliary_loss_clip": 0.01224351,
      "auxiliary_loss_mlp": 0.01025823,
      "balance_loss_clip": 1.04609573,
      "balance_loss_mlp": 1.01889122,
      "epoch": 0.9576143810497204,
      "flos": 32307178435200.0,
      "grad_norm": 2.146396767234672,
      "language_loss": 0.65878403,
      "learning_rate": 1.8766715727287053e-08,
      "loss": 0.68128574,
      "num_input_tokens_seen": 171908200,
      "step": 7964,
      "time_per_iteration": 3.5874691009521484
    },
    {
      "auxiliary_loss_clip": 0.0122724,
      "auxiliary_loss_mlp": 0.0256887,
      "balance_loss_clip": 1.04852915,
      "balance_loss_mlp": 0.99992514,
      "epoch": 0.9577346239403596,
      "flos": 27745733376000.0,
      "grad_norm": 1.8152557456259737,
      "language_loss": 0.79574621,
      "learning_rate": 1.8660404009340546e-08,
      "loss": 0.83370721,
      "num_input_tokens_seen": 171928650,
      "step": 7965,
      "time_per_iteration": 2.658733367919922
    },
    {
      "auxiliary_loss_clip": 0.0111496,
      "auxiliary_loss_mlp": 0.01000115,
      "balance_loss_clip": 1.00538397,
      "balance_loss_mlp": 0.99931067,
      "epoch": 0.9578548668309986,
      "flos": 57468313710720.0,
      "grad_norm": 0.870600881988479,
      "language_loss": 0.59546304,
      "learning_rate": 1.8554392859035485e-08,
      "loss": 0.61661375,
      "num_input_tokens_seen": 171986400,
      "step": 7966,
      "time_per_iteration": 3.1773815155029297
    },
    {
      "auxiliary_loss_clip": 0.01469028,
      "auxiliary_loss_mlp": 0.01027913,
      "balance_loss_clip": 1.03913295,
      "balance_loss_mlp": 1.0210644,
      "epoch": 0.9579751097216377,
      "flos": 19756040503680.0,
      "grad_norm": 1.9166694192507059,
      "language_loss": 0.79193997,
      "learning_rate": 1.8448682292453444e-08,
      "loss": 0.81690937,
      "num_input_tokens_seen": 172005475,
      "step": 7967,
      "time_per_iteration": 2.807110548019409
    },
    {
      "auxiliary_loss_clip": 0.01172925,
      "auxiliary_loss_mlp": 0.01026089,
      "balance_loss_clip": 1.04909658,
      "balance_loss_mlp": 1.01927066,
      "epoch": 0.9580953526122769,
      "flos": 18041270152320.0,
      "grad_norm": 1.7620294599513378,
      "language_loss": 0.65997505,
      "learning_rate": 1.8343272325631154e-08,
      "loss": 0.68196511,
      "num_input_tokens_seen": 172024420,
      "step": 7968,
      "time_per_iteration": 2.565525770187378
    },
    {
      "auxiliary_loss_clip": 0.01475177,
      "auxiliary_loss_mlp": 0.02568875,
      "balance_loss_clip": 1.04333401,
      "balance_loss_mlp": 0.99990082,
      "epoch": 0.9582155955029159,
      "flos": 24270154416000.0,
      "grad_norm": 2.3615611582880778,
      "language_loss": 0.7788074,
      "learning_rate": 1.8238162974558492e-08,
      "loss": 0.81924796,
      "num_input_tokens_seen": 172038350,
      "step": 7969,
      "time_per_iteration": 2.828733205795288
    },
    {
      "auxiliary_loss_clip": 0.01267304,
      "auxiliary_loss_mlp": 0.0103061,
      "balance_loss_clip": 1.04611993,
      "balance_loss_mlp": 1.02402925,
      "epoch": 0.958335838393555,
      "flos": 22783309816320.0,
      "grad_norm": 2.6866496422184447,
      "language_loss": 0.74985814,
      "learning_rate": 1.8133354255181144e-08,
      "loss": 0.77283728,
      "num_input_tokens_seen": 172058665,
      "step": 7970,
      "time_per_iteration": 2.6623237133026123
    },
    {
      "auxiliary_loss_clip": 0.01221893,
      "auxiliary_loss_mlp": 0.01023272,
      "balance_loss_clip": 1.04614902,
      "balance_loss_mlp": 1.01633441,
      "epoch": 0.958456081284194,
      "flos": 16911484698240.0,
      "grad_norm": 1.7563807989600058,
      "language_loss": 0.74637592,
      "learning_rate": 1.802884618339795e-08,
      "loss": 0.76882762,
      "num_input_tokens_seen": 172077470,
      "step": 7971,
      "time_per_iteration": 2.6214537620544434
    },
    {
      "auxiliary_loss_clip": 0.01223678,
      "auxiliary_loss_mlp": 0.01026584,
      "balance_loss_clip": 1.05054355,
      "balance_loss_mlp": 1.01938367,
      "epoch": 0.9585763241748332,
      "flos": 19974951941760.0,
      "grad_norm": 3.1293283736446917,
      "language_loss": 0.80873692,
      "learning_rate": 1.7924638775062894e-08,
      "loss": 0.83123958,
      "num_input_tokens_seen": 172096590,
      "step": 7972,
      "time_per_iteration": 2.6076314449310303
    },
    {
      "auxiliary_loss_clip": 0.01311484,
      "auxiliary_loss_mlp": 0.01026442,
      "balance_loss_clip": 1.04500771,
      "balance_loss_mlp": 1.02021635,
      "epoch": 0.9586965670654722,
      "flos": 21395649646080.0,
      "grad_norm": 2.1519518299768423,
      "language_loss": 0.8161037,
      "learning_rate": 1.7820732045984444e-08,
      "loss": 0.83948296,
      "num_input_tokens_seen": 172116735,
      "step": 7973,
      "time_per_iteration": 2.7457802295684814
    },
    {
      "auxiliary_loss_clip": 0.01222438,
      "auxiliary_loss_mlp": 0.01023654,
      "balance_loss_clip": 1.04676139,
      "balance_loss_mlp": 1.01638556,
      "epoch": 0.9588168099561113,
      "flos": 21435115714560.0,
      "grad_norm": 2.244986768321738,
      "language_loss": 0.74209517,
      "learning_rate": 1.7717126011924655e-08,
      "loss": 0.76455611,
      "num_input_tokens_seen": 172138320,
      "step": 7974,
      "time_per_iteration": 2.6108946800231934
    },
    {
      "auxiliary_loss_clip": 0.01366353,
      "auxiliary_loss_mlp": 0.01025562,
      "balance_loss_clip": 1.03825676,
      "balance_loss_mlp": 1.018466,
      "epoch": 0.9589370528467505,
      "flos": 11763761852160.0,
      "grad_norm": 2.8869007192551703,
      "language_loss": 0.76417172,
      "learning_rate": 1.7613820688600957e-08,
      "loss": 0.78809083,
      "num_input_tokens_seen": 172154225,
      "step": 7975,
      "time_per_iteration": 2.7022299766540527
    },
    {
      "auxiliary_loss_clip": 0.01178862,
      "auxiliary_loss_mlp": 0.01031201,
      "balance_loss_clip": 1.04684937,
      "balance_loss_mlp": 1.02387571,
      "epoch": 0.9590572957373895,
      "flos": 23441516588160.0,
      "grad_norm": 1.9426772540590664,
      "language_loss": 0.78854054,
      "learning_rate": 1.7510816091684588e-08,
      "loss": 0.81064117,
      "num_input_tokens_seen": 172174150,
      "step": 7976,
      "time_per_iteration": 2.6314632892608643
    },
    {
      "auxiliary_loss_clip": 0.01271952,
      "auxiliary_loss_mlp": 0.01022258,
      "balance_loss_clip": 1.04812622,
      "balance_loss_mlp": 1.01522183,
      "epoch": 0.9591775386280286,
      "flos": 22528272274560.0,
      "grad_norm": 3.410570350806108,
      "language_loss": 0.78940141,
      "learning_rate": 1.740811223680083e-08,
      "loss": 0.81234348,
      "num_input_tokens_seen": 172191005,
      "step": 7977,
      "time_per_iteration": 2.6872122287750244
    },
    {
      "auxiliary_loss_clip": 0.01173475,
      "auxiliary_loss_mlp": 0.01027805,
      "balance_loss_clip": 1.04912591,
      "balance_loss_mlp": 1.02078021,
      "epoch": 0.9592977815186677,
      "flos": 18186959715840.0,
      "grad_norm": 3.138113284048057,
      "language_loss": 0.74312019,
      "learning_rate": 1.7305709139530334e-08,
      "loss": 0.76513296,
      "num_input_tokens_seen": 172209785,
      "step": 7978,
      "time_per_iteration": 2.613074779510498
    },
    {
      "auxiliary_loss_clip": 0.01219521,
      "auxiliary_loss_mlp": 0.01027225,
      "balance_loss_clip": 1.04529965,
      "balance_loss_mlp": 1.02054,
      "epoch": 0.9594180244093068,
      "flos": 16537797555840.0,
      "grad_norm": 2.3431478704037905,
      "language_loss": 0.74459267,
      "learning_rate": 1.7203606815407334e-08,
      "loss": 0.7670601,
      "num_input_tokens_seen": 172224380,
      "step": 7979,
      "time_per_iteration": 2.5804033279418945
    },
    {
      "auxiliary_loss_clip": 0.01278109,
      "auxiliary_loss_mlp": 0.01026058,
      "balance_loss_clip": 1.04837084,
      "balance_loss_mlp": 1.01824999,
      "epoch": 0.9595382672999458,
      "flos": 20554334317440.0,
      "grad_norm": 1.899751300715834,
      "language_loss": 0.79582465,
      "learning_rate": 1.7101805279920557e-08,
      "loss": 0.81886631,
      "num_input_tokens_seen": 172242540,
      "step": 7980,
      "time_per_iteration": 3.511331081390381
    },
    {
      "auxiliary_loss_clip": 0.01174965,
      "auxiliary_loss_mlp": 0.01028818,
      "balance_loss_clip": 1.05115008,
      "balance_loss_mlp": 1.021415,
      "epoch": 0.959658510190585,
      "flos": 22638266697600.0,
      "grad_norm": 2.4891130123881515,
      "language_loss": 0.81137204,
      "learning_rate": 1.7000304548513643e-08,
      "loss": 0.83340991,
      "num_input_tokens_seen": 172262645,
      "step": 7981,
      "time_per_iteration": 2.600139617919922
    },
    {
      "auxiliary_loss_clip": 0.01321368,
      "auxiliary_loss_mlp": 0.01028926,
      "balance_loss_clip": 1.04378831,
      "balance_loss_mlp": 1.02238762,
      "epoch": 0.9597787530812241,
      "flos": 19135252725120.0,
      "grad_norm": 2.7033997662228546,
      "language_loss": 0.82428539,
      "learning_rate": 1.6899104636583394e-08,
      "loss": 0.84778833,
      "num_input_tokens_seen": 172280695,
      "step": 7982,
      "time_per_iteration": 2.6944420337677
    },
    {
      "auxiliary_loss_clip": 0.0111595,
      "auxiliary_loss_mlp": 0.01003074,
      "balance_loss_clip": 1.0057627,
      "balance_loss_mlp": 1.00222778,
      "epoch": 0.9598989959718631,
      "flos": 60098124055680.0,
      "grad_norm": 0.7314755389977237,
      "language_loss": 0.61875826,
      "learning_rate": 1.6798205559482638e-08,
      "loss": 0.63994849,
      "num_input_tokens_seen": 172343075,
      "step": 7983,
      "time_per_iteration": 4.220173120498657
    },
    {
      "auxiliary_loss_clip": 0.01328621,
      "auxiliary_loss_mlp": 0.01025317,
      "balance_loss_clip": 1.04870319,
      "balance_loss_mlp": 1.01852489,
      "epoch": 0.9600192388625023,
      "flos": 20886795624960.0,
      "grad_norm": 1.8707217500142637,
      "language_loss": 0.76362097,
      "learning_rate": 1.669760733251713e-08,
      "loss": 0.78716034,
      "num_input_tokens_seen": 172361950,
      "step": 7984,
      "time_per_iteration": 2.712400436401367
    },
    {
      "auxiliary_loss_clip": 0.01321108,
      "auxiliary_loss_mlp": 0.01025966,
      "balance_loss_clip": 1.040887,
      "balance_loss_mlp": 1.01955605,
      "epoch": 0.9601394817531413,
      "flos": 20445740524800.0,
      "grad_norm": 1.8280571409262683,
      "language_loss": 0.82260835,
      "learning_rate": 1.659730997094755e-08,
      "loss": 0.84607911,
      "num_input_tokens_seen": 172380440,
      "step": 7985,
      "time_per_iteration": 3.6606149673461914
    },
    {
      "auxiliary_loss_clip": 0.01217083,
      "auxiliary_loss_mlp": 0.01022707,
      "balance_loss_clip": 1.04581881,
      "balance_loss_mlp": 1.01635575,
      "epoch": 0.9602597246437804,
      "flos": 21507152440320.0,
      "grad_norm": 2.467923902149448,
      "language_loss": 0.62149954,
      "learning_rate": 1.6497313489989283e-08,
      "loss": 0.64389741,
      "num_input_tokens_seen": 172400265,
      "step": 7986,
      "time_per_iteration": 2.617973566055298
    },
    {
      "auxiliary_loss_clip": 0.01372902,
      "auxiliary_loss_mlp": 0.01020924,
      "balance_loss_clip": 1.03690743,
      "balance_loss_mlp": 1.01411414,
      "epoch": 0.9603799675344196,
      "flos": 29935099152000.0,
      "grad_norm": 4.234284893997262,
      "language_loss": 0.70179027,
      "learning_rate": 1.639761790481131e-08,
      "loss": 0.72572851,
      "num_input_tokens_seen": 172421145,
      "step": 7987,
      "time_per_iteration": 2.77366042137146
    },
    {
      "auxiliary_loss_clip": 0.01123781,
      "auxiliary_loss_mlp": 0.0102747,
      "balance_loss_clip": 1.04923379,
      "balance_loss_mlp": 1.02073467,
      "epoch": 0.9605002104250586,
      "flos": 28001525103360.0,
      "grad_norm": 1.9775954121737436,
      "language_loss": 0.79359329,
      "learning_rate": 1.6298223230537754e-08,
      "loss": 0.81510574,
      "num_input_tokens_seen": 172438945,
      "step": 7988,
      "time_per_iteration": 2.6222193241119385
    },
    {
      "auxiliary_loss_clip": 0.01270103,
      "auxiliary_loss_mlp": 0.02568408,
      "balance_loss_clip": 1.04674983,
      "balance_loss_mlp": 0.99990034,
      "epoch": 0.9606204533156977,
      "flos": 35590490870400.0,
      "grad_norm": 2.148295110983144,
      "language_loss": 0.6946851,
      "learning_rate": 1.619912948224611e-08,
      "loss": 0.73307025,
      "num_input_tokens_seen": 172460150,
      "step": 7989,
      "time_per_iteration": 2.80619215965271
    },
    {
      "auxiliary_loss_clip": 0.01320112,
      "auxiliary_loss_mlp": 0.01025915,
      "balance_loss_clip": 1.04422688,
      "balance_loss_mlp": 1.01923048,
      "epoch": 0.9607406962063368,
      "flos": 26574614346240.0,
      "grad_norm": 2.906249277793569,
      "language_loss": 0.60873705,
      "learning_rate": 1.6100336674969682e-08,
      "loss": 0.63219726,
      "num_input_tokens_seen": 172478990,
      "step": 7990,
      "time_per_iteration": 3.648334264755249
    },
    {
      "auxiliary_loss_clip": 0.01272081,
      "auxiliary_loss_mlp": 0.01026188,
      "balance_loss_clip": 1.03987813,
      "balance_loss_mlp": 1.01944935,
      "epoch": 0.9608609390969759,
      "flos": 25331781813120.0,
      "grad_norm": 1.8832657676782472,
      "language_loss": 0.76713121,
      "learning_rate": 1.600184482369449e-08,
      "loss": 0.79011393,
      "num_input_tokens_seen": 172498905,
      "step": 7991,
      "time_per_iteration": 2.7324020862579346
    },
    {
      "auxiliary_loss_clip": 0.01322214,
      "auxiliary_loss_mlp": 0.01027494,
      "balance_loss_clip": 1.04382849,
      "balance_loss_mlp": 1.02037454,
      "epoch": 0.960981181987615,
      "flos": 21069114082560.0,
      "grad_norm": 2.5458002733173832,
      "language_loss": 0.89116263,
      "learning_rate": 1.5903653943362126e-08,
      "loss": 0.91465974,
      "num_input_tokens_seen": 172517900,
      "step": 7992,
      "time_per_iteration": 2.7397284507751465
    },
    {
      "auxiliary_loss_clip": 0.01273453,
      "auxiliary_loss_mlp": 0.010284,
      "balance_loss_clip": 1.04776049,
      "balance_loss_mlp": 1.02159894,
      "epoch": 0.9611014248782541,
      "flos": 17823256554240.0,
      "grad_norm": 2.231809182616771,
      "language_loss": 0.77018934,
      "learning_rate": 1.580576404886802e-08,
      "loss": 0.79320782,
      "num_input_tokens_seen": 172536430,
      "step": 7993,
      "time_per_iteration": 2.669149160385132
    },
    {
      "auxiliary_loss_clip": 0.01222896,
      "auxiliary_loss_mlp": 0.0102652,
      "balance_loss_clip": 1.047822,
      "balance_loss_mlp": 1.01978135,
      "epoch": 0.9612216677688932,
      "flos": 19354631040000.0,
      "grad_norm": 2.2713394546658985,
      "language_loss": 0.79782224,
      "learning_rate": 1.570817515506162e-08,
      "loss": 0.82031643,
      "num_input_tokens_seen": 172555120,
      "step": 7994,
      "time_per_iteration": 2.6409096717834473
    },
    {
      "auxiliary_loss_clip": 0.01172021,
      "auxiliary_loss_mlp": 0.01027268,
      "balance_loss_clip": 1.04945958,
      "balance_loss_mlp": 1.02120042,
      "epoch": 0.9613419106595322,
      "flos": 15808739207040.0,
      "grad_norm": 2.1518083322673114,
      "language_loss": 0.81457639,
      "learning_rate": 1.561088727674753e-08,
      "loss": 0.83656925,
      "num_input_tokens_seen": 172569330,
      "step": 7995,
      "time_per_iteration": 2.552351713180542
    },
    {
      "auxiliary_loss_clip": 0.01384623,
      "auxiliary_loss_mlp": 0.01028097,
      "balance_loss_clip": 1.04476237,
      "balance_loss_mlp": 1.02058411,
      "epoch": 0.9614621535501714,
      "flos": 25702488126720.0,
      "grad_norm": 2.604159877640764,
      "language_loss": 0.71319067,
      "learning_rate": 1.551390042868417e-08,
      "loss": 0.73731786,
      "num_input_tokens_seen": 172591100,
      "step": 7996,
      "time_per_iteration": 2.7985424995422363
    },
    {
      "auxiliary_loss_clip": 0.01224724,
      "auxiliary_loss_mlp": 0.01026884,
      "balance_loss_clip": 1.04973245,
      "balance_loss_mlp": 1.02047086,
      "epoch": 0.9615823964408104,
      "flos": 17819054663040.0,
      "grad_norm": 2.026286008614578,
      "language_loss": 0.71022904,
      "learning_rate": 1.5417214625584207e-08,
      "loss": 0.73274505,
      "num_input_tokens_seen": 172608755,
      "step": 7997,
      "time_per_iteration": 2.606011390686035
    },
    {
      "auxiliary_loss_clip": 0.01219157,
      "auxiliary_loss_mlp": 0.01031431,
      "balance_loss_clip": 1.04491603,
      "balance_loss_mlp": 1.02468967,
      "epoch": 0.9617026393314495,
      "flos": 20190020624640.0,
      "grad_norm": 1.6580307327368407,
      "language_loss": 0.85405791,
      "learning_rate": 1.5320829882114806e-08,
      "loss": 0.87656379,
      "num_input_tokens_seen": 172626830,
      "step": 7998,
      "time_per_iteration": 2.6332411766052246
    },
    {
      "auxiliary_loss_clip": 0.01171859,
      "auxiliary_loss_mlp": 0.01023888,
      "balance_loss_clip": 1.04655695,
      "balance_loss_mlp": 1.01694369,
      "epoch": 0.9618228822220887,
      "flos": 20267013427200.0,
      "grad_norm": 1.9960147379831428,
      "language_loss": 0.79073739,
      "learning_rate": 1.5224746212897378e-08,
      "loss": 0.81269485,
      "num_input_tokens_seen": 172646125,
      "step": 7999,
      "time_per_iteration": 2.5939931869506836
    },
    {
      "auxiliary_loss_clip": 0.01168634,
      "auxiliary_loss_mlp": 0.0102404,
      "balance_loss_clip": 1.04793453,
      "balance_loss_mlp": 1.01762629,
      "epoch": 0.9619431251127277,
      "flos": 21031300039680.0,
      "grad_norm": 1.9827202244288054,
      "language_loss": 0.77409637,
      "learning_rate": 1.512896363250804e-08,
      "loss": 0.79602307,
      "num_input_tokens_seen": 172666235,
      "step": 8000,
      "time_per_iteration": 2.635939121246338
    },
    {
      "auxiliary_loss_clip": 0.01222893,
      "auxiliary_loss_mlp": 0.01024553,
      "balance_loss_clip": 1.04572463,
      "balance_loss_mlp": 1.01771343,
      "epoch": 0.9620633680033668,
      "flos": 22382654538240.0,
      "grad_norm": 1.912805621488122,
      "language_loss": 0.76038957,
      "learning_rate": 1.503348215547673e-08,
      "loss": 0.78286409,
      "num_input_tokens_seen": 172687325,
      "step": 8001,
      "time_per_iteration": 2.6445906162261963
    },
    {
      "auxiliary_loss_clip": 0.01265153,
      "auxiliary_loss_mlp": 0.01021875,
      "balance_loss_clip": 1.04406977,
      "balance_loss_mlp": 1.01511288,
      "epoch": 0.962183610894006,
      "flos": 18471730740480.0,
      "grad_norm": 3.128259983882231,
      "language_loss": 0.80799884,
      "learning_rate": 1.4938301796288078e-08,
      "loss": 0.83086908,
      "num_input_tokens_seen": 172703895,
      "step": 8002,
      "time_per_iteration": 2.6102638244628906
    },
    {
      "auxiliary_loss_clip": 0.01174433,
      "auxiliary_loss_mlp": 0.01026238,
      "balance_loss_clip": 1.05022836,
      "balance_loss_mlp": 1.01937759,
      "epoch": 0.962303853784645,
      "flos": 18435245500800.0,
      "grad_norm": 4.5197700017498015,
      "language_loss": 0.81570733,
      "learning_rate": 1.4843422569380537e-08,
      "loss": 0.83771402,
      "num_input_tokens_seen": 172720650,
      "step": 8003,
      "time_per_iteration": 2.572871685028076
    },
    {
      "auxiliary_loss_clip": 0.0126705,
      "auxiliary_loss_mlp": 0.01024512,
      "balance_loss_clip": 1.0395236,
      "balance_loss_mlp": 1.01757371,
      "epoch": 0.9624240966752841,
      "flos": 26391074826240.0,
      "grad_norm": 1.9145786323281435,
      "language_loss": 0.82879144,
      "learning_rate": 1.4748844489147483e-08,
      "loss": 0.8517071,
      "num_input_tokens_seen": 172737640,
      "step": 8004,
      "time_per_iteration": 2.7164347171783447
    },
    {
      "auxiliary_loss_clip": 0.0126845,
      "auxiliary_loss_mlp": 0.01021568,
      "balance_loss_clip": 1.04453063,
      "balance_loss_mlp": 1.01500273,
      "epoch": 0.9625443395659231,
      "flos": 14647675985280.0,
      "grad_norm": 1.8409563010984247,
      "language_loss": 0.71444297,
      "learning_rate": 1.4654567569936326e-08,
      "loss": 0.73734307,
      "num_input_tokens_seen": 172755215,
      "step": 8005,
      "time_per_iteration": 2.6631295680999756
    },
    {
      "auxiliary_loss_clip": 0.01365608,
      "auxiliary_loss_mlp": 0.01025249,
      "balance_loss_clip": 1.04230797,
      "balance_loss_mlp": 1.01822424,
      "epoch": 0.9626645824565623,
      "flos": 18367626147840.0,
      "grad_norm": 2.5744475587937816,
      "language_loss": 0.82988107,
      "learning_rate": 1.456059182604874e-08,
      "loss": 0.85378969,
      "num_input_tokens_seen": 172774020,
      "step": 8006,
      "time_per_iteration": 3.638010025024414
    },
    {
      "auxiliary_loss_clip": 0.01173285,
      "auxiliary_loss_mlp": 0.01024243,
      "balance_loss_clip": 1.04921007,
      "balance_loss_mlp": 1.01707578,
      "epoch": 0.9627848253472013,
      "flos": 16580424021120.0,
      "grad_norm": 2.0750819151882096,
      "language_loss": 0.76562893,
      "learning_rate": 1.4466917271740653e-08,
      "loss": 0.78760421,
      "num_input_tokens_seen": 172792220,
      "step": 8007,
      "time_per_iteration": 2.692897081375122
    },
    {
      "auxiliary_loss_clip": 0.01270082,
      "auxiliary_loss_mlp": 0.01025778,
      "balance_loss_clip": 1.04458308,
      "balance_loss_mlp": 1.01754618,
      "epoch": 0.9629050682378404,
      "flos": 20886867452160.0,
      "grad_norm": 2.8519348277920207,
      "language_loss": 0.67852074,
      "learning_rate": 1.4373543921222697e-08,
      "loss": 0.70147932,
      "num_input_tokens_seen": 172811805,
      "step": 8008,
      "time_per_iteration": 2.6505165100097656
    },
    {
      "auxiliary_loss_clip": 0.01267027,
      "auxiliary_loss_mlp": 0.01023253,
      "balance_loss_clip": 1.04563951,
      "balance_loss_mlp": 1.01634443,
      "epoch": 0.9630253111284796,
      "flos": 17019252478080.0,
      "grad_norm": 1.9399620880679498,
      "language_loss": 0.77914882,
      "learning_rate": 1.428047178865932e-08,
      "loss": 0.8020516,
      "num_input_tokens_seen": 172828595,
      "step": 8009,
      "time_per_iteration": 3.584616184234619
    },
    {
      "auxiliary_loss_clip": 0.01273137,
      "auxiliary_loss_mlp": 0.01022775,
      "balance_loss_clip": 1.04342496,
      "balance_loss_mlp": 1.01581621,
      "epoch": 0.9631455540191186,
      "flos": 20338942412160.0,
      "grad_norm": 1.9380920742761807,
      "language_loss": 0.74800581,
      "learning_rate": 1.4187700888169451e-08,
      "loss": 0.77096498,
      "num_input_tokens_seen": 172847770,
      "step": 8010,
      "time_per_iteration": 2.695824384689331
    },
    {
      "auxiliary_loss_clip": 0.01109894,
      "auxiliary_loss_mlp": 0.01001452,
      "balance_loss_clip": 1.00577116,
      "balance_loss_mlp": 1.00062358,
      "epoch": 0.9632657969097577,
      "flos": 65956700033280.0,
      "grad_norm": 0.7580303290131469,
      "language_loss": 0.57006705,
      "learning_rate": 1.40952312338265e-08,
      "loss": 0.59118056,
      "num_input_tokens_seen": 172912415,
      "step": 8011,
      "time_per_iteration": 4.156641960144043
    },
    {
      "auxiliary_loss_clip": 0.01321676,
      "auxiliary_loss_mlp": 0.01026589,
      "balance_loss_clip": 1.0438062,
      "balance_loss_mlp": 1.02005076,
      "epoch": 0.9633860398003968,
      "flos": 44419523823360.0,
      "grad_norm": 2.4013239899865155,
      "language_loss": 0.68457973,
      "learning_rate": 1.4003062839657909e-08,
      "loss": 0.70806241,
      "num_input_tokens_seen": 172934895,
      "step": 8012,
      "time_per_iteration": 2.900498867034912
    },
    {
      "auxiliary_loss_clip": 0.0122222,
      "auxiliary_loss_mlp": 0.01022332,
      "balance_loss_clip": 1.04465151,
      "balance_loss_mlp": 1.01600158,
      "epoch": 0.9635062826910359,
      "flos": 24827704300800.0,
      "grad_norm": 1.7812788110584818,
      "language_loss": 0.80174255,
      "learning_rate": 1.391119571964583e-08,
      "loss": 0.82418811,
      "num_input_tokens_seen": 172955835,
      "step": 8013,
      "time_per_iteration": 2.730473518371582
    },
    {
      "auxiliary_loss_clip": 0.0121724,
      "auxiliary_loss_mlp": 0.01027246,
      "balance_loss_clip": 1.04601717,
      "balance_loss_mlp": 1.02059102,
      "epoch": 0.9636265255816749,
      "flos": 15961360095360.0,
      "grad_norm": 2.0479436791996877,
      "language_loss": 0.73074442,
      "learning_rate": 1.3819629887726225e-08,
      "loss": 0.75318933,
      "num_input_tokens_seen": 172973925,
      "step": 8014,
      "time_per_iteration": 2.6018919944763184
    },
    {
      "auxiliary_loss_clip": 0.0127583,
      "auxiliary_loss_mlp": 0.01028437,
      "balance_loss_clip": 1.04663873,
      "balance_loss_mlp": 1.02052426,
      "epoch": 0.9637467684723141,
      "flos": 22601781457920.0,
      "grad_norm": 1.8185009793684477,
      "language_loss": 0.76689059,
      "learning_rate": 1.3728365357789317e-08,
      "loss": 0.7899332,
      "num_input_tokens_seen": 172993290,
      "step": 8015,
      "time_per_iteration": 2.670301914215088
    },
    {
      "auxiliary_loss_clip": 0.01310307,
      "auxiliary_loss_mlp": 0.01025472,
      "balance_loss_clip": 1.03954709,
      "balance_loss_mlp": 1.01836979,
      "epoch": 0.9638670113629532,
      "flos": 17565812801280.0,
      "grad_norm": 2.731551973201352,
      "language_loss": 0.76460171,
      "learning_rate": 1.3637402143680254e-08,
      "loss": 0.78795946,
      "num_input_tokens_seen": 173008190,
      "step": 8016,
      "time_per_iteration": 3.6302926540374756
    },
    {
      "auxiliary_loss_clip": 0.01113605,
      "auxiliary_loss_mlp": 0.01002642,
      "balance_loss_clip": 1.00692463,
      "balance_loss_mlp": 1.00170028,
      "epoch": 0.9639872542535922,
      "flos": 55072139379840.0,
      "grad_norm": 0.7282129443940352,
      "language_loss": 0.55094671,
      "learning_rate": 1.3546740259197998e-08,
      "loss": 0.5721091,
      "num_input_tokens_seen": 173061000,
      "step": 8017,
      "time_per_iteration": 3.2555418014526367
    },
    {
      "auxiliary_loss_clip": 0.01271542,
      "auxiliary_loss_mlp": 0.0102826,
      "balance_loss_clip": 1.04508042,
      "balance_loss_mlp": 1.02081549,
      "epoch": 0.9641074971442314,
      "flos": 24134484746880.0,
      "grad_norm": 2.1284800870413965,
      "language_loss": 0.7004326,
      "learning_rate": 1.3456379718095989e-08,
      "loss": 0.72343063,
      "num_input_tokens_seen": 173081415,
      "step": 8018,
      "time_per_iteration": 2.709378957748413
    },
    {
      "auxiliary_loss_clip": 0.01166532,
      "auxiliary_loss_mlp": 0.01000207,
      "balance_loss_clip": 1.0063076,
      "balance_loss_mlp": 0.99924129,
      "epoch": 0.9642277400348704,
      "flos": 66747416077440.0,
      "grad_norm": 0.835289741210995,
      "language_loss": 0.61960506,
      "learning_rate": 1.3366320534081487e-08,
      "loss": 0.64127243,
      "num_input_tokens_seen": 173144095,
      "step": 8019,
      "time_per_iteration": 3.245222568511963
    },
    {
      "auxiliary_loss_clip": 0.01221218,
      "auxiliary_loss_mlp": 0.01023296,
      "balance_loss_clip": 1.04539585,
      "balance_loss_mlp": 1.01586628,
      "epoch": 0.9643479829255095,
      "flos": 30920272450560.0,
      "grad_norm": 2.6085242201752687,
      "language_loss": 0.75741947,
      "learning_rate": 1.3276562720816675e-08,
      "loss": 0.77986461,
      "num_input_tokens_seen": 173165605,
      "step": 8020,
      "time_per_iteration": 2.762338876724243
    },
    {
      "auxiliary_loss_clip": 0.0117211,
      "auxiliary_loss_mlp": 0.01023186,
      "balance_loss_clip": 1.04718339,
      "balance_loss_mlp": 1.01588476,
      "epoch": 0.9644682258161487,
      "flos": 20048245643520.0,
      "grad_norm": 2.691062703249326,
      "language_loss": 0.82724154,
      "learning_rate": 1.3187106291917549e-08,
      "loss": 0.84919447,
      "num_input_tokens_seen": 173182595,
      "step": 8021,
      "time_per_iteration": 2.6079702377319336
    },
    {
      "auxiliary_loss_clip": 0.01215886,
      "auxiliary_loss_mlp": 0.01026931,
      "balance_loss_clip": 1.0458076,
      "balance_loss_mlp": 1.02061629,
      "epoch": 0.9645884687067877,
      "flos": 21178713456000.0,
      "grad_norm": 1.8844740063849523,
      "language_loss": 0.70944238,
      "learning_rate": 1.309795126095503e-08,
      "loss": 0.73187047,
      "num_input_tokens_seen": 173200895,
      "step": 8022,
      "time_per_iteration": 2.6149537563323975
    },
    {
      "auxiliary_loss_clip": 0.01470972,
      "auxiliary_loss_mlp": 0.01021786,
      "balance_loss_clip": 1.04159212,
      "balance_loss_mlp": 1.01523864,
      "epoch": 0.9647087115974268,
      "flos": 18945967029120.0,
      "grad_norm": 2.803708175335097,
      "language_loss": 0.80613536,
      "learning_rate": 1.3009097641453192e-08,
      "loss": 0.83106291,
      "num_input_tokens_seen": 173218745,
      "step": 8023,
      "time_per_iteration": 2.7755377292633057
    },
    {
      "auxiliary_loss_clip": 0.01274529,
      "auxiliary_loss_mlp": 0.01027598,
      "balance_loss_clip": 1.04856539,
      "balance_loss_mlp": 1.02099681,
      "epoch": 0.9648289544880659,
      "flos": 16545088016640.0,
      "grad_norm": 1.9977007762472057,
      "language_loss": 0.7601856,
      "learning_rate": 1.2920545446891474e-08,
      "loss": 0.78320682,
      "num_input_tokens_seen": 173235465,
      "step": 8024,
      "time_per_iteration": 2.6641483306884766
    },
    {
      "auxiliary_loss_clip": 0.01271583,
      "auxiliary_loss_mlp": 0.01031997,
      "balance_loss_clip": 1.04633355,
      "balance_loss_mlp": 1.02496994,
      "epoch": 0.964949197378705,
      "flos": 24057527857920.0,
      "grad_norm": 1.8065379681806681,
      "language_loss": 0.70583916,
      "learning_rate": 1.2832294690703127e-08,
      "loss": 0.72887504,
      "num_input_tokens_seen": 173254440,
      "step": 8025,
      "time_per_iteration": 2.6810286045074463
    },
    {
      "auxiliary_loss_clip": 0.01221737,
      "auxiliary_loss_mlp": 0.01026939,
      "balance_loss_clip": 1.04742575,
      "balance_loss_mlp": 1.02019143,
      "epoch": 0.965069440269344,
      "flos": 23365565280000.0,
      "grad_norm": 2.007865982775203,
      "language_loss": 0.77648377,
      "learning_rate": 1.2744345386275668e-08,
      "loss": 0.79897052,
      "num_input_tokens_seen": 173273980,
      "step": 8026,
      "time_per_iteration": 2.650714874267578
    },
    {
      "auxiliary_loss_clip": 0.01178388,
      "auxiliary_loss_mlp": 0.01026463,
      "balance_loss_clip": 1.05037665,
      "balance_loss_mlp": 1.01935494,
      "epoch": 0.9651896831599832,
      "flos": 25374875155200.0,
      "grad_norm": 1.7239054824452158,
      "language_loss": 0.78824073,
      "learning_rate": 1.265669754695109e-08,
      "loss": 0.81028926,
      "num_input_tokens_seen": 173293550,
      "step": 8027,
      "time_per_iteration": 2.71108341217041
    },
    {
      "auxiliary_loss_clip": 0.01429875,
      "auxiliary_loss_mlp": 0.01024949,
      "balance_loss_clip": 1.04225469,
      "balance_loss_mlp": 1.01749253,
      "epoch": 0.9653099260506223,
      "flos": 22272875596800.0,
      "grad_norm": 2.031049845470471,
      "language_loss": 0.82018495,
      "learning_rate": 1.2569351186025201e-08,
      "loss": 0.84473312,
      "num_input_tokens_seen": 173312005,
      "step": 8028,
      "time_per_iteration": 2.798649311065674
    },
    {
      "auxiliary_loss_clip": 0.01312469,
      "auxiliary_loss_mlp": 0.01024096,
      "balance_loss_clip": 1.04299259,
      "balance_loss_mlp": 1.01733041,
      "epoch": 0.9654301689412613,
      "flos": 26760847386240.0,
      "grad_norm": 1.8386372917942597,
      "language_loss": 0.75632179,
      "learning_rate": 1.2482306316748737e-08,
      "loss": 0.77968752,
      "num_input_tokens_seen": 173332450,
      "step": 8029,
      "time_per_iteration": 2.7903928756713867
    },
    {
      "auxiliary_loss_clip": 0.01227895,
      "auxiliary_loss_mlp": 0.01025516,
      "balance_loss_clip": 1.04663944,
      "balance_loss_mlp": 1.01853967,
      "epoch": 0.9655504118319005,
      "flos": 17412689122560.0,
      "grad_norm": 2.319885902531926,
      "language_loss": 0.78305197,
      "learning_rate": 1.2395562952326021e-08,
      "loss": 0.8055861,
      "num_input_tokens_seen": 173349610,
      "step": 8030,
      "time_per_iteration": 2.599350929260254
    },
    {
      "auxiliary_loss_clip": 0.01279813,
      "auxiliary_loss_mlp": 0.01022881,
      "balance_loss_clip": 1.04765427,
      "balance_loss_mlp": 1.01616049,
      "epoch": 0.9656706547225395,
      "flos": 22126970551680.0,
      "grad_norm": 3.2849519855377154,
      "language_loss": 0.80975688,
      "learning_rate": 1.2309121105916309e-08,
      "loss": 0.83278376,
      "num_input_tokens_seen": 173367900,
      "step": 8031,
      "time_per_iteration": 2.7232558727264404
    },
    {
      "auxiliary_loss_clip": 0.01225984,
      "auxiliary_loss_mlp": 0.01025707,
      "balance_loss_clip": 1.04812646,
      "balance_loss_mlp": 1.01881647,
      "epoch": 0.9657908976131786,
      "flos": 37049289926400.0,
      "grad_norm": 1.9295563429371245,
      "language_loss": 0.6926859,
      "learning_rate": 1.222298079063222e-08,
      "loss": 0.71520281,
      "num_input_tokens_seen": 173389040,
      "step": 8032,
      "time_per_iteration": 3.664524555206299
    },
    {
      "auxiliary_loss_clip": 0.01220205,
      "auxiliary_loss_mlp": 0.01025249,
      "balance_loss_clip": 1.04737258,
      "balance_loss_mlp": 1.01867199,
      "epoch": 0.9659111405038178,
      "flos": 24389809597440.0,
      "grad_norm": 2.490355063281661,
      "language_loss": 0.72396076,
      "learning_rate": 1.2137142019541524e-08,
      "loss": 0.74641526,
      "num_input_tokens_seen": 173407595,
      "step": 8033,
      "time_per_iteration": 2.728853464126587
    },
    {
      "auxiliary_loss_clip": 0.0117827,
      "auxiliary_loss_mlp": 0.01022573,
      "balance_loss_clip": 1.04614115,
      "balance_loss_mlp": 1.01588535,
      "epoch": 0.9660313833944568,
      "flos": 25009412227200.0,
      "grad_norm": 12.03225706718188,
      "language_loss": 0.73627847,
      "learning_rate": 1.2051604805666027e-08,
      "loss": 0.75828689,
      "num_input_tokens_seen": 173424720,
      "step": 8034,
      "time_per_iteration": 3.719008684158325
    },
    {
      "auxiliary_loss_clip": 0.01173256,
      "auxiliary_loss_mlp": 0.02563863,
      "balance_loss_clip": 1.04994988,
      "balance_loss_mlp": 0.99991363,
      "epoch": 0.9661516262850959,
      "flos": 11801575895040.0,
      "grad_norm": 3.014792055101665,
      "language_loss": 0.78594548,
      "learning_rate": 1.196636916198135e-08,
      "loss": 0.82331669,
      "num_input_tokens_seen": 173442260,
      "step": 8035,
      "time_per_iteration": 2.624998092651367
    },
    {
      "auxiliary_loss_clip": 0.01175205,
      "auxiliary_loss_mlp": 0.01028035,
      "balance_loss_clip": 1.04939461,
      "balance_loss_mlp": 1.02124596,
      "epoch": 0.9662718691757349,
      "flos": 20047778766720.0,
      "grad_norm": 4.998653803924958,
      "language_loss": 0.77007449,
      "learning_rate": 1.1881435101418036e-08,
      "loss": 0.79210687,
      "num_input_tokens_seen": 173461675,
      "step": 8036,
      "time_per_iteration": 2.6511096954345703
    },
    {
      "auxiliary_loss_clip": 0.01165546,
      "auxiliary_loss_mlp": 0.01000912,
      "balance_loss_clip": 1.00743604,
      "balance_loss_mlp": 1.00007713,
      "epoch": 0.9663921120663741,
      "flos": 68027703517440.0,
      "grad_norm": 0.8309565040197836,
      "language_loss": 0.65536457,
      "learning_rate": 1.1796802636860003e-08,
      "loss": 0.67702919,
      "num_input_tokens_seen": 173530205,
      "step": 8037,
      "time_per_iteration": 4.14655065536499
    },
    {
      "auxiliary_loss_clip": 0.01174176,
      "auxiliary_loss_mlp": 0.01025032,
      "balance_loss_clip": 1.04876971,
      "balance_loss_mlp": 1.01816797,
      "epoch": 0.9665123549570132,
      "flos": 26322916769280.0,
      "grad_norm": 2.4813926197855123,
      "language_loss": 0.742522,
      "learning_rate": 1.1712471781146316e-08,
      "loss": 0.76451409,
      "num_input_tokens_seen": 173549540,
      "step": 8038,
      "time_per_iteration": 2.598094940185547
    },
    {
      "auxiliary_loss_clip": 0.01171086,
      "auxiliary_loss_mlp": 0.01023887,
      "balance_loss_clip": 1.04652286,
      "balance_loss_mlp": 1.01691663,
      "epoch": 0.9666325978476522,
      "flos": 43941121557120.0,
      "grad_norm": 1.8009121807248005,
      "language_loss": 0.67362493,
      "learning_rate": 1.1628442547069628e-08,
      "loss": 0.69557464,
      "num_input_tokens_seen": 173571740,
      "step": 8039,
      "time_per_iteration": 2.860253095626831
    },
    {
      "auxiliary_loss_clip": 0.01225527,
      "auxiliary_loss_mlp": 0.02569025,
      "balance_loss_clip": 1.04669929,
      "balance_loss_mlp": 0.99992919,
      "epoch": 0.9667528407382914,
      "flos": 21543422198400.0,
      "grad_norm": 1.857871092875102,
      "language_loss": 0.77246845,
      "learning_rate": 1.1544714947377521e-08,
      "loss": 0.81041396,
      "num_input_tokens_seen": 173589425,
      "step": 8040,
      "time_per_iteration": 2.625199556350708
    },
    {
      "auxiliary_loss_clip": 0.01174835,
      "auxiliary_loss_mlp": 0.01028001,
      "balance_loss_clip": 1.05007052,
      "balance_loss_mlp": 1.02099121,
      "epoch": 0.9668730836289304,
      "flos": 23878585278720.0,
      "grad_norm": 2.0841548490694137,
      "language_loss": 0.70080316,
      "learning_rate": 1.1461288994770945e-08,
      "loss": 0.72283149,
      "num_input_tokens_seen": 173608500,
      "step": 8041,
      "time_per_iteration": 2.641010284423828
    },
    {
      "auxiliary_loss_clip": 0.01173848,
      "auxiliary_loss_mlp": 0.01020308,
      "balance_loss_clip": 1.04716635,
      "balance_loss_mlp": 1.01350439,
      "epoch": 0.9669933265195695,
      "flos": 28293011971200.0,
      "grad_norm": 1.793043112607677,
      "language_loss": 0.77351862,
      "learning_rate": 1.1378164701906002e-08,
      "loss": 0.79546022,
      "num_input_tokens_seen": 173630265,
      "step": 8042,
      "time_per_iteration": 3.5909347534179688
    },
    {
      "auxiliary_loss_clip": 0.01176626,
      "auxiliary_loss_mlp": 0.01022163,
      "balance_loss_clip": 1.05062556,
      "balance_loss_mlp": 1.01491511,
      "epoch": 0.9671135694102087,
      "flos": 22454763091200.0,
      "grad_norm": 1.8641250072020366,
      "language_loss": 0.66707027,
      "learning_rate": 1.1295342081392156e-08,
      "loss": 0.68905813,
      "num_input_tokens_seen": 173649625,
      "step": 8043,
      "time_per_iteration": 2.5905184745788574
    },
    {
      "auxiliary_loss_clip": 0.01170769,
      "auxiliary_loss_mlp": 0.01024407,
      "balance_loss_clip": 1.04595518,
      "balance_loss_mlp": 1.01777029,
      "epoch": 0.9672338123008477,
      "flos": 20155941596160.0,
      "grad_norm": 1.641520468140671,
      "language_loss": 0.69892949,
      "learning_rate": 1.1212821145793804e-08,
      "loss": 0.72088122,
      "num_input_tokens_seen": 173669240,
      "step": 8044,
      "time_per_iteration": 2.692209005355835
    },
    {
      "auxiliary_loss_clip": 0.01268084,
      "auxiliary_loss_mlp": 0.01024125,
      "balance_loss_clip": 1.04406261,
      "balance_loss_mlp": 1.01753879,
      "epoch": 0.9673540551914868,
      "flos": 16977487939200.0,
      "grad_norm": 2.1498043369071906,
      "language_loss": 0.78617084,
      "learning_rate": 1.1130601907629156e-08,
      "loss": 0.80909288,
      "num_input_tokens_seen": 173686970,
      "step": 8045,
      "time_per_iteration": 2.6269166469573975
    },
    {
      "auxiliary_loss_clip": 0.01118599,
      "auxiliary_loss_mlp": 0.01000967,
      "balance_loss_clip": 1.00546825,
      "balance_loss_mlp": 1.00005555,
      "epoch": 0.9674742980821259,
      "flos": 61892903952000.0,
      "grad_norm": 0.8116654381175274,
      "language_loss": 0.64719212,
      "learning_rate": 1.1048684379370899e-08,
      "loss": 0.66838777,
      "num_input_tokens_seen": 173747655,
      "step": 8046,
      "time_per_iteration": 3.1674466133117676
    },
    {
      "auxiliary_loss_clip": 0.01263372,
      "auxiliary_loss_mlp": 0.01023445,
      "balance_loss_clip": 1.04480517,
      "balance_loss_mlp": 1.01668012,
      "epoch": 0.967594540972765,
      "flos": 18697824898560.0,
      "grad_norm": 3.947911547100608,
      "language_loss": 0.74619585,
      "learning_rate": 1.0967068573445759e-08,
      "loss": 0.76906407,
      "num_input_tokens_seen": 173765140,
      "step": 8047,
      "time_per_iteration": 2.6678624153137207
    },
    {
      "auxiliary_loss_clip": 0.01269402,
      "auxiliary_loss_mlp": 0.01021831,
      "balance_loss_clip": 1.04453385,
      "balance_loss_mlp": 1.01486576,
      "epoch": 0.967714783863404,
      "flos": 20777411733120.0,
      "grad_norm": 2.73761264754445,
      "language_loss": 0.65491283,
      "learning_rate": 1.0885754502234945e-08,
      "loss": 0.67782521,
      "num_input_tokens_seen": 173784800,
      "step": 8048,
      "time_per_iteration": 2.709085702896118
    },
    {
      "auxiliary_loss_clip": 0.01216764,
      "auxiliary_loss_mlp": 0.01025017,
      "balance_loss_clip": 1.04380882,
      "balance_loss_mlp": 1.01834679,
      "epoch": 0.9678350267540432,
      "flos": 23185473465600.0,
      "grad_norm": 1.7384385773953785,
      "language_loss": 0.77656507,
      "learning_rate": 1.08047421780737e-08,
      "loss": 0.79898286,
      "num_input_tokens_seen": 173803990,
      "step": 8049,
      "time_per_iteration": 2.70662260055542
    },
    {
      "auxiliary_loss_clip": 0.01277419,
      "auxiliary_loss_mlp": 0.02566559,
      "balance_loss_clip": 1.0452491,
      "balance_loss_mlp": 0.99991512,
      "epoch": 0.9679552696446823,
      "flos": 21726063878400.0,
      "grad_norm": 2.1582994485004545,
      "language_loss": 0.73777401,
      "learning_rate": 1.0724031613251305e-08,
      "loss": 0.77621388,
      "num_input_tokens_seen": 173821890,
      "step": 8050,
      "time_per_iteration": 2.6845579147338867
    },
    {
      "auxiliary_loss_clip": 0.01227868,
      "auxiliary_loss_mlp": 0.01020935,
      "balance_loss_clip": 1.04757953,
      "balance_loss_mlp": 1.0139221,
      "epoch": 0.9680755125353213,
      "flos": 26869046129280.0,
      "grad_norm": 2.5593081684382337,
      "language_loss": 0.66222471,
      "learning_rate": 1.0643622820011744e-08,
      "loss": 0.68471277,
      "num_input_tokens_seen": 173842945,
      "step": 8051,
      "time_per_iteration": 2.636754035949707
    },
    {
      "auxiliary_loss_clip": 0.01175734,
      "auxiliary_loss_mlp": 0.01025589,
      "balance_loss_clip": 1.04795444,
      "balance_loss_mlp": 1.01833558,
      "epoch": 0.9681957554259605,
      "flos": 28325008010880.0,
      "grad_norm": 2.7550081878871366,
      "language_loss": 0.67435497,
      "learning_rate": 1.0563515810552814e-08,
      "loss": 0.69636822,
      "num_input_tokens_seen": 173859915,
      "step": 8052,
      "time_per_iteration": 2.636791467666626
    },
    {
      "auxiliary_loss_clip": 0.01176127,
      "auxiliary_loss_mlp": 0.01028166,
      "balance_loss_clip": 1.05177498,
      "balance_loss_mlp": 1.02129364,
      "epoch": 0.9683159983165995,
      "flos": 20557674282240.0,
      "grad_norm": 1.6594236730144831,
      "language_loss": 0.73443973,
      "learning_rate": 1.0483710597026795e-08,
      "loss": 0.7564826,
      "num_input_tokens_seen": 173879775,
      "step": 8053,
      "time_per_iteration": 2.657710552215576
    },
    {
      "auxiliary_loss_clip": 0.01320835,
      "auxiliary_loss_mlp": 0.01024195,
      "balance_loss_clip": 1.04490542,
      "balance_loss_mlp": 1.01782596,
      "epoch": 0.9684362412072386,
      "flos": 24207958016640.0,
      "grad_norm": 2.222342392048259,
      "language_loss": 0.73996222,
      "learning_rate": 1.0404207191540227e-08,
      "loss": 0.76341254,
      "num_input_tokens_seen": 173900230,
      "step": 8054,
      "time_per_iteration": 2.7994613647460938
    },
    {
      "auxiliary_loss_clip": 0.01172314,
      "auxiliary_loss_mlp": 0.01024097,
      "balance_loss_clip": 1.04891658,
      "balance_loss_mlp": 1.01751399,
      "epoch": 0.9685564840978778,
      "flos": 22346241125760.0,
      "grad_norm": 1.9640284189325186,
      "language_loss": 0.74442106,
      "learning_rate": 1.0325005606153236e-08,
      "loss": 0.7663852,
      "num_input_tokens_seen": 173919690,
      "step": 8055,
      "time_per_iteration": 2.6557979583740234
    },
    {
      "auxiliary_loss_clip": 0.01276763,
      "auxiliary_loss_mlp": 0.01028888,
      "balance_loss_clip": 1.04323053,
      "balance_loss_mlp": 1.0214349,
      "epoch": 0.9686767269885168,
      "flos": 14386389477120.0,
      "grad_norm": 2.589112986932865,
      "language_loss": 0.79501343,
      "learning_rate": 1.0246105852881104e-08,
      "loss": 0.81806993,
      "num_input_tokens_seen": 173934790,
      "step": 8056,
      "time_per_iteration": 2.6910030841827393
    },
    {
      "auxiliary_loss_clip": 0.01174553,
      "auxiliary_loss_mlp": 0.01023622,
      "balance_loss_clip": 1.04900265,
      "balance_loss_mlp": 1.0166899,
      "epoch": 0.9687969698791559,
      "flos": 21287630471040.0,
      "grad_norm": 2.1623159046520337,
      "language_loss": 0.79051566,
      "learning_rate": 1.0167507943692476e-08,
      "loss": 0.81249738,
      "num_input_tokens_seen": 173953875,
      "step": 8057,
      "time_per_iteration": 2.598576545715332
    },
    {
      "auxiliary_loss_clip": 0.01222178,
      "auxiliary_loss_mlp": 0.01031032,
      "balance_loss_clip": 1.04995942,
      "balance_loss_mlp": 1.02405262,
      "epoch": 0.968917212769795,
      "flos": 19828328624640.0,
      "grad_norm": 2.746410091008232,
      "language_loss": 0.7153483,
      "learning_rate": 1.008921189051093e-08,
      "loss": 0.73788035,
      "num_input_tokens_seen": 173971220,
      "step": 8058,
      "time_per_iteration": 2.6294760704040527
    },
    {
      "auxiliary_loss_clip": 0.01175139,
      "auxiliary_loss_mlp": 0.01028996,
      "balance_loss_clip": 1.05061531,
      "balance_loss_mlp": 1.02159286,
      "epoch": 0.9690374556604341,
      "flos": 21681749473920.0,
      "grad_norm": 2.342723055330534,
      "language_loss": 0.77297932,
      "learning_rate": 1.0011217705213848e-08,
      "loss": 0.79502064,
      "num_input_tokens_seen": 173989095,
      "step": 8059,
      "time_per_iteration": 3.4895198345184326
    },
    {
      "auxiliary_loss_clip": 0.01220229,
      "auxiliary_loss_mlp": 0.01024994,
      "balance_loss_clip": 1.04891276,
      "balance_loss_mlp": 1.01838064,
      "epoch": 0.9691576985510731,
      "flos": 32635437851520.0,
      "grad_norm": 2.0483492869229885,
      "language_loss": 0.74750006,
      "learning_rate": 9.933525399632658e-09,
      "loss": 0.7699523,
      "num_input_tokens_seen": 174007330,
      "step": 8060,
      "time_per_iteration": 2.695963144302368
    },
    {
      "auxiliary_loss_clip": 0.01274578,
      "auxiliary_loss_mlp": 0.0102967,
      "balance_loss_clip": 1.04855406,
      "balance_loss_mlp": 1.0220288,
      "epoch": 0.9692779414417123,
      "flos": 35663174040960.0,
      "grad_norm": 2.221707014412274,
      "language_loss": 0.65427953,
      "learning_rate": 9.856134985553488e-09,
      "loss": 0.67732203,
      "num_input_tokens_seen": 174027055,
      "step": 8061,
      "time_per_iteration": 3.7503786087036133
    },
    {
      "auxiliary_loss_clip": 0.01170253,
      "auxiliary_loss_mlp": 0.01029805,
      "balance_loss_clip": 1.04753637,
      "balance_loss_mlp": 1.02270865,
      "epoch": 0.9693981843323514,
      "flos": 28366952117760.0,
      "grad_norm": 1.8209859863779456,
      "language_loss": 0.73679531,
      "learning_rate": 9.77904647471628e-09,
      "loss": 0.75879586,
      "num_input_tokens_seen": 174050235,
      "step": 8062,
      "time_per_iteration": 3.5465800762176514
    },
    {
      "auxiliary_loss_clip": 0.01360182,
      "auxiliary_loss_mlp": 0.01018685,
      "balance_loss_clip": 1.04184532,
      "balance_loss_mlp": 1.01242018,
      "epoch": 0.9695184272229904,
      "flos": 23622865378560.0,
      "grad_norm": 2.243483289963047,
      "language_loss": 0.74176961,
      "learning_rate": 9.702259878815454e-09,
      "loss": 0.76555824,
      "num_input_tokens_seen": 174070560,
      "step": 8063,
      "time_per_iteration": 2.7463488578796387
    },
    {
      "auxiliary_loss_clip": 0.0122457,
      "auxiliary_loss_mlp": 0.01027634,
      "balance_loss_clip": 1.04961872,
      "balance_loss_mlp": 1.02028739,
      "epoch": 0.9696386701136296,
      "flos": 23294677789440.0,
      "grad_norm": 2.2445158737513786,
      "language_loss": 0.74112439,
      "learning_rate": 9.625775209499254e-09,
      "loss": 0.76364642,
      "num_input_tokens_seen": 174090565,
      "step": 8064,
      "time_per_iteration": 2.6496777534484863
    },
    {
      "auxiliary_loss_clip": 0.01313065,
      "auxiliary_loss_mlp": 0.01026567,
      "balance_loss_clip": 1.04082966,
      "balance_loss_mlp": 1.01897883,
      "epoch": 0.9697589130042686,
      "flos": 15121876360320.0,
      "grad_norm": 4.17214859314391,
      "language_loss": 0.74477655,
      "learning_rate": 9.549592478370172e-09,
      "loss": 0.76817286,
      "num_input_tokens_seen": 174108745,
      "step": 8065,
      "time_per_iteration": 2.792856454849243
    },
    {
      "auxiliary_loss_clip": 0.01222679,
      "auxiliary_loss_mlp": 0.01023006,
      "balance_loss_clip": 1.0456531,
      "balance_loss_mlp": 1.01650906,
      "epoch": 0.9698791558949077,
      "flos": 18879532824960.0,
      "grad_norm": 1.7745754562034666,
      "language_loss": 0.79300159,
      "learning_rate": 9.473711696985632e-09,
      "loss": 0.81545848,
      "num_input_tokens_seen": 174128075,
      "step": 8066,
      "time_per_iteration": 2.6848390102386475
    },
    {
      "auxiliary_loss_clip": 0.01168318,
      "auxiliary_loss_mlp": 0.01025387,
      "balance_loss_clip": 1.04480743,
      "balance_loss_mlp": 1.0183419,
      "epoch": 0.9699993987855468,
      "flos": 17931455297280.0,
      "grad_norm": 2.0580225806043058,
      "language_loss": 0.76392978,
      "learning_rate": 9.398132876856201e-09,
      "loss": 0.78586686,
      "num_input_tokens_seen": 174147040,
      "step": 8067,
      "time_per_iteration": 2.662001371383667
    },
    {
      "auxiliary_loss_clip": 0.01267688,
      "auxiliary_loss_mlp": 0.00999851,
      "balance_loss_clip": 1.00556421,
      "balance_loss_mlp": 0.99895108,
      "epoch": 0.9701196416761859,
      "flos": 67182186297600.0,
      "grad_norm": 0.7775726462656665,
      "language_loss": 0.60780251,
      "learning_rate": 9.322856029447379e-09,
      "loss": 0.63047791,
      "num_input_tokens_seen": 174208225,
      "step": 8068,
      "time_per_iteration": 4.136422395706177
    },
    {
      "auxiliary_loss_clip": 0.01170418,
      "auxiliary_loss_mlp": 0.01029436,
      "balance_loss_clip": 1.04858756,
      "balance_loss_mlp": 1.02321291,
      "epoch": 0.970239884566825,
      "flos": 24277804012800.0,
      "grad_norm": 2.099567174610099,
      "language_loss": 0.80512297,
      "learning_rate": 9.247881166178695e-09,
      "loss": 0.8271215,
      "num_input_tokens_seen": 174226935,
      "step": 8069,
      "time_per_iteration": 2.603888988494873
    },
    {
      "auxiliary_loss_clip": 0.0132883,
      "auxiliary_loss_mlp": 0.01019421,
      "balance_loss_clip": 1.0457443,
      "balance_loss_mlp": 1.01240277,
      "epoch": 0.970360127457464,
      "flos": 25301689194240.0,
      "grad_norm": 2.1321640052720436,
      "language_loss": 0.76410896,
      "learning_rate": 9.173208298423274e-09,
      "loss": 0.78759146,
      "num_input_tokens_seen": 174248140,
      "step": 8070,
      "time_per_iteration": 2.787639617919922
    },
    {
      "auxiliary_loss_clip": 0.01274785,
      "auxiliary_loss_mlp": 0.02564941,
      "balance_loss_clip": 1.04665685,
      "balance_loss_mlp": 0.99992526,
      "epoch": 0.9704803703481032,
      "flos": 29572473398400.0,
      "grad_norm": 1.8564581719871136,
      "language_loss": 0.76400197,
      "learning_rate": 9.09883743750961e-09,
      "loss": 0.80239928,
      "num_input_tokens_seen": 174271030,
      "step": 8071,
      "time_per_iteration": 2.825347423553467
    },
    {
      "auxiliary_loss_clip": 0.01264719,
      "auxiliary_loss_mlp": 0.01022444,
      "balance_loss_clip": 1.04594445,
      "balance_loss_mlp": 1.01598275,
      "epoch": 0.9706006132387422,
      "flos": 17380046638080.0,
      "grad_norm": 1.6880253693450873,
      "language_loss": 0.84088731,
      "learning_rate": 9.024768594719124e-09,
      "loss": 0.86375892,
      "num_input_tokens_seen": 174289410,
      "step": 8072,
      "time_per_iteration": 2.6846513748168945
    },
    {
      "auxiliary_loss_clip": 0.01326721,
      "auxiliary_loss_mlp": 0.01027764,
      "balance_loss_clip": 1.04704702,
      "balance_loss_mlp": 1.02068877,
      "epoch": 0.9707208561293813,
      "flos": 18186421011840.0,
      "grad_norm": 3.9933943582614937,
      "language_loss": 0.72934031,
      "learning_rate": 8.95100178128816e-09,
      "loss": 0.75288522,
      "num_input_tokens_seen": 174308550,
      "step": 8073,
      "time_per_iteration": 2.68218994140625
    },
    {
      "auxiliary_loss_clip": 0.01272464,
      "auxiliary_loss_mlp": 0.0103013,
      "balance_loss_clip": 1.04469419,
      "balance_loss_mlp": 1.02281678,
      "epoch": 0.9708410990200205,
      "flos": 31248388212480.0,
      "grad_norm": 2.2485615667602237,
      "language_loss": 0.7038933,
      "learning_rate": 8.877537008407321e-09,
      "loss": 0.72691923,
      "num_input_tokens_seen": 174328600,
      "step": 8074,
      "time_per_iteration": 2.7670435905456543
    },
    {
      "auxiliary_loss_clip": 0.01272142,
      "auxiliary_loss_mlp": 0.01021211,
      "balance_loss_clip": 1.04497325,
      "balance_loss_mlp": 1.01455927,
      "epoch": 0.9709613419106595,
      "flos": 30554450386560.0,
      "grad_norm": 1.4928044543411279,
      "language_loss": 0.68599439,
      "learning_rate": 8.804374287221028e-09,
      "loss": 0.70892787,
      "num_input_tokens_seen": 174349835,
      "step": 8075,
      "time_per_iteration": 2.718585968017578
    },
    {
      "auxiliary_loss_clip": 0.01315105,
      "auxiliary_loss_mlp": 0.01022672,
      "balance_loss_clip": 1.03773594,
      "balance_loss_mlp": 1.01642191,
      "epoch": 0.9710815848012986,
      "flos": 23730166281600.0,
      "grad_norm": 1.9185330347536966,
      "language_loss": 0.84149885,
      "learning_rate": 8.731513628827958e-09,
      "loss": 0.86487657,
      "num_input_tokens_seen": 174369200,
      "step": 8076,
      "time_per_iteration": 2.8405776023864746
    },
    {
      "auxiliary_loss_clip": 0.01222345,
      "auxiliary_loss_mlp": 0.01021076,
      "balance_loss_clip": 1.04719925,
      "balance_loss_mlp": 1.01437354,
      "epoch": 0.9712018276919377,
      "flos": 23761875012480.0,
      "grad_norm": 2.6578701387971124,
      "language_loss": 0.82519102,
      "learning_rate": 8.658955044280825e-09,
      "loss": 0.84762526,
      "num_input_tokens_seen": 174388125,
      "step": 8077,
      "time_per_iteration": 2.6409029960632324
    },
    {
      "auxiliary_loss_clip": 0.01225486,
      "auxiliary_loss_mlp": 0.01025731,
      "balance_loss_clip": 1.05053616,
      "balance_loss_mlp": 1.01868916,
      "epoch": 0.9713220705825768,
      "flos": 23330983461120.0,
      "grad_norm": 1.799761148301575,
      "language_loss": 0.7741133,
      "learning_rate": 8.586698544587268e-09,
      "loss": 0.79662549,
      "num_input_tokens_seen": 174409735,
      "step": 8078,
      "time_per_iteration": 2.7247602939605713
    },
    {
      "auxiliary_loss_clip": 0.01269128,
      "auxiliary_loss_mlp": 0.01025593,
      "balance_loss_clip": 1.0435816,
      "balance_loss_mlp": 1.01861334,
      "epoch": 0.9714423134732159,
      "flos": 22200946611840.0,
      "grad_norm": 2.059661096480692,
      "language_loss": 0.73917639,
      "learning_rate": 8.514744140707853e-09,
      "loss": 0.76212358,
      "num_input_tokens_seen": 174428875,
      "step": 8079,
      "time_per_iteration": 2.6683847904205322
    },
    {
      "auxiliary_loss_clip": 0.01170823,
      "auxiliary_loss_mlp": 0.01025527,
      "balance_loss_clip": 1.04808009,
      "balance_loss_mlp": 1.01894665,
      "epoch": 0.971562556363855,
      "flos": 20229917656320.0,
      "grad_norm": 1.8333373131195876,
      "language_loss": 0.76256466,
      "learning_rate": 8.443091843558515e-09,
      "loss": 0.78452814,
      "num_input_tokens_seen": 174447960,
      "step": 8080,
      "time_per_iteration": 2.6159791946411133
    },
    {
      "auxiliary_loss_clip": 0.01271182,
      "auxiliary_loss_mlp": 0.01028475,
      "balance_loss_clip": 1.04580021,
      "balance_loss_mlp": 1.02120614,
      "epoch": 0.9716827992544941,
      "flos": 24970197553920.0,
      "grad_norm": 2.1594355172163726,
      "language_loss": 0.64745605,
      "learning_rate": 8.37174166400878e-09,
      "loss": 0.67045259,
      "num_input_tokens_seen": 174463535,
      "step": 8081,
      "time_per_iteration": 2.7242605686187744
    },
    {
      "auxiliary_loss_clip": 0.01175161,
      "auxiliary_loss_mlp": 0.01025416,
      "balance_loss_clip": 1.0518496,
      "balance_loss_mlp": 1.01861835,
      "epoch": 0.9718030421451331,
      "flos": 24681476033280.0,
      "grad_norm": 3.9349581571842736,
      "language_loss": 0.85379863,
      "learning_rate": 8.300693612881992e-09,
      "loss": 0.87580442,
      "num_input_tokens_seen": 174483600,
      "step": 8082,
      "time_per_iteration": 2.6209068298339844
    },
    {
      "auxiliary_loss_clip": 0.01220607,
      "auxiliary_loss_mlp": 0.02567984,
      "balance_loss_clip": 1.04899859,
      "balance_loss_mlp": 0.99992162,
      "epoch": 0.9719232850357723,
      "flos": 22090700793600.0,
      "grad_norm": 4.173892651063349,
      "language_loss": 0.81389475,
      "learning_rate": 8.22994770095664e-09,
      "loss": 0.85178071,
      "num_input_tokens_seen": 174502175,
      "step": 8083,
      "time_per_iteration": 2.6583497524261475
    },
    {
      "auxiliary_loss_clip": 0.01278595,
      "auxiliary_loss_mlp": 0.01027425,
      "balance_loss_clip": 1.05227232,
      "balance_loss_mlp": 1.01997185,
      "epoch": 0.9720435279264114,
      "flos": 23656908493440.0,
      "grad_norm": 2.1058478498947704,
      "language_loss": 0.75437373,
      "learning_rate": 8.159503938964585e-09,
      "loss": 0.77743393,
      "num_input_tokens_seen": 174519495,
      "step": 8084,
      "time_per_iteration": 3.581129550933838
    },
    {
      "auxiliary_loss_clip": 0.01317405,
      "auxiliary_loss_mlp": 0.01021667,
      "balance_loss_clip": 1.04387951,
      "balance_loss_mlp": 1.01511312,
      "epoch": 0.9721637708170504,
      "flos": 28365910623360.0,
      "grad_norm": 1.9835223632426529,
      "language_loss": 0.70566148,
      "learning_rate": 8.089362337592164e-09,
      "loss": 0.72905219,
      "num_input_tokens_seen": 174543120,
      "step": 8085,
      "time_per_iteration": 2.8353919982910156
    },
    {
      "auxiliary_loss_clip": 0.01266206,
      "auxiliary_loss_mlp": 0.01025418,
      "balance_loss_clip": 1.04677868,
      "balance_loss_mlp": 1.01862025,
      "epoch": 0.9722840137076896,
      "flos": 29130807767040.0,
      "grad_norm": 1.5685078356174094,
      "language_loss": 0.72482204,
      "learning_rate": 8.019522907479536e-09,
      "loss": 0.74773836,
      "num_input_tokens_seen": 174563480,
      "step": 8086,
      "time_per_iteration": 2.904819965362549
    },
    {
      "auxiliary_loss_clip": 0.01224958,
      "auxiliary_loss_mlp": 0.01021784,
      "balance_loss_clip": 1.04844141,
      "balance_loss_mlp": 1.01539779,
      "epoch": 0.9724042565983286,
      "flos": 19243954258560.0,
      "grad_norm": 2.203743715705549,
      "language_loss": 0.77358413,
      "learning_rate": 7.949985659221558e-09,
      "loss": 0.79605156,
      "num_input_tokens_seen": 174580745,
      "step": 8087,
      "time_per_iteration": 3.533684253692627
    },
    {
      "auxiliary_loss_clip": 0.01274409,
      "auxiliary_loss_mlp": 0.01023766,
      "balance_loss_clip": 1.04587162,
      "balance_loss_mlp": 1.0171411,
      "epoch": 0.9725244994889677,
      "flos": 23039676161280.0,
      "grad_norm": 2.2094302506806116,
      "language_loss": 0.79168725,
      "learning_rate": 7.880750603366904e-09,
      "loss": 0.81466901,
      "num_input_tokens_seen": 174599615,
      "step": 8088,
      "time_per_iteration": 2.74710750579834
    },
    {
      "auxiliary_loss_clip": 0.01330054,
      "auxiliary_loss_mlp": 0.01026416,
      "balance_loss_clip": 1.04416013,
      "balance_loss_mlp": 1.01887035,
      "epoch": 0.9726447423796069,
      "flos": 23367468700800.0,
      "grad_norm": 2.2250387810220285,
      "language_loss": 0.79522294,
      "learning_rate": 7.811817750418282e-09,
      "loss": 0.81878769,
      "num_input_tokens_seen": 174618375,
      "step": 8089,
      "time_per_iteration": 3.603872060775757
    },
    {
      "auxiliary_loss_clip": 0.01315217,
      "auxiliary_loss_mlp": 0.01028178,
      "balance_loss_clip": 1.04503441,
      "balance_loss_mlp": 1.02108502,
      "epoch": 0.9727649852702459,
      "flos": 26541648639360.0,
      "grad_norm": 1.5735079734233341,
      "language_loss": 0.80234194,
      "learning_rate": 7.743187110833105e-09,
      "loss": 0.82577586,
      "num_input_tokens_seen": 174641135,
      "step": 8090,
      "time_per_iteration": 2.8274173736572266
    },
    {
      "auxiliary_loss_clip": 0.01271954,
      "auxiliary_loss_mlp": 0.01024299,
      "balance_loss_clip": 1.04309607,
      "balance_loss_mlp": 1.01764965,
      "epoch": 0.972885228160885,
      "flos": 20522338277760.0,
      "grad_norm": 1.4741514796056676,
      "language_loss": 0.80824244,
      "learning_rate": 7.674858695022602e-09,
      "loss": 0.83120501,
      "num_input_tokens_seen": 174659490,
      "step": 8091,
      "time_per_iteration": 2.667311429977417
    },
    {
      "auxiliary_loss_clip": 0.01175085,
      "auxiliary_loss_mlp": 0.01024326,
      "balance_loss_clip": 1.04975629,
      "balance_loss_mlp": 1.01720297,
      "epoch": 0.9730054710515241,
      "flos": 17566064196480.0,
      "grad_norm": 2.5853046464808886,
      "language_loss": 0.76385319,
      "learning_rate": 7.606832513351591e-09,
      "loss": 0.78584731,
      "num_input_tokens_seen": 174677440,
      "step": 8092,
      "time_per_iteration": 2.6024863719940186
    },
    {
      "auxiliary_loss_clip": 0.01059433,
      "auxiliary_loss_mlp": 0.0250491,
      "balance_loss_clip": 1.00553203,
      "balance_loss_mlp": 0.99987334,
      "epoch": 0.9731257139421632,
      "flos": 68972010117120.0,
      "grad_norm": 0.8243247616888904,
      "language_loss": 0.63899928,
      "learning_rate": 7.539108576140264e-09,
      "loss": 0.67464268,
      "num_input_tokens_seen": 174741550,
      "step": 8093,
      "time_per_iteration": 3.257070302963257
    },
    {
      "auxiliary_loss_clip": 0.01365756,
      "auxiliary_loss_mlp": 0.01021974,
      "balance_loss_clip": 1.04346263,
      "balance_loss_mlp": 1.01540875,
      "epoch": 0.9732459568328022,
      "flos": 18478841633280.0,
      "grad_norm": 2.064540588734389,
      "language_loss": 0.70204836,
      "learning_rate": 7.471686893661732e-09,
      "loss": 0.72592568,
      "num_input_tokens_seen": 174759845,
      "step": 8094,
      "time_per_iteration": 3.644542694091797
    },
    {
      "auxiliary_loss_clip": 0.01270293,
      "auxiliary_loss_mlp": 0.01023353,
      "balance_loss_clip": 1.04895413,
      "balance_loss_mlp": 1.01709199,
      "epoch": 0.9733661997234414,
      "flos": 20883886623360.0,
      "grad_norm": 1.987032716125693,
      "language_loss": 0.64435399,
      "learning_rate": 7.4045674761442636e-09,
      "loss": 0.66729045,
      "num_input_tokens_seen": 174777175,
      "step": 8095,
      "time_per_iteration": 2.6821351051330566
    },
    {
      "auxiliary_loss_clip": 0.01172094,
      "auxiliary_loss_mlp": 0.02565719,
      "balance_loss_clip": 1.04933703,
      "balance_loss_mlp": 0.99991035,
      "epoch": 0.9734864426140805,
      "flos": 23766795175680.0,
      "grad_norm": 1.8894643921673775,
      "language_loss": 0.73856938,
      "learning_rate": 7.337750333769488e-09,
      "loss": 0.77594745,
      "num_input_tokens_seen": 174796980,
      "step": 8096,
      "time_per_iteration": 2.580782413482666
    },
    {
      "auxiliary_loss_clip": 0.01276015,
      "auxiliary_loss_mlp": 0.01021324,
      "balance_loss_clip": 1.04219389,
      "balance_loss_mlp": 1.0144577,
      "epoch": 0.9736066855047195,
      "flos": 35042422176000.0,
      "grad_norm": 1.8119797976548588,
      "language_loss": 0.7297368,
      "learning_rate": 7.2712354766737425e-09,
      "loss": 0.75271016,
      "num_input_tokens_seen": 174817310,
      "step": 8097,
      "time_per_iteration": 2.7738685607910156
    },
    {
      "auxiliary_loss_clip": 0.01321744,
      "auxiliary_loss_mlp": 0.01026502,
      "balance_loss_clip": 1.04853201,
      "balance_loss_mlp": 1.01971006,
      "epoch": 0.9737269283953586,
      "flos": 20410620001920.0,
      "grad_norm": 2.2569761156082744,
      "language_loss": 0.80883372,
      "learning_rate": 7.2050229149469565e-09,
      "loss": 0.83231616,
      "num_input_tokens_seen": 174837320,
      "step": 8098,
      "time_per_iteration": 2.7790071964263916
    },
    {
      "auxiliary_loss_clip": 0.01322319,
      "auxiliary_loss_mlp": 0.0102751,
      "balance_loss_clip": 1.04011869,
      "balance_loss_mlp": 1.02114415,
      "epoch": 0.9738471712859977,
      "flos": 28911680847360.0,
      "grad_norm": 2.8687149388324262,
      "language_loss": 0.63674074,
      "learning_rate": 7.139112658633984e-09,
      "loss": 0.66023898,
      "num_input_tokens_seen": 174857470,
      "step": 8099,
      "time_per_iteration": 2.751539707183838
    },
    {
      "auxiliary_loss_clip": 0.01317728,
      "auxiliary_loss_mlp": 0.01020667,
      "balance_loss_clip": 1.04534495,
      "balance_loss_mlp": 1.01407754,
      "epoch": 0.9739674141766368,
      "flos": 27782326356480.0,
      "grad_norm": 2.0893943711549023,
      "language_loss": 0.70336163,
      "learning_rate": 7.073504717733048e-09,
      "loss": 0.72674561,
      "num_input_tokens_seen": 174877035,
      "step": 8100,
      "time_per_iteration": 2.77908992767334
    },
    {
      "auxiliary_loss_clip": 0.01172908,
      "auxiliary_loss_mlp": 0.01005029,
      "balance_loss_clip": 1.00688207,
      "balance_loss_mlp": 1.00408733,
      "epoch": 0.9740876570672758,
      "flos": 68863057188480.0,
      "grad_norm": 0.8278038706050304,
      "language_loss": 0.57168984,
      "learning_rate": 7.008199102196855e-09,
      "loss": 0.59346926,
      "num_input_tokens_seen": 174938460,
      "step": 8101,
      "time_per_iteration": 3.2262117862701416
    },
    {
      "auxiliary_loss_clip": 0.01157486,
      "auxiliary_loss_mlp": 0.01002725,
      "balance_loss_clip": 1.00659549,
      "balance_loss_mlp": 1.0018611,
      "epoch": 0.974207899957915,
      "flos": 58236622646400.0,
      "grad_norm": 0.7988962252651206,
      "language_loss": 0.58890647,
      "learning_rate": 6.9431958219321464e-09,
      "loss": 0.61050856,
      "num_input_tokens_seen": 174994625,
      "step": 8102,
      "time_per_iteration": 3.1899397373199463
    },
    {
      "auxiliary_loss_clip": 0.01268683,
      "auxiliary_loss_mlp": 0.01022234,
      "balance_loss_clip": 1.04328573,
      "balance_loss_mlp": 1.01563549,
      "epoch": 0.9743281428485541,
      "flos": 22600057605120.0,
      "grad_norm": 1.5224925279099604,
      "language_loss": 0.77706081,
      "learning_rate": 6.878494886800146e-09,
      "loss": 0.79997003,
      "num_input_tokens_seen": 175015400,
      "step": 8103,
      "time_per_iteration": 2.727565288543701
    },
    {
      "auxiliary_loss_clip": 0.01172085,
      "auxiliary_loss_mlp": 0.0102364,
      "balance_loss_clip": 1.04633248,
      "balance_loss_mlp": 1.01668143,
      "epoch": 0.9744483857391931,
      "flos": 20008815488640.0,
      "grad_norm": 2.04076562862349,
      "language_loss": 0.76383328,
      "learning_rate": 6.814096306615669e-09,
      "loss": 0.78579056,
      "num_input_tokens_seen": 175033540,
      "step": 8104,
      "time_per_iteration": 2.690594434738159
    },
    {
      "auxiliary_loss_clip": 0.01278807,
      "auxiliary_loss_mlp": 0.01033929,
      "balance_loss_clip": 1.04651427,
      "balance_loss_mlp": 1.02629066,
      "epoch": 0.9745686286298323,
      "flos": 17675268520320.0,
      "grad_norm": 2.886516735077991,
      "language_loss": 0.65310633,
      "learning_rate": 6.750000091148011e-09,
      "loss": 0.67623377,
      "num_input_tokens_seen": 175050835,
      "step": 8105,
      "time_per_iteration": 2.659674644470215
    },
    {
      "auxiliary_loss_clip": 0.01173217,
      "auxiliary_loss_mlp": 0.01030419,
      "balance_loss_clip": 1.0502317,
      "balance_loss_mlp": 1.0235641,
      "epoch": 0.9746888715204713,
      "flos": 29460252332160.0,
      "grad_norm": 5.740910309039396,
      "language_loss": 0.72491604,
      "learning_rate": 6.686206250120729e-09,
      "loss": 0.74695241,
      "num_input_tokens_seen": 175072330,
      "step": 8106,
      "time_per_iteration": 2.6384987831115723
    },
    {
      "auxiliary_loss_clip": 0.01323857,
      "auxiliary_loss_mlp": 0.01026275,
      "balance_loss_clip": 1.0424962,
      "balance_loss_mlp": 1.01969206,
      "epoch": 0.9748091144111104,
      "flos": 18479308510080.0,
      "grad_norm": 6.671808009581269,
      "language_loss": 0.74327302,
      "learning_rate": 6.622714793210749e-09,
      "loss": 0.7667743,
      "num_input_tokens_seen": 175091250,
      "step": 8107,
      "time_per_iteration": 2.6937525272369385
    },
    {
      "auxiliary_loss_clip": 0.01172067,
      "auxiliary_loss_mlp": 0.01021546,
      "balance_loss_clip": 1.0482111,
      "balance_loss_mlp": 1.0150671,
      "epoch": 0.9749293573017496,
      "flos": 20665154753280.0,
      "grad_norm": 1.7627876023356712,
      "language_loss": 0.78521097,
      "learning_rate": 6.559525730050364e-09,
      "loss": 0.80714709,
      "num_input_tokens_seen": 175111350,
      "step": 8108,
      "time_per_iteration": 2.5429210662841797
    },
    {
      "auxiliary_loss_clip": 0.01319367,
      "auxiliary_loss_mlp": 0.01023054,
      "balance_loss_clip": 1.04529226,
      "balance_loss_mlp": 1.01638722,
      "epoch": 0.9750496001923886,
      "flos": 18478590238080.0,
      "grad_norm": 1.8790345632150771,
      "language_loss": 0.76119643,
      "learning_rate": 6.496639070224574e-09,
      "loss": 0.78462058,
      "num_input_tokens_seen": 175129835,
      "step": 8109,
      "time_per_iteration": 2.64664888381958
    },
    {
      "auxiliary_loss_clip": 0.01225587,
      "auxiliary_loss_mlp": 0.01028594,
      "balance_loss_clip": 1.04772305,
      "balance_loss_mlp": 1.02203763,
      "epoch": 0.9751698430830277,
      "flos": 19572967860480.0,
      "grad_norm": 2.5282871089098293,
      "language_loss": 0.8384459,
      "learning_rate": 6.4340548232739714e-09,
      "loss": 0.86098766,
      "num_input_tokens_seen": 175146035,
      "step": 8110,
      "time_per_iteration": 3.537400245666504
    },
    {
      "auxiliary_loss_clip": 0.01327295,
      "auxiliary_loss_mlp": 0.0102736,
      "balance_loss_clip": 1.04418004,
      "balance_loss_mlp": 1.02108693,
      "epoch": 0.9752900859736668,
      "flos": 23550325862400.0,
      "grad_norm": 1.7768572852416236,
      "language_loss": 0.79411519,
      "learning_rate": 6.371772998692071e-09,
      "loss": 0.8176617,
      "num_input_tokens_seen": 175165290,
      "step": 8111,
      "time_per_iteration": 2.74153208732605
    },
    {
      "auxiliary_loss_clip": 0.01322445,
      "auxiliary_loss_mlp": 0.01022688,
      "balance_loss_clip": 1.04243147,
      "balance_loss_mlp": 1.01550317,
      "epoch": 0.9754103288643059,
      "flos": 20303211358080.0,
      "grad_norm": 3.301652686147625,
      "language_loss": 0.64821875,
      "learning_rate": 6.309793605927094e-09,
      "loss": 0.67167008,
      "num_input_tokens_seen": 175183610,
      "step": 8112,
      "time_per_iteration": 2.693718671798706
    },
    {
      "auxiliary_loss_clip": 0.01176939,
      "auxiliary_loss_mlp": 0.01027166,
      "balance_loss_clip": 1.04614913,
      "balance_loss_mlp": 1.02101231,
      "epoch": 0.975530571754945,
      "flos": 19350680544000.0,
      "grad_norm": 2.148641129515742,
      "language_loss": 0.80508924,
      "learning_rate": 6.248116654381297e-09,
      "loss": 0.82713032,
      "num_input_tokens_seen": 175202080,
      "step": 8113,
      "time_per_iteration": 3.5914552211761475
    },
    {
      "auxiliary_loss_clip": 0.01275093,
      "auxiliary_loss_mlp": 0.01024461,
      "balance_loss_clip": 1.04424906,
      "balance_loss_mlp": 1.0181073,
      "epoch": 0.9756508146455841,
      "flos": 23583399310080.0,
      "grad_norm": 2.3031475074120706,
      "language_loss": 0.72816038,
      "learning_rate": 6.186742153410751e-09,
      "loss": 0.75115585,
      "num_input_tokens_seen": 175221575,
      "step": 8114,
      "time_per_iteration": 2.7147908210754395
    },
    {
      "auxiliary_loss_clip": 0.01271675,
      "auxiliary_loss_mlp": 0.01032669,
      "balance_loss_clip": 1.04536605,
      "balance_loss_mlp": 1.02557564,
      "epoch": 0.9757710575362232,
      "flos": 22966921163520.0,
      "grad_norm": 6.431900452535017,
      "language_loss": 0.87507898,
      "learning_rate": 6.125670112326453e-09,
      "loss": 0.89812243,
      "num_input_tokens_seen": 175240835,
      "step": 8115,
      "time_per_iteration": 3.5621843338012695
    },
    {
      "auxiliary_loss_clip": 0.01226397,
      "auxiliary_loss_mlp": 0.01024962,
      "balance_loss_clip": 1.04778028,
      "balance_loss_mlp": 1.01814318,
      "epoch": 0.9758913004268622,
      "flos": 27966009530880.0,
      "grad_norm": 1.5947281458875502,
      "language_loss": 0.70410728,
      "learning_rate": 6.064900540392548e-09,
      "loss": 0.72662085,
      "num_input_tokens_seen": 175262930,
      "step": 8116,
      "time_per_iteration": 2.7153406143188477
    },
    {
      "auxiliary_loss_clip": 0.01265116,
      "auxiliary_loss_mlp": 0.01027228,
      "balance_loss_clip": 1.04768705,
      "balance_loss_mlp": 1.0210886,
      "epoch": 0.9760115433175014,
      "flos": 22200156512640.0,
      "grad_norm": 1.9783632060827394,
      "language_loss": 0.78825289,
      "learning_rate": 6.0044334468278835e-09,
      "loss": 0.81117636,
      "num_input_tokens_seen": 175282275,
      "step": 8117,
      "time_per_iteration": 2.7277421951293945
    },
    {
      "auxiliary_loss_clip": 0.01367694,
      "auxiliary_loss_mlp": 0.01031055,
      "balance_loss_clip": 1.04208255,
      "balance_loss_mlp": 1.0233686,
      "epoch": 0.9761317862081405,
      "flos": 26250736389120.0,
      "grad_norm": 1.7978796383149325,
      "language_loss": 0.7161262,
      "learning_rate": 5.944268840805345e-09,
      "loss": 0.74011368,
      "num_input_tokens_seen": 175303020,
      "step": 8118,
      "time_per_iteration": 2.7255361080169678
    },
    {
      "auxiliary_loss_clip": 0.01317571,
      "auxiliary_loss_mlp": 0.0102212,
      "balance_loss_clip": 1.04298282,
      "balance_loss_mlp": 1.01615906,
      "epoch": 0.9762520290987795,
      "flos": 26575440359040.0,
      "grad_norm": 4.822984836200634,
      "language_loss": 0.64116704,
      "learning_rate": 5.88440673145163e-09,
      "loss": 0.66456389,
      "num_input_tokens_seen": 175324070,
      "step": 8119,
      "time_per_iteration": 2.7915236949920654
    },
    {
      "auxiliary_loss_clip": 0.01220199,
      "auxiliary_loss_mlp": 0.01026081,
      "balance_loss_clip": 1.05035663,
      "balance_loss_mlp": 1.01922047,
      "epoch": 0.9763722719894187,
      "flos": 18005036307840.0,
      "grad_norm": 3.191448174318995,
      "language_loss": 0.82834876,
      "learning_rate": 5.824847127848142e-09,
      "loss": 0.85081154,
      "num_input_tokens_seen": 175342595,
      "step": 8120,
      "time_per_iteration": 3.515204906463623
    },
    {
      "auxiliary_loss_clip": 0.01373791,
      "auxiliary_loss_mlp": 0.0103252,
      "balance_loss_clip": 1.04596436,
      "balance_loss_mlp": 1.0256114,
      "epoch": 0.9764925148800577,
      "flos": 22455660931200.0,
      "grad_norm": 1.8335337308634592,
      "language_loss": 0.78982365,
      "learning_rate": 5.765590039029433e-09,
      "loss": 0.81388676,
      "num_input_tokens_seen": 175361915,
      "step": 8121,
      "time_per_iteration": 2.7313811779022217
    },
    {
      "auxiliary_loss_clip": 0.01173664,
      "auxiliary_loss_mlp": 0.01026049,
      "balance_loss_clip": 1.05110669,
      "balance_loss_mlp": 1.01927197,
      "epoch": 0.9766127577706968,
      "flos": 36757084786560.0,
      "grad_norm": 2.4584418863121416,
      "language_loss": 0.7116769,
      "learning_rate": 5.706635473985422e-09,
      "loss": 0.73367405,
      "num_input_tokens_seen": 175385785,
      "step": 8122,
      "time_per_iteration": 2.6834144592285156
    },
    {
      "auxiliary_loss_clip": 0.01218873,
      "auxiliary_loss_mlp": 0.01024027,
      "balance_loss_clip": 1.04488683,
      "balance_loss_mlp": 1.01692176,
      "epoch": 0.976733000661336,
      "flos": 22309971367680.0,
      "grad_norm": 2.1595264601935886,
      "language_loss": 0.85083747,
      "learning_rate": 5.6479834416591764e-09,
      "loss": 0.87326646,
      "num_input_tokens_seen": 175405145,
      "step": 8123,
      "time_per_iteration": 2.6524124145507812
    },
    {
      "auxiliary_loss_clip": 0.01221033,
      "auxiliary_loss_mlp": 0.02569138,
      "balance_loss_clip": 1.048069,
      "balance_loss_mlp": 0.99987769,
      "epoch": 0.976853243551975,
      "flos": 25810938264960.0,
      "grad_norm": 2.5939874355795367,
      "language_loss": 0.68607068,
      "learning_rate": 5.589633950947803e-09,
      "loss": 0.72397232,
      "num_input_tokens_seen": 175422645,
      "step": 8124,
      "time_per_iteration": 2.6467039585113525
    },
    {
      "auxiliary_loss_clip": 0.01271243,
      "auxiliary_loss_mlp": 0.01025784,
      "balance_loss_clip": 1.04639804,
      "balance_loss_mlp": 1.01827347,
      "epoch": 0.9769734864426141,
      "flos": 21397445326080.0,
      "grad_norm": 1.9917340242041177,
      "language_loss": 0.70092523,
      "learning_rate": 5.5315870107035535e-09,
      "loss": 0.72389555,
      "num_input_tokens_seen": 175440695,
      "step": 8125,
      "time_per_iteration": 2.6610519886016846
    },
    {
      "auxiliary_loss_clip": 0.01272935,
      "auxiliary_loss_mlp": 0.01025066,
      "balance_loss_clip": 1.04869461,
      "balance_loss_mlp": 1.01805639,
      "epoch": 0.9770937293332532,
      "flos": 13990977584640.0,
      "grad_norm": 2.0604930519317213,
      "language_loss": 0.78960466,
      "learning_rate": 5.473842629731607e-09,
      "loss": 0.81258476,
      "num_input_tokens_seen": 175459195,
      "step": 8126,
      "time_per_iteration": 2.7171144485473633
    },
    {
      "auxiliary_loss_clip": 0.01279565,
      "auxiliary_loss_mlp": 0.02568369,
      "balance_loss_clip": 1.04635072,
      "balance_loss_mlp": 0.99994457,
      "epoch": 0.9772139722238923,
      "flos": 17931994001280.0,
      "grad_norm": 1.9912109999352892,
      "language_loss": 0.78261828,
      "learning_rate": 5.416400816792066e-09,
      "loss": 0.82109761,
      "num_input_tokens_seen": 175476710,
      "step": 8127,
      "time_per_iteration": 2.6238460540771484
    },
    {
      "auxiliary_loss_clip": 0.01171202,
      "auxiliary_loss_mlp": 0.01025033,
      "balance_loss_clip": 1.04819536,
      "balance_loss_mlp": 1.01784492,
      "epoch": 0.9773342151145313,
      "flos": 20446171488000.0,
      "grad_norm": 2.625799802977219,
      "language_loss": 0.7864815,
      "learning_rate": 5.359261580598407e-09,
      "loss": 0.8084439,
      "num_input_tokens_seen": 175492550,
      "step": 8128,
      "time_per_iteration": 2.59236216545105
    },
    {
      "auxiliary_loss_clip": 0.01220888,
      "auxiliary_loss_mlp": 0.01024604,
      "balance_loss_clip": 1.04707026,
      "balance_loss_mlp": 1.01672482,
      "epoch": 0.9774544580051704,
      "flos": 11837306949120.0,
      "grad_norm": 2.301133808003227,
      "language_loss": 0.78066677,
      "learning_rate": 5.302424929819027e-09,
      "loss": 0.80312169,
      "num_input_tokens_seen": 175506560,
      "step": 8129,
      "time_per_iteration": 2.575076103210449
    },
    {
      "auxiliary_loss_clip": 0.01224774,
      "auxiliary_loss_mlp": 0.01023147,
      "balance_loss_clip": 1.04411221,
      "balance_loss_mlp": 1.01603937,
      "epoch": 0.9775747008958096,
      "flos": 13479932833920.0,
      "grad_norm": 2.3345552991396996,
      "language_loss": 0.73527825,
      "learning_rate": 5.24589087307592e-09,
      "loss": 0.75775743,
      "num_input_tokens_seen": 175524180,
      "step": 8130,
      "time_per_iteration": 2.630694627761841
    },
    {
      "auxiliary_loss_clip": 0.01174957,
      "auxiliary_loss_mlp": 0.01025575,
      "balance_loss_clip": 1.04823565,
      "balance_loss_mlp": 1.01838374,
      "epoch": 0.9776949437864486,
      "flos": 59532314042880.0,
      "grad_norm": 2.4688961882876854,
      "language_loss": 0.6480757,
      "learning_rate": 5.189659418944891e-09,
      "loss": 0.67008102,
      "num_input_tokens_seen": 175554355,
      "step": 8131,
      "time_per_iteration": 3.0589349269866943
    },
    {
      "auxiliary_loss_clip": 0.01173222,
      "auxiliary_loss_mlp": 0.01028106,
      "balance_loss_clip": 1.04964876,
      "balance_loss_mlp": 1.02104282,
      "epoch": 0.9778151866770877,
      "flos": 21178605715200.0,
      "grad_norm": 1.9564665642922583,
      "language_loss": 0.78578341,
      "learning_rate": 5.133730575956674e-09,
      "loss": 0.80779666,
      "num_input_tokens_seen": 175574025,
      "step": 8132,
      "time_per_iteration": 2.6388168334960938
    },
    {
      "auxiliary_loss_clip": 0.01272742,
      "auxiliary_loss_mlp": 0.01024266,
      "balance_loss_clip": 1.04585528,
      "balance_loss_mlp": 1.01774561,
      "epoch": 0.9779354295677268,
      "flos": 20886795624960.0,
      "grad_norm": 2.94121351394592,
      "language_loss": 0.72344553,
      "learning_rate": 5.0781043525953696e-09,
      "loss": 0.74641562,
      "num_input_tokens_seen": 175592090,
      "step": 8133,
      "time_per_iteration": 2.6564722061157227
    },
    {
      "auxiliary_loss_clip": 0.01266018,
      "auxiliary_loss_mlp": 0.01025051,
      "balance_loss_clip": 1.04704213,
      "balance_loss_mlp": 1.01853633,
      "epoch": 0.9780556724583659,
      "flos": 23440618748160.0,
      "grad_norm": 1.847890416653465,
      "language_loss": 0.74091196,
      "learning_rate": 5.0227807572995605e-09,
      "loss": 0.76382262,
      "num_input_tokens_seen": 175614065,
      "step": 8134,
      "time_per_iteration": 2.70780873298645
    },
    {
      "auxiliary_loss_clip": 0.0127298,
      "auxiliary_loss_mlp": 0.01024892,
      "balance_loss_clip": 1.04519677,
      "balance_loss_mlp": 1.01827574,
      "epoch": 0.9781759153490049,
      "flos": 20923244951040.0,
      "grad_norm": 3.7079227045462817,
      "language_loss": 0.67633152,
      "learning_rate": 4.967759798461646e-09,
      "loss": 0.69931018,
      "num_input_tokens_seen": 175632410,
      "step": 8135,
      "time_per_iteration": 2.638004779815674
    },
    {
      "auxiliary_loss_clip": 0.0117154,
      "auxiliary_loss_mlp": 0.01022112,
      "balance_loss_clip": 1.04987288,
      "balance_loss_mlp": 1.01595759,
      "epoch": 0.9782961582396441,
      "flos": 28293191539200.0,
      "grad_norm": 2.051541617860036,
      "language_loss": 0.75004125,
      "learning_rate": 4.913041484428282e-09,
      "loss": 0.77197778,
      "num_input_tokens_seen": 175652885,
      "step": 8136,
      "time_per_iteration": 2.674104690551758
    },
    {
      "auxiliary_loss_clip": 0.01225637,
      "auxiliary_loss_mlp": 0.01024227,
      "balance_loss_clip": 1.04892349,
      "balance_loss_mlp": 1.01780462,
      "epoch": 0.9784164011302832,
      "flos": 25552955808000.0,
      "grad_norm": 2.0098225567106742,
      "language_loss": 0.73827291,
      "learning_rate": 4.858625823500384e-09,
      "loss": 0.76077157,
      "num_input_tokens_seen": 175670585,
      "step": 8137,
      "time_per_iteration": 3.595191240310669
    },
    {
      "auxiliary_loss_clip": 0.01225292,
      "auxiliary_loss_mlp": 0.01022951,
      "balance_loss_clip": 1.04703152,
      "balance_loss_mlp": 1.01611996,
      "epoch": 0.9785366440209222,
      "flos": 29965945956480.0,
      "grad_norm": 2.2428913617810364,
      "language_loss": 0.73451686,
      "learning_rate": 4.80451282393246e-09,
      "loss": 0.75699925,
      "num_input_tokens_seen": 175690570,
      "step": 8138,
      "time_per_iteration": 3.527982234954834
    },
    {
      "auxiliary_loss_clip": 0.01172602,
      "auxiliary_loss_mlp": 0.01024184,
      "balance_loss_clip": 1.04644108,
      "balance_loss_mlp": 1.01725817,
      "epoch": 0.9786568869115614,
      "flos": 32343591847680.0,
      "grad_norm": 3.792971419525503,
      "language_loss": 0.67405462,
      "learning_rate": 4.750702493933722e-09,
      "loss": 0.69602251,
      "num_input_tokens_seen": 175710455,
      "step": 8139,
      "time_per_iteration": 2.748850107192993
    },
    {
      "auxiliary_loss_clip": 0.01276676,
      "auxiliary_loss_mlp": 0.025655,
      "balance_loss_clip": 1.04895186,
      "balance_loss_mlp": 0.99993134,
      "epoch": 0.9787771298022004,
      "flos": 23331414424320.0,
      "grad_norm": 2.3537783296508974,
      "language_loss": 0.8501485,
      "learning_rate": 4.697194841666974e-09,
      "loss": 0.88857025,
      "num_input_tokens_seen": 175729380,
      "step": 8140,
      "time_per_iteration": 2.6363420486450195
    },
    {
      "auxiliary_loss_clip": 0.01222637,
      "auxiliary_loss_mlp": 0.01024198,
      "balance_loss_clip": 1.04742908,
      "balance_loss_mlp": 1.01653266,
      "epoch": 0.9788973726928395,
      "flos": 21468548298240.0,
      "grad_norm": 3.2536707362221367,
      "language_loss": 0.82157284,
      "learning_rate": 4.6439898752492764e-09,
      "loss": 0.84404123,
      "num_input_tokens_seen": 175749520,
      "step": 8141,
      "time_per_iteration": 3.56811785697937
    },
    {
      "auxiliary_loss_clip": 0.01113895,
      "auxiliary_loss_mlp": 0.02505272,
      "balance_loss_clip": 1.00646019,
      "balance_loss_mlp": 0.99989361,
      "epoch": 0.9790176155834787,
      "flos": 68897459439360.0,
      "grad_norm": 0.75043556857342,
      "language_loss": 0.63661683,
      "learning_rate": 4.591087602751731e-09,
      "loss": 0.67280853,
      "num_input_tokens_seen": 175811380,
      "step": 8142,
      "time_per_iteration": 3.3031511306762695
    },
    {
      "auxiliary_loss_clip": 0.01218842,
      "auxiliary_loss_mlp": 0.0102109,
      "balance_loss_clip": 1.04692125,
      "balance_loss_mlp": 1.01423287,
      "epoch": 0.9791378584741177,
      "flos": 21430877909760.0,
      "grad_norm": 1.6888868371381534,
      "language_loss": 0.72147393,
      "learning_rate": 4.538488032199916e-09,
      "loss": 0.7438733,
      "num_input_tokens_seen": 175829480,
      "step": 8143,
      "time_per_iteration": 2.6398508548736572
    },
    {
      "auxiliary_loss_clip": 0.01226777,
      "auxiliary_loss_mlp": 0.01030088,
      "balance_loss_clip": 1.04656982,
      "balance_loss_mlp": 1.0231024,
      "epoch": 0.9792581013647568,
      "flos": 20153032594560.0,
      "grad_norm": 2.1645152874799054,
      "language_loss": 0.69227058,
      "learning_rate": 4.486191171572784e-09,
      "loss": 0.71483922,
      "num_input_tokens_seen": 175846750,
      "step": 8144,
      "time_per_iteration": 2.670585870742798
    },
    {
      "auxiliary_loss_clip": 0.01226561,
      "auxiliary_loss_mlp": 0.01023141,
      "balance_loss_clip": 1.04906023,
      "balance_loss_mlp": 1.01669443,
      "epoch": 0.9793783442553959,
      "flos": 23728191033600.0,
      "grad_norm": 4.948835463403625,
      "language_loss": 0.77575958,
      "learning_rate": 4.434197028803766e-09,
      "loss": 0.79825664,
      "num_input_tokens_seen": 175865975,
      "step": 8145,
      "time_per_iteration": 2.6177213191986084
    },
    {
      "auxiliary_loss_clip": 0.01326631,
      "auxiliary_loss_mlp": 0.01032407,
      "balance_loss_clip": 1.04319286,
      "balance_loss_mlp": 1.02567816,
      "epoch": 0.979498587146035,
      "flos": 23038742407680.0,
      "grad_norm": 2.1390658832230183,
      "language_loss": 0.82168937,
      "learning_rate": 4.3825056117805514e-09,
      "loss": 0.84527969,
      "num_input_tokens_seen": 175881860,
      "step": 8146,
      "time_per_iteration": 3.596791982650757
    },
    {
      "auxiliary_loss_clip": 0.01173867,
      "auxiliary_loss_mlp": 0.01030863,
      "balance_loss_clip": 1.04887772,
      "balance_loss_mlp": 1.02346587,
      "epoch": 0.979618830036674,
      "flos": 14318841951360.0,
      "grad_norm": 3.4817894796991773,
      "language_loss": 0.79234946,
      "learning_rate": 4.331116928344425e-09,
      "loss": 0.8143968,
      "num_input_tokens_seen": 175898175,
      "step": 8147,
      "time_per_iteration": 2.5755534172058105
    },
    {
      "auxiliary_loss_clip": 0.01279424,
      "auxiliary_loss_mlp": 0.02566897,
      "balance_loss_clip": 1.04761004,
      "balance_loss_mlp": 0.99990994,
      "epoch": 0.9797390729273132,
      "flos": 16727514215040.0,
      "grad_norm": 2.162891058742617,
      "language_loss": 0.62662053,
      "learning_rate": 4.28003098629115e-09,
      "loss": 0.66508371,
      "num_input_tokens_seen": 175914310,
      "step": 8148,
      "time_per_iteration": 2.609384298324585
    },
    {
      "auxiliary_loss_clip": 0.01319795,
      "auxiliary_loss_mlp": 0.01020046,
      "balance_loss_clip": 1.03893626,
      "balance_loss_mlp": 1.01318526,
      "epoch": 0.9798593158179523,
      "flos": 24532661986560.0,
      "grad_norm": 1.8532458841370534,
      "language_loss": 0.78794241,
      "learning_rate": 4.229247793370305e-09,
      "loss": 0.81134081,
      "num_input_tokens_seen": 175933435,
      "step": 8149,
      "time_per_iteration": 2.759444236755371
    },
    {
      "auxiliary_loss_clip": 0.01173106,
      "auxiliary_loss_mlp": 0.01028326,
      "balance_loss_clip": 1.04908299,
      "balance_loss_mlp": 1.02126312,
      "epoch": 0.9799795587085913,
      "flos": 27308808339840.0,
      "grad_norm": 1.9452679776619388,
      "language_loss": 0.70489651,
      "learning_rate": 4.178767357285951e-09,
      "loss": 0.72691089,
      "num_input_tokens_seen": 175955065,
      "step": 8150,
      "time_per_iteration": 2.6126294136047363
    },
    {
      "auxiliary_loss_clip": 0.01224681,
      "auxiliary_loss_mlp": 0.02566124,
      "balance_loss_clip": 1.04816163,
      "balance_loss_mlp": 0.99990237,
      "epoch": 0.9800998015992305,
      "flos": 26286575184000.0,
      "grad_norm": 2.719879798257293,
      "language_loss": 0.71867979,
      "learning_rate": 4.128589685695516e-09,
      "loss": 0.75658786,
      "num_input_tokens_seen": 175975490,
      "step": 8151,
      "time_per_iteration": 2.713348150253296
    },
    {
      "auxiliary_loss_clip": 0.01172446,
      "auxiliary_loss_mlp": 0.01027975,
      "balance_loss_clip": 1.0489521,
      "balance_loss_mlp": 1.02073312,
      "epoch": 0.9802200444898695,
      "flos": 16723635546240.0,
      "grad_norm": 1.8883844856498497,
      "language_loss": 0.84246379,
      "learning_rate": 4.078714786211135e-09,
      "loss": 0.86446798,
      "num_input_tokens_seen": 175991340,
      "step": 8152,
      "time_per_iteration": 2.515617609024048
    },
    {
      "auxiliary_loss_clip": 0.01218018,
      "auxiliary_loss_mlp": 0.0102296,
      "balance_loss_clip": 1.04706669,
      "balance_loss_mlp": 1.0164963,
      "epoch": 0.9803402873805086,
      "flos": 24900459298560.0,
      "grad_norm": 2.20408790809063,
      "language_loss": 0.76812756,
      "learning_rate": 4.029142666398977e-09,
      "loss": 0.79053736,
      "num_input_tokens_seen": 176011505,
      "step": 8153,
      "time_per_iteration": 2.7215042114257812
    },
    {
      "auxiliary_loss_clip": 0.01172649,
      "auxiliary_loss_mlp": 0.01022216,
      "balance_loss_clip": 1.05108571,
      "balance_loss_mlp": 1.01610935,
      "epoch": 0.9804605302711478,
      "flos": 22564937082240.0,
      "grad_norm": 2.090763925885501,
      "language_loss": 0.80400276,
      "learning_rate": 3.979873333778805e-09,
      "loss": 0.8259514,
      "num_input_tokens_seen": 176029680,
      "step": 8154,
      "time_per_iteration": 2.5991363525390625
    },
    {
      "auxiliary_loss_clip": 0.01276492,
      "auxiliary_loss_mlp": 0.01028251,
      "balance_loss_clip": 1.04729414,
      "balance_loss_mlp": 1.02103329,
      "epoch": 0.9805807731617868,
      "flos": 38905368382080.0,
      "grad_norm": 1.979390765325647,
      "language_loss": 0.74103242,
      "learning_rate": 3.930906795824862e-09,
      "loss": 0.76407987,
      "num_input_tokens_seen": 176050355,
      "step": 8155,
      "time_per_iteration": 2.7860875129699707
    },
    {
      "auxiliary_loss_clip": 0.01217515,
      "auxiliary_loss_mlp": 0.01025229,
      "balance_loss_clip": 1.04692578,
      "balance_loss_mlp": 1.01848149,
      "epoch": 0.9807010160524259,
      "flos": 17821999578240.0,
      "grad_norm": 1.9867082688524853,
      "language_loss": 0.765136,
      "learning_rate": 3.882243059965207e-09,
      "loss": 0.78756344,
      "num_input_tokens_seen": 176068070,
      "step": 8156,
      "time_per_iteration": 2.5889415740966797
    },
    {
      "auxiliary_loss_clip": 0.012192,
      "auxiliary_loss_mlp": 0.01023844,
      "balance_loss_clip": 1.04529333,
      "balance_loss_mlp": 1.01633632,
      "epoch": 0.980821258943065,
      "flos": 13552975140480.0,
      "grad_norm": 3.5832252652977896,
      "language_loss": 0.65992737,
      "learning_rate": 3.833882133582156e-09,
      "loss": 0.68235779,
      "num_input_tokens_seen": 176083730,
      "step": 8157,
      "time_per_iteration": 2.614975690841675
    },
    {
      "auxiliary_loss_clip": 0.01224633,
      "auxiliary_loss_mlp": 0.01026221,
      "balance_loss_clip": 1.04654217,
      "balance_loss_mlp": 1.01933372,
      "epoch": 0.9809415018337041,
      "flos": 21689794120320.0,
      "grad_norm": 2.5789150738741964,
      "language_loss": 0.78288484,
      "learning_rate": 3.785824024012285e-09,
      "loss": 0.80539334,
      "num_input_tokens_seen": 176102730,
      "step": 8158,
      "time_per_iteration": 2.615952491760254
    },
    {
      "auxiliary_loss_clip": 0.01265092,
      "auxiliary_loss_mlp": 0.01022478,
      "balance_loss_clip": 1.04788446,
      "balance_loss_mlp": 1.01590693,
      "epoch": 0.9810617447243432,
      "flos": 23294857357440.0,
      "grad_norm": 1.7257324468670343,
      "language_loss": 0.78469872,
      "learning_rate": 3.738068738545541e-09,
      "loss": 0.80757445,
      "num_input_tokens_seen": 176121815,
      "step": 8159,
      "time_per_iteration": 2.6631081104278564
    },
    {
      "auxiliary_loss_clip": 0.01226036,
      "auxiliary_loss_mlp": 0.01025912,
      "balance_loss_clip": 1.04758382,
      "balance_loss_mlp": 1.01888514,
      "epoch": 0.9811819876149822,
      "flos": 18332038748160.0,
      "grad_norm": 2.3251792899182764,
      "language_loss": 0.78414917,
      "learning_rate": 3.6906162844265733e-09,
      "loss": 0.80666864,
      "num_input_tokens_seen": 176138900,
      "step": 8160,
      "time_per_iteration": 2.6103334426879883
    },
    {
      "auxiliary_loss_clip": 0.01269131,
      "auxiliary_loss_mlp": 0.01024699,
      "balance_loss_clip": 1.0437268,
      "balance_loss_mlp": 1.01753163,
      "epoch": 0.9813022305056214,
      "flos": 22601961025920.0,
      "grad_norm": 1.7736737430572151,
      "language_loss": 0.70585728,
      "learning_rate": 3.643466668853845e-09,
      "loss": 0.72879553,
      "num_input_tokens_seen": 176156925,
      "step": 8161,
      "time_per_iteration": 2.6477913856506348
    },
    {
      "auxiliary_loss_clip": 0.01272676,
      "auxiliary_loss_mlp": 0.01025035,
      "balance_loss_clip": 1.04368806,
      "balance_loss_mlp": 1.01808524,
      "epoch": 0.9814224733962604,
      "flos": 25413335642880.0,
      "grad_norm": 2.8394578556735595,
      "language_loss": 0.75463331,
      "learning_rate": 3.59661989898008e-09,
      "loss": 0.77761042,
      "num_input_tokens_seen": 176177980,
      "step": 8162,
      "time_per_iteration": 2.716294527053833
    },
    {
      "auxiliary_loss_clip": 0.01312619,
      "auxiliary_loss_mlp": 0.01025384,
      "balance_loss_clip": 1.04467058,
      "balance_loss_mlp": 1.01879442,
      "epoch": 0.9815427162868995,
      "flos": 25007185584000.0,
      "grad_norm": 2.0459836969205134,
      "language_loss": 0.76412016,
      "learning_rate": 3.5500759819115934e-09,
      "loss": 0.78750014,
      "num_input_tokens_seen": 176198345,
      "step": 8163,
      "time_per_iteration": 3.570202112197876
    },
    {
      "auxiliary_loss_clip": 0.01173495,
      "auxiliary_loss_mlp": 0.01031447,
      "balance_loss_clip": 1.04988694,
      "balance_loss_mlp": 1.02487254,
      "epoch": 0.9816629591775387,
      "flos": 20662604887680.0,
      "grad_norm": 2.020009534367411,
      "language_loss": 0.81205851,
      "learning_rate": 3.5038349247094034e-09,
      "loss": 0.83410788,
      "num_input_tokens_seen": 176215605,
      "step": 8164,
      "time_per_iteration": 3.4798805713653564
    },
    {
      "auxiliary_loss_clip": 0.01273208,
      "auxiliary_loss_mlp": 0.01025981,
      "balance_loss_clip": 1.04458666,
      "balance_loss_mlp": 1.01925468,
      "epoch": 0.9817832020681777,
      "flos": 17712220636800.0,
      "grad_norm": 6.262593811881118,
      "language_loss": 0.77390009,
      "learning_rate": 3.4578967343878994e-09,
      "loss": 0.79689199,
      "num_input_tokens_seen": 176231810,
      "step": 8165,
      "time_per_iteration": 2.6599955558776855
    },
    {
      "auxiliary_loss_clip": 0.01269929,
      "auxiliary_loss_mlp": 0.01022838,
      "balance_loss_clip": 1.04697299,
      "balance_loss_mlp": 1.01598358,
      "epoch": 0.9819034449588168,
      "flos": 22530032040960.0,
      "grad_norm": 2.301382984170266,
      "language_loss": 0.80930525,
      "learning_rate": 3.4122614179161733e-09,
      "loss": 0.83223295,
      "num_input_tokens_seen": 176251770,
      "step": 8166,
      "time_per_iteration": 2.638740301132202
    },
    {
      "auxiliary_loss_clip": 0.0131764,
      "auxiliary_loss_mlp": 0.01024889,
      "balance_loss_clip": 1.04340744,
      "balance_loss_mlp": 1.01876795,
      "epoch": 0.9820236878494559,
      "flos": 20011221699840.0,
      "grad_norm": 2.184244554571534,
      "language_loss": 0.78485596,
      "learning_rate": 3.36692898221691e-09,
      "loss": 0.8082813,
      "num_input_tokens_seen": 176270135,
      "step": 8167,
      "time_per_iteration": 3.5635814666748047
    },
    {
      "auxiliary_loss_clip": 0.01224919,
      "auxiliary_loss_mlp": 0.01018697,
      "balance_loss_clip": 1.04802823,
      "balance_loss_mlp": 1.01261425,
      "epoch": 0.982143930740095,
      "flos": 18807316531200.0,
      "grad_norm": 1.9238882069283803,
      "language_loss": 0.73705482,
      "learning_rate": 3.3218994341668305e-09,
      "loss": 0.75949097,
      "num_input_tokens_seen": 176289065,
      "step": 8168,
      "time_per_iteration": 2.5854713916778564
    },
    {
      "auxiliary_loss_clip": 0.01172377,
      "auxiliary_loss_mlp": 0.01020588,
      "balance_loss_clip": 1.05139637,
      "balance_loss_mlp": 1.01421309,
      "epoch": 0.982264173630734,
      "flos": 26578026138240.0,
      "grad_norm": 1.669297662324535,
      "language_loss": 0.75424314,
      "learning_rate": 3.2771727805971373e-09,
      "loss": 0.77617282,
      "num_input_tokens_seen": 176310450,
      "step": 8169,
      "time_per_iteration": 2.6621246337890625
    },
    {
      "auxiliary_loss_clip": 0.01367737,
      "auxiliary_loss_mlp": 0.01023575,
      "balance_loss_clip": 1.03941703,
      "balance_loss_mlp": 1.01676238,
      "epoch": 0.9823844165213732,
      "flos": 22014462176640.0,
      "grad_norm": 1.7351309740527119,
      "language_loss": 0.771083,
      "learning_rate": 3.232749028292847e-09,
      "loss": 0.79499614,
      "num_input_tokens_seen": 176327415,
      "step": 8170,
      "time_per_iteration": 2.7080976963043213
    },
    {
      "auxiliary_loss_clip": 0.01172868,
      "auxiliary_loss_mlp": 0.01028055,
      "balance_loss_clip": 1.04773366,
      "balance_loss_mlp": 1.0207442,
      "epoch": 0.9825046594120123,
      "flos": 21908166854400.0,
      "grad_norm": 1.7804051480996195,
      "language_loss": 0.88193125,
      "learning_rate": 3.188628183992792e-09,
      "loss": 0.9039405,
      "num_input_tokens_seen": 176347680,
      "step": 8171,
      "time_per_iteration": 2.5987253189086914
    },
    {
      "auxiliary_loss_clip": 0.01116465,
      "auxiliary_loss_mlp": 0.01000301,
      "balance_loss_clip": 1.00557315,
      "balance_loss_mlp": 0.99941874,
      "epoch": 0.9826249023026513,
      "flos": 59494610718720.0,
      "grad_norm": 0.7368035878670971,
      "language_loss": 0.62478185,
      "learning_rate": 3.1448102543902844e-09,
      "loss": 0.6459496,
      "num_input_tokens_seen": 176411595,
      "step": 8172,
      "time_per_iteration": 4.1643383502960205
    },
    {
      "auxiliary_loss_clip": 0.01268924,
      "auxiliary_loss_mlp": 0.01025972,
      "balance_loss_clip": 1.04656959,
      "balance_loss_mlp": 1.01934719,
      "epoch": 0.9827451451932905,
      "flos": 16071031296000.0,
      "grad_norm": 2.193108206931262,
      "language_loss": 0.67748189,
      "learning_rate": 3.1012952461324515e-09,
      "loss": 0.70043087,
      "num_input_tokens_seen": 176430570,
      "step": 8173,
      "time_per_iteration": 2.6003880500793457
    },
    {
      "auxiliary_loss_clip": 0.01218764,
      "auxiliary_loss_mlp": 0.01025701,
      "balance_loss_clip": 1.04982734,
      "balance_loss_mlp": 1.01849174,
      "epoch": 0.9828653880839295,
      "flos": 20262775622400.0,
      "grad_norm": 2.119283976370514,
      "language_loss": 0.73960018,
      "learning_rate": 3.0580831658204575e-09,
      "loss": 0.76204479,
      "num_input_tokens_seen": 176448150,
      "step": 8174,
      "time_per_iteration": 2.596463918685913
    },
    {
      "auxiliary_loss_clip": 0.01221367,
      "auxiliary_loss_mlp": 0.01023407,
      "balance_loss_clip": 1.04902327,
      "balance_loss_mlp": 1.01656413,
      "epoch": 0.9829856309745686,
      "flos": 21616141282560.0,
      "grad_norm": 2.9082518850920325,
      "language_loss": 0.77834213,
      "learning_rate": 3.015174020009281e-09,
      "loss": 0.80078983,
      "num_input_tokens_seen": 176467475,
      "step": 8175,
      "time_per_iteration": 2.5999507904052734
    },
    {
      "auxiliary_loss_clip": 0.01325648,
      "auxiliary_loss_mlp": 0.01028973,
      "balance_loss_clip": 1.04312086,
      "balance_loss_mlp": 1.02238381,
      "epoch": 0.9831058738652078,
      "flos": 23764209396480.0,
      "grad_norm": 1.8113715005267594,
      "language_loss": 0.74932551,
      "learning_rate": 2.9725678152086043e-09,
      "loss": 0.77287173,
      "num_input_tokens_seen": 176486045,
      "step": 8176,
      "time_per_iteration": 2.6963109970092773
    },
    {
      "auxiliary_loss_clip": 0.01263062,
      "auxiliary_loss_mlp": 0.01023851,
      "balance_loss_clip": 1.04280055,
      "balance_loss_mlp": 1.01660585,
      "epoch": 0.9832261167558468,
      "flos": 11320911072000.0,
      "grad_norm": 2.8229668881490704,
      "language_loss": 0.82422501,
      "learning_rate": 2.930264557881257e-09,
      "loss": 0.84709418,
      "num_input_tokens_seen": 176501230,
      "step": 8177,
      "time_per_iteration": 2.648855686187744
    },
    {
      "auxiliary_loss_clip": 0.01059521,
      "auxiliary_loss_mlp": 0.01000028,
      "balance_loss_clip": 1.00561905,
      "balance_loss_mlp": 0.99919969,
      "epoch": 0.9833463596464859,
      "flos": 60000304343040.0,
      "grad_norm": 0.8324063270151959,
      "language_loss": 0.58128452,
      "learning_rate": 2.8882642544452163e-09,
      "loss": 0.60188007,
      "num_input_tokens_seen": 176565955,
      "step": 8178,
      "time_per_iteration": 3.167478561401367
    },
    {
      "auxiliary_loss_clip": 0.01265823,
      "auxiliary_loss_mlp": 0.01027173,
      "balance_loss_clip": 1.04411542,
      "balance_loss_mlp": 1.02057481,
      "epoch": 0.983466602537125,
      "flos": 13626699805440.0,
      "grad_norm": 3.1350885250536518,
      "language_loss": 0.74560237,
      "learning_rate": 2.8465669112716083e-09,
      "loss": 0.7685324,
      "num_input_tokens_seen": 176583480,
      "step": 8179,
      "time_per_iteration": 2.6301181316375732
    },
    {
      "auxiliary_loss_clip": 0.01225502,
      "auxiliary_loss_mlp": 0.0256667,
      "balance_loss_clip": 1.04691458,
      "balance_loss_mlp": 0.99991322,
      "epoch": 0.9835868454277641,
      "flos": 22926844563840.0,
      "grad_norm": 1.9440262614287958,
      "language_loss": 0.76735222,
      "learning_rate": 2.8051725346858177e-09,
      "loss": 0.80527395,
      "num_input_tokens_seen": 176603740,
      "step": 8180,
      "time_per_iteration": 2.6243069171905518
    },
    {
      "auxiliary_loss_clip": 0.01171122,
      "auxiliary_loss_mlp": 0.01022736,
      "balance_loss_clip": 1.04537868,
      "balance_loss_mlp": 1.01607251,
      "epoch": 0.9837070883184031,
      "flos": 27673409341440.0,
      "grad_norm": 2.1974190148497064,
      "language_loss": 0.71184713,
      "learning_rate": 2.7640811309674883e-09,
      "loss": 0.73378569,
      "num_input_tokens_seen": 176623240,
      "step": 8181,
      "time_per_iteration": 2.6329071521759033
    },
    {
      "auxiliary_loss_clip": 0.01314847,
      "auxiliary_loss_mlp": 0.01027026,
      "balance_loss_clip": 1.04355681,
      "balance_loss_mlp": 1.02018595,
      "epoch": 0.9838273312090423,
      "flos": 29241951425280.0,
      "grad_norm": 1.688612520813857,
      "language_loss": 0.80811161,
      "learning_rate": 2.7232927063498557e-09,
      "loss": 0.83153033,
      "num_input_tokens_seen": 176643615,
      "step": 8182,
      "time_per_iteration": 2.7302498817443848
    },
    {
      "auxiliary_loss_clip": 0.01221585,
      "auxiliary_loss_mlp": 0.0102678,
      "balance_loss_clip": 1.04752254,
      "balance_loss_mlp": 1.0196867,
      "epoch": 0.9839475740996814,
      "flos": 40110207304320.0,
      "grad_norm": 5.746203946027384,
      "language_loss": 0.69310331,
      "learning_rate": 2.682807267020859e-09,
      "loss": 0.71558702,
      "num_input_tokens_seen": 176666375,
      "step": 8183,
      "time_per_iteration": 2.773212194442749
    },
    {
      "auxiliary_loss_clip": 0.01219006,
      "auxiliary_loss_mlp": 0.01027458,
      "balance_loss_clip": 1.04577994,
      "balance_loss_mlp": 1.02093983,
      "epoch": 0.9840678169903204,
      "flos": 24169389788160.0,
      "grad_norm": 1.7319794494844207,
      "language_loss": 0.62761998,
      "learning_rate": 2.642624819121808e-09,
      "loss": 0.65008461,
      "num_input_tokens_seen": 176686525,
      "step": 8184,
      "time_per_iteration": 2.6370620727539062
    },
    {
      "auxiliary_loss_clip": 0.01271645,
      "auxiliary_loss_mlp": 0.01024344,
      "balance_loss_clip": 1.04804611,
      "balance_loss_mlp": 1.01764441,
      "epoch": 0.9841880598809596,
      "flos": 14684484447360.0,
      "grad_norm": 1.9885309791199768,
      "language_loss": 0.61712414,
      "learning_rate": 2.6027453687487154e-09,
      "loss": 0.64008403,
      "num_input_tokens_seen": 176703615,
      "step": 8185,
      "time_per_iteration": 2.6404874324798584
    },
    {
      "auxiliary_loss_clip": 0.01269643,
      "auxiliary_loss_mlp": 0.01027738,
      "balance_loss_clip": 1.04666805,
      "balance_loss_mlp": 1.02087498,
      "epoch": 0.9843083027715986,
      "flos": 22344768668160.0,
      "grad_norm": 2.315364890694146,
      "language_loss": 0.53986239,
      "learning_rate": 2.5631689219509643e-09,
      "loss": 0.56283623,
      "num_input_tokens_seen": 176722295,
      "step": 8186,
      "time_per_iteration": 2.628795623779297
    },
    {
      "auxiliary_loss_clip": 0.01167284,
      "auxiliary_loss_mlp": 0.01023346,
      "balance_loss_clip": 1.04695392,
      "balance_loss_mlp": 1.01718307,
      "epoch": 0.9844285456622377,
      "flos": 21800111765760.0,
      "grad_norm": 1.7752884578119619,
      "language_loss": 0.83641064,
      "learning_rate": 2.523895484732197e-09,
      "loss": 0.85831696,
      "num_input_tokens_seen": 176741750,
      "step": 8187,
      "time_per_iteration": 2.702096939086914
    },
    {
      "auxiliary_loss_clip": 0.0122705,
      "auxiliary_loss_mlp": 0.01023417,
      "balance_loss_clip": 1.04641771,
      "balance_loss_mlp": 1.01613355,
      "epoch": 0.9845487885528769,
      "flos": 18035380321920.0,
      "grad_norm": 2.1843653065248247,
      "language_loss": 0.74768573,
      "learning_rate": 2.4849250630505357e-09,
      "loss": 0.77019036,
      "num_input_tokens_seen": 176759995,
      "step": 8188,
      "time_per_iteration": 2.6488585472106934
    },
    {
      "auxiliary_loss_clip": 0.01458816,
      "auxiliary_loss_mlp": 0.0102396,
      "balance_loss_clip": 1.03801394,
      "balance_loss_mlp": 1.0173111,
      "epoch": 0.9846690314435159,
      "flos": 25228610974080.0,
      "grad_norm": 1.7948988955524519,
      "language_loss": 0.73444366,
      "learning_rate": 2.4462576628172528e-09,
      "loss": 0.75927138,
      "num_input_tokens_seen": 176778625,
      "step": 8189,
      "time_per_iteration": 3.644676923751831
    },
    {
      "auxiliary_loss_clip": 0.01221307,
      "auxiliary_loss_mlp": 0.01028987,
      "balance_loss_clip": 1.04898906,
      "balance_loss_mlp": 1.02226388,
      "epoch": 0.984789274334155,
      "flos": 18552171248640.0,
      "grad_norm": 3.993905874798312,
      "language_loss": 0.73859215,
      "learning_rate": 2.407893289898766e-09,
      "loss": 0.76109505,
      "num_input_tokens_seen": 176797655,
      "step": 8190,
      "time_per_iteration": 3.5194203853607178
    },
    {
      "auxiliary_loss_clip": 0.01314452,
      "auxiliary_loss_mlp": 0.0102452,
      "balance_loss_clip": 1.04283881,
      "balance_loss_mlp": 1.01748919,
      "epoch": 0.984909517224794,
      "flos": 27345437233920.0,
      "grad_norm": 1.9002799080798765,
      "language_loss": 0.84223616,
      "learning_rate": 2.3698319501144202e-09,
      "loss": 0.86562586,
      "num_input_tokens_seen": 176818640,
      "step": 8191,
      "time_per_iteration": 2.688042402267456
    },
    {
      "auxiliary_loss_clip": 0.01127644,
      "auxiliary_loss_mlp": 0.01027416,
      "balance_loss_clip": 1.04848576,
      "balance_loss_mlp": 1.02038288,
      "epoch": 0.9850297601154332,
      "flos": 18734058743040.0,
      "grad_norm": 2.6367740969489186,
      "language_loss": 0.73392749,
      "learning_rate": 2.3320736492382644e-09,
      "loss": 0.75547802,
      "num_input_tokens_seen": 176837475,
      "step": 8192,
      "time_per_iteration": 2.6411659717559814
    },
    {
      "auxiliary_loss_clip": 0.01171305,
      "auxiliary_loss_mlp": 0.01024751,
      "balance_loss_clip": 1.04954433,
      "balance_loss_mlp": 1.01810467,
      "epoch": 0.9851500030060723,
      "flos": 22308247514880.0,
      "grad_norm": 1.963331951677283,
      "language_loss": 0.68415773,
      "learning_rate": 2.29461839299816e-09,
      "loss": 0.70611823,
      "num_input_tokens_seen": 176857190,
      "step": 8193,
      "time_per_iteration": 2.5515103340148926
    },
    {
      "auxiliary_loss_clip": 0.01222865,
      "auxiliary_loss_mlp": 0.01022837,
      "balance_loss_clip": 1.04580092,
      "balance_loss_mlp": 1.0162003,
      "epoch": 0.9852702458967113,
      "flos": 26353691746560.0,
      "grad_norm": 1.6802903589310003,
      "language_loss": 0.80061376,
      "learning_rate": 2.257466187076229e-09,
      "loss": 0.82307082,
      "num_input_tokens_seen": 176876395,
      "step": 8194,
      "time_per_iteration": 3.6386654376983643
    },
    {
      "auxiliary_loss_clip": 0.01227516,
      "auxiliary_loss_mlp": 0.0256735,
      "balance_loss_clip": 1.04685593,
      "balance_loss_mlp": 0.99988729,
      "epoch": 0.9853904887873505,
      "flos": 20883599314560.0,
      "grad_norm": 1.8735699581959333,
      "language_loss": 0.71602541,
      "learning_rate": 2.2206170371081854e-09,
      "loss": 0.75397408,
      "num_input_tokens_seen": 176894980,
      "step": 8195,
      "time_per_iteration": 2.6481716632843018
    },
    {
      "auxiliary_loss_clip": 0.01277961,
      "auxiliary_loss_mlp": 0.01028542,
      "balance_loss_clip": 1.04489827,
      "balance_loss_mlp": 1.02153552,
      "epoch": 0.9855107316779895,
      "flos": 25263444188160.0,
      "grad_norm": 1.6996012714802455,
      "language_loss": 0.84916824,
      "learning_rate": 2.1840709486842247e-09,
      "loss": 0.87223327,
      "num_input_tokens_seen": 176914600,
      "step": 8196,
      "time_per_iteration": 2.683431386947632
    },
    {
      "auxiliary_loss_clip": 0.01266676,
      "auxiliary_loss_mlp": 0.0102661,
      "balance_loss_clip": 1.04505968,
      "balance_loss_mlp": 1.0197556,
      "epoch": 0.9856309745686286,
      "flos": 19062102677760.0,
      "grad_norm": 2.3295398535460765,
      "language_loss": 0.79216373,
      "learning_rate": 2.1478279273481335e-09,
      "loss": 0.81509662,
      "num_input_tokens_seen": 176933085,
      "step": 8197,
      "time_per_iteration": 2.6710562705993652
    },
    {
      "auxiliary_loss_clip": 0.01219722,
      "auxiliary_loss_mlp": 0.01026353,
      "balance_loss_clip": 1.04716134,
      "balance_loss_mlp": 1.01950455,
      "epoch": 0.9857512174592677,
      "flos": 34130758060800.0,
      "grad_norm": 2.413742378614482,
      "language_loss": 0.804075,
      "learning_rate": 2.1118879785981815e-09,
      "loss": 0.82653582,
      "num_input_tokens_seen": 176953225,
      "step": 8198,
      "time_per_iteration": 3.593425989151001
    },
    {
      "auxiliary_loss_clip": 0.01269376,
      "auxiliary_loss_mlp": 0.01021922,
      "balance_loss_clip": 1.04597378,
      "balance_loss_mlp": 1.01586056,
      "epoch": 0.9858714603499068,
      "flos": 25994693266560.0,
      "grad_norm": 1.7278513234268609,
      "language_loss": 0.79426163,
      "learning_rate": 2.0762511078862288e-09,
      "loss": 0.81717467,
      "num_input_tokens_seen": 176973570,
      "step": 8199,
      "time_per_iteration": 2.7478859424591064
    },
    {
      "auxiliary_loss_clip": 0.01177046,
      "auxiliary_loss_mlp": 0.01024084,
      "balance_loss_clip": 1.04583216,
      "balance_loss_mlp": 1.01760209,
      "epoch": 0.9859917032405459,
      "flos": 23696230907520.0,
      "grad_norm": 1.8937516543248345,
      "language_loss": 0.65087193,
      "learning_rate": 2.0409173206186183e-09,
      "loss": 0.67288327,
      "num_input_tokens_seen": 176992810,
      "step": 8200,
      "time_per_iteration": 2.682705879211426
    },
    {
      "auxiliary_loss_clip": 0.01316087,
      "auxiliary_loss_mlp": 0.01024378,
      "balance_loss_clip": 1.04459035,
      "balance_loss_mlp": 1.01791143,
      "epoch": 0.986111946131185,
      "flos": 19938287134080.0,
      "grad_norm": 1.9343675961697597,
      "language_loss": 0.87153578,
      "learning_rate": 2.0058866221550617e-09,
      "loss": 0.89494038,
      "num_input_tokens_seen": 177011050,
      "step": 8201,
      "time_per_iteration": 2.669773817062378
    },
    {
      "auxiliary_loss_clip": 0.01173268,
      "auxiliary_loss_mlp": 0.0102707,
      "balance_loss_clip": 1.04786348,
      "balance_loss_mlp": 1.02034378,
      "epoch": 0.9862321890218241,
      "flos": 19828831415040.0,
      "grad_norm": 2.0397143325119758,
      "language_loss": 0.74986255,
      "learning_rate": 1.971159017809976e-09,
      "loss": 0.7718659,
      "num_input_tokens_seen": 177029340,
      "step": 8202,
      "time_per_iteration": 2.594998598098755
    },
    {
      "auxiliary_loss_clip": 0.01224286,
      "auxiliary_loss_mlp": 0.01027091,
      "balance_loss_clip": 1.05027461,
      "balance_loss_mlp": 1.02037311,
      "epoch": 0.9863524319124631,
      "flos": 21652051904640.0,
      "grad_norm": 2.3700589870879685,
      "language_loss": 0.77780116,
      "learning_rate": 1.93673451285159e-09,
      "loss": 0.8003149,
      "num_input_tokens_seen": 177048390,
      "step": 8203,
      "time_per_iteration": 2.5950088500976562
    },
    {
      "auxiliary_loss_clip": 0.01174185,
      "auxiliary_loss_mlp": 0.0100122,
      "balance_loss_clip": 1.00547957,
      "balance_loss_mlp": 1.00028384,
      "epoch": 0.9864726748031023,
      "flos": 52769977920000.0,
      "grad_norm": 0.7312513764108556,
      "language_loss": 0.56499016,
      "learning_rate": 1.9026131125019495e-09,
      "loss": 0.58674431,
      "num_input_tokens_seen": 177105760,
      "step": 8204,
      "time_per_iteration": 3.16135311126709
    },
    {
      "auxiliary_loss_clip": 0.01217334,
      "auxiliary_loss_mlp": 0.01023858,
      "balance_loss_clip": 1.04685938,
      "balance_loss_mlp": 1.01694369,
      "epoch": 0.9865929176937414,
      "flos": 23364631526400.0,
      "grad_norm": 1.7939022631613357,
      "language_loss": 0.87034291,
      "learning_rate": 1.8687948219371363e-09,
      "loss": 0.89275479,
      "num_input_tokens_seen": 177124985,
      "step": 8205,
      "time_per_iteration": 2.60935115814209
    },
    {
      "auxiliary_loss_clip": 0.01175308,
      "auxiliary_loss_mlp": 0.01023182,
      "balance_loss_clip": 1.04719067,
      "balance_loss_mlp": 1.015764,
      "epoch": 0.9867131605843804,
      "flos": 21616679986560.0,
      "grad_norm": 2.4334043267458174,
      "language_loss": 0.88678014,
      "learning_rate": 1.835279646287491e-09,
      "loss": 0.90876502,
      "num_input_tokens_seen": 177142995,
      "step": 8206,
      "time_per_iteration": 2.7042205333709717
    },
    {
      "auxiliary_loss_clip": 0.01230518,
      "auxiliary_loss_mlp": 0.0103291,
      "balance_loss_clip": 1.05004013,
      "balance_loss_mlp": 1.02509594,
      "epoch": 0.9868334034750196,
      "flos": 22271403139200.0,
      "grad_norm": 1.9427171900058084,
      "language_loss": 0.76774049,
      "learning_rate": 1.8020675906371685e-09,
      "loss": 0.79037476,
      "num_input_tokens_seen": 177162390,
      "step": 8207,
      "time_per_iteration": 2.5905849933624268
    },
    {
      "auxiliary_loss_clip": 0.01366996,
      "auxiliary_loss_mlp": 0.010222,
      "balance_loss_clip": 1.04358029,
      "balance_loss_mlp": 1.01529825,
      "epoch": 0.9869536463656586,
      "flos": 25809573548160.0,
      "grad_norm": 2.313334673613612,
      "language_loss": 0.75281304,
      "learning_rate": 1.7691586600243612e-09,
      "loss": 0.77670503,
      "num_input_tokens_seen": 177181290,
      "step": 8208,
      "time_per_iteration": 2.7866811752319336
    },
    {
      "auxiliary_loss_clip": 0.01273573,
      "auxiliary_loss_mlp": 0.0102301,
      "balance_loss_clip": 1.04987645,
      "balance_loss_mlp": 1.01612866,
      "epoch": 0.9870738892562977,
      "flos": 16398500613120.0,
      "grad_norm": 2.7345257778663368,
      "language_loss": 0.86862069,
      "learning_rate": 1.7365528594415202e-09,
      "loss": 0.89158654,
      "num_input_tokens_seen": 177195360,
      "step": 8209,
      "time_per_iteration": 2.604391098022461
    },
    {
      "auxiliary_loss_clip": 0.01227697,
      "auxiliary_loss_mlp": 0.02566965,
      "balance_loss_clip": 1.04884052,
      "balance_loss_mlp": 0.9998911,
      "epoch": 0.9871941321469369,
      "flos": 35481358373760.0,
      "grad_norm": 2.9401786672086727,
      "language_loss": 0.67725205,
      "learning_rate": 1.7042501938346888e-09,
      "loss": 0.71519864,
      "num_input_tokens_seen": 177218090,
      "step": 8210,
      "time_per_iteration": 2.7910826206207275
    },
    {
      "auxiliary_loss_clip": 0.01263259,
      "auxiliary_loss_mlp": 0.01021993,
      "balance_loss_clip": 1.03987753,
      "balance_loss_mlp": 1.01529956,
      "epoch": 0.9873143750375759,
      "flos": 21434217874560.0,
      "grad_norm": 1.8956955233090456,
      "language_loss": 0.76597416,
      "learning_rate": 1.6722506681043913e-09,
      "loss": 0.7888267,
      "num_input_tokens_seen": 177237050,
      "step": 8211,
      "time_per_iteration": 2.629702091217041
    },
    {
      "auxiliary_loss_clip": 0.0127307,
      "auxiliary_loss_mlp": 0.01026078,
      "balance_loss_clip": 1.04553747,
      "balance_loss_mlp": 1.01866317,
      "epoch": 0.987434617928215,
      "flos": 16326499800960.0,
      "grad_norm": 2.1066789811790425,
      "language_loss": 0.69506222,
      "learning_rate": 1.640554287104745e-09,
      "loss": 0.7180537,
      "num_input_tokens_seen": 177255325,
      "step": 8212,
      "time_per_iteration": 2.680727005004883
    },
    {
      "auxiliary_loss_clip": 0.01324759,
      "auxiliary_loss_mlp": 0.01025293,
      "balance_loss_clip": 1.04134464,
      "balance_loss_mlp": 1.01781297,
      "epoch": 0.9875548608188541,
      "flos": 17851984456320.0,
      "grad_norm": 4.076527286907154,
      "language_loss": 0.80271655,
      "learning_rate": 1.609161055644348e-09,
      "loss": 0.82621706,
      "num_input_tokens_seen": 177271250,
      "step": 8213,
      "time_per_iteration": 2.649904251098633
    },
    {
      "auxiliary_loss_clip": 0.01227216,
      "auxiliary_loss_mlp": 0.0102863,
      "balance_loss_clip": 1.04597056,
      "balance_loss_mlp": 1.02095246,
      "epoch": 0.9876751037094932,
      "flos": 26132876887680.0,
      "grad_norm": 2.507946621713477,
      "language_loss": 0.68489987,
      "learning_rate": 1.5780709784849467e-09,
      "loss": 0.70745826,
      "num_input_tokens_seen": 177288270,
      "step": 8214,
      "time_per_iteration": 3.5112483501434326
    },
    {
      "auxiliary_loss_clip": 0.01420342,
      "auxiliary_loss_mlp": 0.01028844,
      "balance_loss_clip": 1.04600859,
      "balance_loss_mlp": 1.02177227,
      "epoch": 0.9877953466001322,
      "flos": 15991344973440.0,
      "grad_norm": 1.89778637619767,
      "language_loss": 0.82993495,
      "learning_rate": 1.5472840603436565e-09,
      "loss": 0.85442686,
      "num_input_tokens_seen": 177305500,
      "step": 8215,
      "time_per_iteration": 2.740823268890381
    },
    {
      "auxiliary_loss_clip": 0.01171054,
      "auxiliary_loss_mlp": 0.01024151,
      "balance_loss_clip": 1.04617429,
      "balance_loss_mlp": 1.01728106,
      "epoch": 0.9879155894907714,
      "flos": 18806777827200.0,
      "grad_norm": 2.398418260037538,
      "language_loss": 0.78281039,
      "learning_rate": 1.5168003058900757e-09,
      "loss": 0.80476242,
      "num_input_tokens_seen": 177323500,
      "step": 8216,
      "time_per_iteration": 3.566293239593506
    },
    {
      "auxiliary_loss_clip": 0.01317684,
      "auxiliary_loss_mlp": 0.0102703,
      "balance_loss_clip": 1.04302907,
      "balance_loss_mlp": 1.02063465,
      "epoch": 0.9880358323814105,
      "flos": 22382044007040.0,
      "grad_norm": 2.100627483710409,
      "language_loss": 0.92048597,
      "learning_rate": 1.4866197197491715e-09,
      "loss": 0.94393313,
      "num_input_tokens_seen": 177342860,
      "step": 8217,
      "time_per_iteration": 2.7280917167663574
    },
    {
      "auxiliary_loss_clip": 0.01226903,
      "auxiliary_loss_mlp": 0.02570217,
      "balance_loss_clip": 1.04804778,
      "balance_loss_mlp": 0.99991781,
      "epoch": 0.9881560752720495,
      "flos": 15668831733120.0,
      "grad_norm": 3.985760403283924,
      "language_loss": 0.78854191,
      "learning_rate": 1.4567423064988371e-09,
      "loss": 0.82651317,
      "num_input_tokens_seen": 177360210,
      "step": 8218,
      "time_per_iteration": 2.623809576034546
    },
    {
      "auxiliary_loss_clip": 0.01174633,
      "auxiliary_loss_mlp": 0.0102267,
      "balance_loss_clip": 1.04905128,
      "balance_loss_mlp": 1.01551163,
      "epoch": 0.9882763181626887,
      "flos": 21500113374720.0,
      "grad_norm": 2.7320185722078776,
      "language_loss": 0.78169346,
      "learning_rate": 1.4271680706718913e-09,
      "loss": 0.80366647,
      "num_input_tokens_seen": 177377885,
      "step": 8219,
      "time_per_iteration": 3.4504363536834717
    },
    {
      "auxiliary_loss_clip": 0.01224163,
      "auxiliary_loss_mlp": 0.01025758,
      "balance_loss_clip": 1.04962265,
      "balance_loss_mlp": 1.018471,
      "epoch": 0.9883965610533277,
      "flos": 28034598551040.0,
      "grad_norm": 1.7771574709593578,
      "language_loss": 0.82679844,
      "learning_rate": 1.3978970167543013e-09,
      "loss": 0.84929764,
      "num_input_tokens_seen": 177398065,
      "step": 8220,
      "time_per_iteration": 2.660292148590088
    },
    {
      "auxiliary_loss_clip": 0.01265053,
      "auxiliary_loss_mlp": 0.01025771,
      "balance_loss_clip": 1.04522204,
      "balance_loss_mlp": 1.01868093,
      "epoch": 0.9885168039439668,
      "flos": 14098601710080.0,
      "grad_norm": 2.515813970298919,
      "language_loss": 0.77623558,
      "learning_rate": 1.3689291491867372e-09,
      "loss": 0.79914379,
      "num_input_tokens_seen": 177416380,
      "step": 8221,
      "time_per_iteration": 2.6117310523986816
    },
    {
      "auxiliary_loss_clip": 0.01174805,
      "auxiliary_loss_mlp": 0.01026218,
      "balance_loss_clip": 1.04826903,
      "balance_loss_mlp": 1.01888645,
      "epoch": 0.988637046834606,
      "flos": 26432013352320.0,
      "grad_norm": 2.111116640557565,
      "language_loss": 0.73672062,
      "learning_rate": 1.3402644723636836e-09,
      "loss": 0.75873089,
      "num_input_tokens_seen": 177438410,
      "step": 8222,
      "time_per_iteration": 2.669499397277832
    },
    {
      "auxiliary_loss_clip": 0.01266656,
      "auxiliary_loss_mlp": 0.01032064,
      "balance_loss_clip": 1.04659224,
      "balance_loss_mlp": 1.02527499,
      "epoch": 0.988757289725245,
      "flos": 25229113764480.0,
      "grad_norm": 3.4725819723061995,
      "language_loss": 0.83869159,
      "learning_rate": 1.311902990633218e-09,
      "loss": 0.86167878,
      "num_input_tokens_seen": 177457375,
      "step": 8223,
      "time_per_iteration": 2.657245635986328
    },
    {
      "auxiliary_loss_clip": 0.01266104,
      "auxiliary_loss_mlp": 0.01021835,
      "balance_loss_clip": 1.04010022,
      "balance_loss_mlp": 1.01559734,
      "epoch": 0.9888775326158841,
      "flos": 26359042872960.0,
      "grad_norm": 1.732375790744388,
      "language_loss": 0.71430099,
      "learning_rate": 1.2838447082978987e-09,
      "loss": 0.73718035,
      "num_input_tokens_seen": 177478530,
      "step": 8224,
      "time_per_iteration": 3.694157600402832
    },
    {
      "auxiliary_loss_clip": 0.01220423,
      "auxiliary_loss_mlp": 0.01026045,
      "balance_loss_clip": 1.04554367,
      "balance_loss_mlp": 1.0192287,
      "epoch": 0.9889977755065231,
      "flos": 24316120846080.0,
      "grad_norm": 2.3694209103489343,
      "language_loss": 0.83081496,
      "learning_rate": 1.2560896296143208e-09,
      "loss": 0.85327959,
      "num_input_tokens_seen": 177496995,
      "step": 8225,
      "time_per_iteration": 2.6197404861450195
    },
    {
      "auxiliary_loss_clip": 0.01173284,
      "auxiliary_loss_mlp": 0.01029892,
      "balance_loss_clip": 1.04891682,
      "balance_loss_mlp": 1.02282929,
      "epoch": 0.9891180183971623,
      "flos": 18951066760320.0,
      "grad_norm": 2.6583911917708973,
      "language_loss": 0.82342261,
      "learning_rate": 1.2286377587926722e-09,
      "loss": 0.84545445,
      "num_input_tokens_seen": 177513785,
      "step": 8226,
      "time_per_iteration": 2.5583908557891846
    },
    {
      "auxiliary_loss_clip": 0.01173521,
      "auxiliary_loss_mlp": 0.01027688,
      "balance_loss_clip": 1.04880452,
      "balance_loss_mlp": 1.02108645,
      "epoch": 0.9892382612878013,
      "flos": 26176580760960.0,
      "grad_norm": 1.9897324795743438,
      "language_loss": 0.74661732,
      "learning_rate": 1.2014890999973992e-09,
      "loss": 0.76862943,
      "num_input_tokens_seen": 177530705,
      "step": 8227,
      "time_per_iteration": 2.6416029930114746
    },
    {
      "auxiliary_loss_clip": 0.01170308,
      "auxiliary_loss_mlp": 0.010229,
      "balance_loss_clip": 1.04706407,
      "balance_loss_mlp": 1.01615596,
      "epoch": 0.9893585041784404,
      "flos": 25449605400960.0,
      "grad_norm": 1.7743212575479907,
      "language_loss": 0.78537828,
      "learning_rate": 1.1746436573472073e-09,
      "loss": 0.80731034,
      "num_input_tokens_seen": 177552440,
      "step": 8228,
      "time_per_iteration": 2.5901246070861816
    },
    {
      "auxiliary_loss_clip": 0.01280113,
      "auxiliary_loss_mlp": 0.01028452,
      "balance_loss_clip": 1.04609549,
      "balance_loss_mlp": 1.02144885,
      "epoch": 0.9894787470690796,
      "flos": 20189302352640.0,
      "grad_norm": 1.8575825202908962,
      "language_loss": 0.69300544,
      "learning_rate": 1.1481014349141726e-09,
      "loss": 0.71609104,
      "num_input_tokens_seen": 177569660,
      "step": 8229,
      "time_per_iteration": 2.67826247215271
    },
    {
      "auxiliary_loss_clip": 0.01273137,
      "auxiliary_loss_mlp": 0.01028946,
      "balance_loss_clip": 1.04795599,
      "balance_loss_mlp": 1.02191854,
      "epoch": 0.9895989899597186,
      "flos": 24644308435200.0,
      "grad_norm": 1.8811195682849324,
      "language_loss": 0.84608221,
      "learning_rate": 1.121862436724852e-09,
      "loss": 0.86910307,
      "num_input_tokens_seen": 177588500,
      "step": 8230,
      "time_per_iteration": 2.638256311416626
    },
    {
      "auxiliary_loss_clip": 0.01224311,
      "auxiliary_loss_mlp": 0.01029081,
      "balance_loss_clip": 1.04841447,
      "balance_loss_mlp": 1.02241397,
      "epoch": 0.9897192328503577,
      "flos": 21799034357760.0,
      "grad_norm": 1.885545091762791,
      "language_loss": 0.70507789,
      "learning_rate": 1.0959266667598388e-09,
      "loss": 0.7276119,
      "num_input_tokens_seen": 177607315,
      "step": 8231,
      "time_per_iteration": 2.635660409927368
    },
    {
      "auxiliary_loss_clip": 0.01322394,
      "auxiliary_loss_mlp": 0.01031942,
      "balance_loss_clip": 1.04683506,
      "balance_loss_mlp": 1.02425313,
      "epoch": 0.9898394757409968,
      "flos": 21325229032320.0,
      "grad_norm": 1.972129997185183,
      "language_loss": 0.74974751,
      "learning_rate": 1.0702941289533196e-09,
      "loss": 0.77329087,
      "num_input_tokens_seen": 177625990,
      "step": 8232,
      "time_per_iteration": 2.6900534629821777
    },
    {
      "auxiliary_loss_clip": 0.01316678,
      "auxiliary_loss_mlp": 0.01021132,
      "balance_loss_clip": 1.04497766,
      "balance_loss_mlp": 1.01478076,
      "epoch": 0.9899597186316359,
      "flos": 18545024442240.0,
      "grad_norm": 2.2036394324297115,
      "language_loss": 0.89015859,
      "learning_rate": 1.0449648271939615e-09,
      "loss": 0.91353673,
      "num_input_tokens_seen": 177642335,
      "step": 8233,
      "time_per_iteration": 2.6865315437316895
    },
    {
      "auxiliary_loss_clip": 0.0137753,
      "auxiliary_loss_mlp": 0.02565647,
      "balance_loss_clip": 1.04720604,
      "balance_loss_mlp": 0.9998979,
      "epoch": 0.990079961522275,
      "flos": 23766723348480.0,
      "grad_norm": 1.6030436833891195,
      "language_loss": 0.72988456,
      "learning_rate": 1.0199387653240243e-09,
      "loss": 0.76931632,
      "num_input_tokens_seen": 177662025,
      "step": 8234,
      "time_per_iteration": 2.7271647453308105
    },
    {
      "auxiliary_loss_clip": 0.01264605,
      "auxiliary_loss_mlp": 0.01026316,
      "balance_loss_clip": 1.04368567,
      "balance_loss_mlp": 1.01963782,
      "epoch": 0.9902002044129141,
      "flos": 16399182971520.0,
      "grad_norm": 1.6319127973955871,
      "language_loss": 0.70712698,
      "learning_rate": 9.952159471400267e-10,
      "loss": 0.7300362,
      "num_input_tokens_seen": 177679065,
      "step": 8235,
      "time_per_iteration": 2.6614911556243896
    },
    {
      "auxiliary_loss_clip": 0.01122228,
      "auxiliary_loss_mlp": 0.02563062,
      "balance_loss_clip": 1.04588819,
      "balance_loss_mlp": 0.99990261,
      "epoch": 0.9903204473035532,
      "flos": 22559657783040.0,
      "grad_norm": 1.9563788606392252,
      "language_loss": 0.84666598,
      "learning_rate": 9.707963763923022e-10,
      "loss": 0.88351887,
      "num_input_tokens_seen": 177698115,
      "step": 8236,
      "time_per_iteration": 2.596771240234375
    },
    {
      "auxiliary_loss_clip": 0.01271789,
      "auxiliary_loss_mlp": 0.01025963,
      "balance_loss_clip": 1.04394782,
      "balance_loss_mlp": 1.01915932,
      "epoch": 0.9904406901941922,
      "flos": 16144001775360.0,
      "grad_norm": 2.0200909166865495,
      "language_loss": 0.79420328,
      "learning_rate": 9.466800567854427e-10,
      "loss": 0.81718081,
      "num_input_tokens_seen": 177716715,
      "step": 8237,
      "time_per_iteration": 2.6371898651123047
    },
    {
      "auxiliary_loss_clip": 0.01224468,
      "auxiliary_loss_mlp": 0.01026196,
      "balance_loss_clip": 1.04247534,
      "balance_loss_mlp": 1.01833367,
      "epoch": 0.9905609330848314,
      "flos": 26651499408000.0,
      "grad_norm": 2.2746768825630195,
      "language_loss": 0.68423557,
      "learning_rate": 9.228669919778553e-10,
      "loss": 0.70674223,
      "num_input_tokens_seen": 177735640,
      "step": 8238,
      "time_per_iteration": 2.7217857837677
    },
    {
      "auxiliary_loss_clip": 0.01171497,
      "auxiliary_loss_mlp": 0.0102755,
      "balance_loss_clip": 1.0458076,
      "balance_loss_mlp": 1.02057338,
      "epoch": 0.9906811759754705,
      "flos": 23111820627840.0,
      "grad_norm": 2.3064991145048483,
      "language_loss": 0.79578,
      "learning_rate": 8.993571855817617e-10,
      "loss": 0.81777048,
      "num_input_tokens_seen": 177754470,
      "step": 8239,
      "time_per_iteration": 2.6792516708374023
    },
    {
      "auxiliary_loss_clip": 0.01218331,
      "auxiliary_loss_mlp": 0.01025219,
      "balance_loss_clip": 1.04624391,
      "balance_loss_mlp": 1.01821232,
      "epoch": 0.9908014188661095,
      "flos": 22090593052800.0,
      "grad_norm": 1.7849503901281862,
      "language_loss": 0.74983919,
      "learning_rate": 8.761506411638642e-10,
      "loss": 0.77227473,
      "num_input_tokens_seen": 177773935,
      "step": 8240,
      "time_per_iteration": 3.5605368614196777
    },
    {
      "auxiliary_loss_clip": 0.01277972,
      "auxiliary_loss_mlp": 0.01029418,
      "balance_loss_clip": 1.05011463,
      "balance_loss_mlp": 1.02218246,
      "epoch": 0.9909216617567487,
      "flos": 19242948677760.0,
      "grad_norm": 1.737419819130022,
      "language_loss": 0.73721492,
      "learning_rate": 8.53247362244236e-10,
      "loss": 0.76028883,
      "num_input_tokens_seen": 177792745,
      "step": 8241,
      "time_per_iteration": 2.6367921829223633
    },
    {
      "auxiliary_loss_clip": 0.01271682,
      "auxiliary_loss_mlp": 0.01027354,
      "balance_loss_clip": 1.04777193,
      "balance_loss_mlp": 1.02069008,
      "epoch": 0.9910419046473877,
      "flos": 23621213352960.0,
      "grad_norm": 1.5718640704487163,
      "language_loss": 0.68035972,
      "learning_rate": 8.306473522976532e-10,
      "loss": 0.70335007,
      "num_input_tokens_seen": 177812150,
      "step": 8242,
      "time_per_iteration": 3.6063671112060547
    },
    {
      "auxiliary_loss_clip": 0.01173365,
      "auxiliary_loss_mlp": 0.01026641,
      "balance_loss_clip": 1.0494498,
      "balance_loss_mlp": 1.02027822,
      "epoch": 0.9911621475380268,
      "flos": 22711380831360.0,
      "grad_norm": 1.9368527430780065,
      "language_loss": 0.71724153,
      "learning_rate": 8.083506147522623e-10,
      "loss": 0.7392416,
      "num_input_tokens_seen": 177831545,
      "step": 8243,
      "time_per_iteration": 2.5927305221557617
    },
    {
      "auxiliary_loss_clip": 0.01217082,
      "auxiliary_loss_mlp": 0.01027708,
      "balance_loss_clip": 1.04446566,
      "balance_loss_mlp": 1.02064514,
      "epoch": 0.991282390428666,
      "flos": 13516956777600.0,
      "grad_norm": 2.4056212426080728,
      "language_loss": 0.85026175,
      "learning_rate": 7.863571529906909e-10,
      "loss": 0.87270963,
      "num_input_tokens_seen": 177847130,
      "step": 8244,
      "time_per_iteration": 2.61842679977417
    },
    {
      "auxiliary_loss_clip": 0.01115219,
      "auxiliary_loss_mlp": 0.00999409,
      "balance_loss_clip": 1.0056721,
      "balance_loss_mlp": 0.99857444,
      "epoch": 0.991402633319305,
      "flos": 61830492071040.0,
      "grad_norm": 0.724793159400049,
      "language_loss": 0.59664649,
      "learning_rate": 7.646669703489372e-10,
      "loss": 0.61779279,
      "num_input_tokens_seen": 177911440,
      "step": 8245,
      "time_per_iteration": 4.145562410354614
    },
    {
      "auxiliary_loss_clip": 0.01575579,
      "auxiliary_loss_mlp": 0.010274,
      "balance_loss_clip": 1.03677368,
      "balance_loss_mlp": 1.02053654,
      "epoch": 0.9915228762099441,
      "flos": 18770148933120.0,
      "grad_norm": 1.853079492858798,
      "language_loss": 0.57296336,
      "learning_rate": 7.432800701177023e-10,
      "loss": 0.59899318,
      "num_input_tokens_seen": 177929440,
      "step": 8246,
      "time_per_iteration": 2.97361421585083
    },
    {
      "auxiliary_loss_clip": 0.01168451,
      "auxiliary_loss_mlp": 0.00997664,
      "balance_loss_clip": 1.00769687,
      "balance_loss_mlp": 0.99684763,
      "epoch": 0.9916431191005832,
      "flos": 65936660244480.0,
      "grad_norm": 0.7930911764136114,
      "language_loss": 0.57730973,
      "learning_rate": 7.221964555415017e-10,
      "loss": 0.59897089,
      "num_input_tokens_seen": 177989100,
      "step": 8247,
      "time_per_iteration": 3.352449417114258
    },
    {
      "auxiliary_loss_clip": 0.0126747,
      "auxiliary_loss_mlp": 0.01022459,
      "balance_loss_clip": 1.04517436,
      "balance_loss_mlp": 1.01605403,
      "epoch": 0.9917633619912223,
      "flos": 16581573256320.0,
      "grad_norm": 2.1213030856450126,
      "language_loss": 0.75026691,
      "learning_rate": 7.01416129818222e-10,
      "loss": 0.77316618,
      "num_input_tokens_seen": 178006720,
      "step": 8248,
      "time_per_iteration": 2.6481921672821045
    },
    {
      "auxiliary_loss_clip": 0.01330198,
      "auxiliary_loss_mlp": 0.01028322,
      "balance_loss_clip": 1.04645324,
      "balance_loss_mlp": 1.02140498,
      "epoch": 0.9918836048818613,
      "flos": 25411108999680.0,
      "grad_norm": 1.839589443311687,
      "language_loss": 0.58645129,
      "learning_rate": 6.809390961006745e-10,
      "loss": 0.61003649,
      "num_input_tokens_seen": 178026850,
      "step": 8249,
      "time_per_iteration": 2.9813663959503174
    },
    {
      "auxiliary_loss_clip": 0.01272382,
      "auxiliary_loss_mlp": 0.01027807,
      "balance_loss_clip": 1.04713845,
      "balance_loss_mlp": 1.02115822,
      "epoch": 0.9920038477725005,
      "flos": 25046867134080.0,
      "grad_norm": 2.5427472222099223,
      "language_loss": 0.68750244,
      "learning_rate": 6.607653574948191e-10,
      "loss": 0.71050429,
      "num_input_tokens_seen": 178047630,
      "step": 8250,
      "time_per_iteration": 3.798618793487549
    },
    {
      "auxiliary_loss_clip": 0.0121537,
      "auxiliary_loss_mlp": 0.01028206,
      "balance_loss_clip": 1.04439056,
      "balance_loss_mlp": 1.02193546,
      "epoch": 0.9921240906631396,
      "flos": 21829773421440.0,
      "grad_norm": 5.274102653192486,
      "language_loss": 0.81994361,
      "learning_rate": 6.408949170613187e-10,
      "loss": 0.84237933,
      "num_input_tokens_seen": 178066895,
      "step": 8251,
      "time_per_iteration": 2.674574613571167
    },
    {
      "auxiliary_loss_clip": 0.01271978,
      "auxiliary_loss_mlp": 0.01028859,
      "balance_loss_clip": 1.04492271,
      "balance_loss_mlp": 1.02168298,
      "epoch": 0.9922443335537786,
      "flos": 24864225454080.0,
      "grad_norm": 1.6100947657901903,
      "language_loss": 0.8188374,
      "learning_rate": 6.213277778144288e-10,
      "loss": 0.84184575,
      "num_input_tokens_seen": 178088540,
      "step": 8252,
      "time_per_iteration": 2.693537950515747
    },
    {
      "auxiliary_loss_clip": 0.01320276,
      "auxiliary_loss_mlp": 0.01030383,
      "balance_loss_clip": 1.04041672,
      "balance_loss_mlp": 1.02333474,
      "epoch": 0.9923645764444178,
      "flos": 21613088626560.0,
      "grad_norm": 2.3849497898534944,
      "language_loss": 0.67109972,
      "learning_rate": 6.020639427224416e-10,
      "loss": 0.6946063,
      "num_input_tokens_seen": 178106185,
      "step": 8253,
      "time_per_iteration": 2.8271055221557617
    },
    {
      "auxiliary_loss_clip": 0.01273498,
      "auxiliary_loss_mlp": 0.01026123,
      "balance_loss_clip": 1.04859829,
      "balance_loss_mlp": 1.01957583,
      "epoch": 0.9924848193350568,
      "flos": 25001798544000.0,
      "grad_norm": 2.1366953501704193,
      "language_loss": 0.72762918,
      "learning_rate": 5.831034147076864e-10,
      "loss": 0.75062543,
      "num_input_tokens_seen": 178123435,
      "step": 8254,
      "time_per_iteration": 2.6746108531951904
    },
    {
      "auxiliary_loss_clip": 0.01108994,
      "auxiliary_loss_mlp": 0.01001465,
      "balance_loss_clip": 1.00664258,
      "balance_loss_mlp": 1.00064218,
      "epoch": 0.9926050622256959,
      "flos": 68912543151360.0,
      "grad_norm": 0.6874464100316846,
      "language_loss": 0.5567174,
      "learning_rate": 5.644461966463065e-10,
      "loss": 0.57782197,
      "num_input_tokens_seen": 178191045,
      "step": 8255,
      "time_per_iteration": 3.318091869354248
    },
    {
      "auxiliary_loss_clip": 0.01268741,
      "auxiliary_loss_mlp": 0.010272,
      "balance_loss_clip": 1.04662466,
      "balance_loss_mlp": 1.02065253,
      "epoch": 0.9927253051163349,
      "flos": 20923675914240.0,
      "grad_norm": 2.0580963218512225,
      "language_loss": 0.75998747,
      "learning_rate": 5.460922913687049e-10,
      "loss": 0.78294694,
      "num_input_tokens_seen": 178210135,
      "step": 8256,
      "time_per_iteration": 2.647670030593872
    },
    {
      "auxiliary_loss_clip": 0.01368802,
      "auxiliary_loss_mlp": 0.02569728,
      "balance_loss_clip": 1.03998804,
      "balance_loss_mlp": 0.99993229,
      "epoch": 0.9928455480069741,
      "flos": 22308211601280.0,
      "grad_norm": 2.107610087229513,
      "language_loss": 0.75877893,
      "learning_rate": 5.280417016593208e-10,
      "loss": 0.79816425,
      "num_input_tokens_seen": 178229925,
      "step": 8257,
      "time_per_iteration": 2.6987483501434326
    },
    {
      "auxiliary_loss_clip": 0.01220066,
      "auxiliary_loss_mlp": 0.02565173,
      "balance_loss_clip": 1.04970539,
      "balance_loss_mlp": 0.99989796,
      "epoch": 0.9929657908976132,
      "flos": 17383889393280.0,
      "grad_norm": 1.7952312184226038,
      "language_loss": 0.74766886,
      "learning_rate": 5.102944302559642e-10,
      "loss": 0.78552127,
      "num_input_tokens_seen": 178247420,
      "step": 8258,
      "time_per_iteration": 2.6118004322052
    },
    {
      "auxiliary_loss_clip": 0.01432986,
      "auxiliary_loss_mlp": 0.01028194,
      "balance_loss_clip": 1.04680967,
      "balance_loss_mlp": 1.02079153,
      "epoch": 0.9930860337882522,
      "flos": 22674680110080.0,
      "grad_norm": 6.961249800784524,
      "language_loss": 0.79535997,
      "learning_rate": 4.9285047985137e-10,
      "loss": 0.81997174,
      "num_input_tokens_seen": 178266840,
      "step": 8259,
      "time_per_iteration": 2.7769243717193604
    },
    {
      "auxiliary_loss_clip": 0.01225458,
      "auxiliary_loss_mlp": 0.01027094,
      "balance_loss_clip": 1.04714823,
      "balance_loss_mlp": 1.02017951,
      "epoch": 0.9932062766788914,
      "flos": 28147789284480.0,
      "grad_norm": 1.8475480361653513,
      "language_loss": 0.74706525,
      "learning_rate": 4.757098530916436e-10,
      "loss": 0.76959074,
      "num_input_tokens_seen": 178287285,
      "step": 8260,
      "time_per_iteration": 2.7071943283081055
    },
    {
      "auxiliary_loss_clip": 0.01222568,
      "auxiliary_loss_mlp": 0.01024379,
      "balance_loss_clip": 1.048684,
      "balance_loss_mlp": 1.0169847,
      "epoch": 0.9933265195695304,
      "flos": 20156659868160.0,
      "grad_norm": 2.4990950358612647,
      "language_loss": 0.77614993,
      "learning_rate": 4.5887255257670563e-10,
      "loss": 0.79861939,
      "num_input_tokens_seen": 178304325,
      "step": 8261,
      "time_per_iteration": 2.586324453353882
    },
    {
      "auxiliary_loss_clip": 0.01171778,
      "auxiliary_loss_mlp": 0.01029701,
      "balance_loss_clip": 1.04646397,
      "balance_loss_mlp": 1.02270365,
      "epoch": 0.9934467624601695,
      "flos": 21362037494400.0,
      "grad_norm": 2.087594577338146,
      "language_loss": 0.77043825,
      "learning_rate": 4.4233858086117906e-10,
      "loss": 0.79245305,
      "num_input_tokens_seen": 178322850,
      "step": 8262,
      "time_per_iteration": 2.575388193130493
    },
    {
      "auxiliary_loss_clip": 0.01364587,
      "auxiliary_loss_mlp": 0.01022373,
      "balance_loss_clip": 1.0470382,
      "balance_loss_mlp": 1.01560199,
      "epoch": 0.9935670053508087,
      "flos": 19756040503680.0,
      "grad_norm": 2.1979293126992623,
      "language_loss": 0.67533958,
      "learning_rate": 4.261079404528356e-10,
      "loss": 0.69920921,
      "num_input_tokens_seen": 178342330,
      "step": 8263,
      "time_per_iteration": 2.704423666000366
    },
    {
      "auxiliary_loss_clip": 0.01222743,
      "auxiliary_loss_mlp": 0.0102541,
      "balance_loss_clip": 1.0470705,
      "balance_loss_mlp": 1.01859152,
      "epoch": 0.9936872482414477,
      "flos": 21978838863360.0,
      "grad_norm": 1.7723709137403312,
      "language_loss": 0.69250828,
      "learning_rate": 4.1018063381437205e-10,
      "loss": 0.71498978,
      "num_input_tokens_seen": 178362715,
      "step": 8264,
      "time_per_iteration": 2.6404080390930176
    },
    {
      "auxiliary_loss_clip": 0.01106696,
      "auxiliary_loss_mlp": 0.01002037,
      "balance_loss_clip": 1.00616717,
      "balance_loss_mlp": 1.00122643,
      "epoch": 0.9938074911320868,
      "flos": 69810667839360.0,
      "grad_norm": 0.8664121817183589,
      "language_loss": 0.61051381,
      "learning_rate": 3.9455666336141167e-10,
      "loss": 0.6316011,
      "num_input_tokens_seen": 178426495,
      "step": 8265,
      "time_per_iteration": 4.189847230911255
    },
    {
      "auxiliary_loss_clip": 0.01171756,
      "auxiliary_loss_mlp": 0.01025891,
      "balance_loss_clip": 1.049824,
      "balance_loss_mlp": 1.01834512,
      "epoch": 0.9939277340227259,
      "flos": 15084170058240.0,
      "grad_norm": 2.5981942749163447,
      "language_loss": 0.83243465,
      "learning_rate": 3.7923603146450267e-10,
      "loss": 0.85441113,
      "num_input_tokens_seen": 178442555,
      "step": 8266,
      "time_per_iteration": 2.554323196411133
    },
    {
      "auxiliary_loss_clip": 0.0131973,
      "auxiliary_loss_mlp": 0.0102767,
      "balance_loss_clip": 1.04146838,
      "balance_loss_mlp": 1.02066612,
      "epoch": 0.994047976913365,
      "flos": 17712364291200.0,
      "grad_norm": 2.1075153648663907,
      "language_loss": 0.80755651,
      "learning_rate": 3.642187404473418e-10,
      "loss": 0.83103049,
      "num_input_tokens_seen": 178460715,
      "step": 8267,
      "time_per_iteration": 2.6651298999786377
    },
    {
      "auxiliary_loss_clip": 0.01224003,
      "auxiliary_loss_mlp": 0.01021822,
      "balance_loss_clip": 1.04696393,
      "balance_loss_mlp": 1.01557803,
      "epoch": 0.994168219804004,
      "flos": 19171558396800.0,
      "grad_norm": 2.202421420730303,
      "language_loss": 0.85887378,
      "learning_rate": 3.495047925885508e-10,
      "loss": 0.88133204,
      "num_input_tokens_seen": 178479050,
      "step": 8268,
      "time_per_iteration": 3.6232926845550537
    },
    {
      "auxiliary_loss_clip": 0.01272489,
      "auxiliary_loss_mlp": 0.01024424,
      "balance_loss_clip": 1.04426634,
      "balance_loss_mlp": 1.01785612,
      "epoch": 0.9942884626946432,
      "flos": 17851589406720.0,
      "grad_norm": 2.0403948585528457,
      "language_loss": 0.82862425,
      "learning_rate": 3.350941901199e-10,
      "loss": 0.85159338,
      "num_input_tokens_seen": 178495970,
      "step": 8269,
      "time_per_iteration": 2.6619069576263428
    },
    {
      "auxiliary_loss_clip": 0.01276345,
      "auxiliary_loss_mlp": 0.01025707,
      "balance_loss_clip": 1.04672098,
      "balance_loss_mlp": 1.018888,
      "epoch": 0.9944087055852823,
      "flos": 18796578364800.0,
      "grad_norm": 4.381460049079689,
      "language_loss": 0.83176243,
      "learning_rate": 3.2098693522764066e-10,
      "loss": 0.85478294,
      "num_input_tokens_seen": 178509170,
      "step": 8270,
      "time_per_iteration": 2.5806376934051514
    },
    {
      "auxiliary_loss_clip": 0.01278876,
      "auxiliary_loss_mlp": 0.0256693,
      "balance_loss_clip": 1.04664814,
      "balance_loss_mlp": 0.9999125,
      "epoch": 0.9945289484759213,
      "flos": 20996969616000.0,
      "grad_norm": 1.9862124639852408,
      "language_loss": 0.8114925,
      "learning_rate": 3.071830300516165e-10,
      "loss": 0.84995055,
      "num_input_tokens_seen": 178527000,
      "step": 8271,
      "time_per_iteration": 3.549713134765625
    },
    {
      "auxiliary_loss_clip": 0.01229657,
      "auxiliary_loss_mlp": 0.01024282,
      "balance_loss_clip": 1.0484159,
      "balance_loss_mlp": 1.01682544,
      "epoch": 0.9946491913665605,
      "flos": 14756952136320.0,
      "grad_norm": 1.9923505991223773,
      "language_loss": 0.70948619,
      "learning_rate": 2.9368247668615234e-10,
      "loss": 0.73202562,
      "num_input_tokens_seen": 178545590,
      "step": 8272,
      "time_per_iteration": 2.610046148300171
    },
    {
      "auxiliary_loss_clip": 0.0117876,
      "auxiliary_loss_mlp": 0.01028548,
      "balance_loss_clip": 1.05091119,
      "balance_loss_mlp": 1.02103758,
      "epoch": 0.9947694342571995,
      "flos": 12669931186560.0,
      "grad_norm": 2.8938081311334796,
      "language_loss": 0.61349654,
      "learning_rate": 2.804852771789434e-10,
      "loss": 0.63556957,
      "num_input_tokens_seen": 178558890,
      "step": 8273,
      "time_per_iteration": 2.525644302368164
    },
    {
      "auxiliary_loss_clip": 0.01172011,
      "auxiliary_loss_mlp": 0.01024565,
      "balance_loss_clip": 1.0489403,
      "balance_loss_mlp": 1.01820481,
      "epoch": 0.9948896771478386,
      "flos": 18843442634880.0,
      "grad_norm": 1.8177024596603115,
      "language_loss": 0.55789065,
      "learning_rate": 2.675914335321661e-10,
      "loss": 0.5798564,
      "num_input_tokens_seen": 178577645,
      "step": 8274,
      "time_per_iteration": 2.5766186714172363
    },
    {
      "auxiliary_loss_clip": 0.01226155,
      "auxiliary_loss_mlp": 0.01026836,
      "balance_loss_clip": 1.04766548,
      "balance_loss_mlp": 1.01988912,
      "epoch": 0.9950099200384778,
      "flos": 24900207903360.0,
      "grad_norm": 3.897416602262242,
      "language_loss": 0.79329693,
      "learning_rate": 2.550009477018111e-10,
      "loss": 0.81582689,
      "num_input_tokens_seen": 178596415,
      "step": 8275,
      "time_per_iteration": 2.6283960342407227
    },
    {
      "auxiliary_loss_clip": 0.01270895,
      "auxiliary_loss_mlp": 0.0256862,
      "balance_loss_clip": 1.04708946,
      "balance_loss_mlp": 0.99990827,
      "epoch": 0.9951301629291168,
      "flos": 23733613987200.0,
      "grad_norm": 2.5261573803526938,
      "language_loss": 0.62699085,
      "learning_rate": 2.4271382159790634e-10,
      "loss": 0.66538602,
      "num_input_tokens_seen": 178613845,
      "step": 8276,
      "time_per_iteration": 3.6178090572357178
    },
    {
      "auxiliary_loss_clip": 0.01331627,
      "auxiliary_loss_mlp": 0.01027785,
      "balance_loss_clip": 1.04379201,
      "balance_loss_mlp": 1.02112114,
      "epoch": 0.9952504058197559,
      "flos": 22236893147520.0,
      "grad_norm": 3.284350914487527,
      "language_loss": 0.86181271,
      "learning_rate": 2.3073005708429406e-10,
      "loss": 0.88540685,
      "num_input_tokens_seen": 178633490,
      "step": 8277,
      "time_per_iteration": 2.7749431133270264
    },
    {
      "auxiliary_loss_clip": 0.01317153,
      "auxiliary_loss_mlp": 0.01027849,
      "balance_loss_clip": 1.04683495,
      "balance_loss_mlp": 1.02118218,
      "epoch": 0.995370648710395,
      "flos": 21211032718080.0,
      "grad_norm": 1.7289501268489524,
      "language_loss": 0.71922004,
      "learning_rate": 2.190496559788535e-10,
      "loss": 0.74267006,
      "num_input_tokens_seen": 178651775,
      "step": 8278,
      "time_per_iteration": 2.6886322498321533
    },
    {
      "auxiliary_loss_clip": 0.01266364,
      "auxiliary_loss_mlp": 0.01029764,
      "balance_loss_clip": 1.04628491,
      "balance_loss_mlp": 1.02275133,
      "epoch": 0.9954908916010341,
      "flos": 14866731077760.0,
      "grad_norm": 2.5164284264135977,
      "language_loss": 0.76700217,
      "learning_rate": 2.0767262005372265e-10,
      "loss": 0.78996348,
      "num_input_tokens_seen": 178669290,
      "step": 8279,
      "time_per_iteration": 2.626100540161133
    },
    {
      "auxiliary_loss_clip": 0.01226965,
      "auxiliary_loss_mlp": 0.0102771,
      "balance_loss_clip": 1.04328895,
      "balance_loss_mlp": 1.02073669,
      "epoch": 0.9956111344916732,
      "flos": 19208259118080.0,
      "grad_norm": 1.835738529140405,
      "language_loss": 0.75421786,
      "learning_rate": 1.965989510346322e-10,
      "loss": 0.77676463,
      "num_input_tokens_seen": 178688410,
      "step": 8280,
      "time_per_iteration": 2.706963062286377
    },
    {
      "auxiliary_loss_clip": 0.01366565,
      "auxiliary_loss_mlp": 0.01021671,
      "balance_loss_clip": 1.04222035,
      "balance_loss_mlp": 1.01467967,
      "epoch": 0.9957313773823123,
      "flos": 20047060494720.0,
      "grad_norm": 3.07666879868278,
      "language_loss": 0.71233273,
      "learning_rate": 1.8582865060134955e-10,
      "loss": 0.73621505,
      "num_input_tokens_seen": 178706600,
      "step": 8281,
      "time_per_iteration": 2.7134578227996826
    },
    {
      "auxiliary_loss_clip": 0.01059461,
      "auxiliary_loss_mlp": 0.01000082,
      "balance_loss_clip": 1.00554872,
      "balance_loss_mlp": 0.99922991,
      "epoch": 0.9958516202729514,
      "flos": 57483253768320.0,
      "grad_norm": 0.7837749649795689,
      "language_loss": 0.55701721,
      "learning_rate": 1.7536172038790098e-10,
      "loss": 0.57761264,
      "num_input_tokens_seen": 178766910,
      "step": 8282,
      "time_per_iteration": 3.2197062969207764
    },
    {
      "auxiliary_loss_clip": 0.01270615,
      "auxiliary_loss_mlp": 0.01026035,
      "balance_loss_clip": 1.04886556,
      "balance_loss_mlp": 1.01929092,
      "epoch": 0.9959718631635904,
      "flos": 27782900974080.0,
      "grad_norm": 2.555805455767541,
      "language_loss": 0.69221461,
      "learning_rate": 1.651981619819054e-10,
      "loss": 0.71518111,
      "num_input_tokens_seen": 178784060,
      "step": 8283,
      "time_per_iteration": 2.7209367752075195
    },
    {
      "auxiliary_loss_clip": 0.01371646,
      "auxiliary_loss_mlp": 0.0102467,
      "balance_loss_clip": 1.04457593,
      "balance_loss_mlp": 1.01774073,
      "epoch": 0.9960921060542296,
      "flos": 24024095274240.0,
      "grad_norm": 2.7460937630509363,
      "language_loss": 0.71049869,
      "learning_rate": 1.5533797692546257e-10,
      "loss": 0.73446178,
      "num_input_tokens_seen": 178802795,
      "step": 8284,
      "time_per_iteration": 2.7168235778808594
    },
    {
      "auxiliary_loss_clip": 0.01221276,
      "auxiliary_loss_mlp": 0.01022514,
      "balance_loss_clip": 1.04583442,
      "balance_loss_mlp": 1.01430297,
      "epoch": 0.9962123489448687,
      "flos": 18697393935360.0,
      "grad_norm": 2.13330092953392,
      "language_loss": 0.84591275,
      "learning_rate": 1.4578116671404296e-10,
      "loss": 0.86835068,
      "num_input_tokens_seen": 178821075,
      "step": 8285,
      "time_per_iteration": 2.6027705669403076
    },
    {
      "auxiliary_loss_clip": 0.01218959,
      "auxiliary_loss_mlp": 0.01025328,
      "balance_loss_clip": 1.04902744,
      "balance_loss_mlp": 1.01898956,
      "epoch": 0.9963325918355077,
      "flos": 20010754823040.0,
      "grad_norm": 2.0591794784701762,
      "language_loss": 0.71305752,
      "learning_rate": 1.3652773279759777e-10,
      "loss": 0.73550045,
      "num_input_tokens_seen": 178837725,
      "step": 8286,
      "time_per_iteration": 2.590189218521118
    },
    {
      "auxiliary_loss_clip": 0.0122405,
      "auxiliary_loss_mlp": 0.01024476,
      "balance_loss_clip": 1.04729974,
      "balance_loss_mlp": 1.0173533,
      "epoch": 0.9964528347261468,
      "flos": 33108488991360.0,
      "grad_norm": 2.215568208095501,
      "language_loss": 0.63308001,
      "learning_rate": 1.2757767657989305e-10,
      "loss": 0.65556532,
      "num_input_tokens_seen": 178861515,
      "step": 8287,
      "time_per_iteration": 2.7125871181488037
    },
    {
      "auxiliary_loss_clip": 0.01219129,
      "auxiliary_loss_mlp": 0.01027652,
      "balance_loss_clip": 1.04864883,
      "balance_loss_mlp": 1.0209049,
      "epoch": 0.9965730776167859,
      "flos": 23109342589440.0,
      "grad_norm": 2.0780448475358346,
      "language_loss": 0.87128896,
      "learning_rate": 1.1893099941850948e-10,
      "loss": 0.89375675,
      "num_input_tokens_seen": 178880410,
      "step": 8288,
      "time_per_iteration": 2.6276051998138428
    },
    {
      "auxiliary_loss_clip": 0.01273331,
      "auxiliary_loss_mlp": 0.01023768,
      "balance_loss_clip": 1.04424393,
      "balance_loss_mlp": 1.01669025,
      "epoch": 0.996693320507425,
      "flos": 22965843755520.0,
      "grad_norm": 2.1174305018533968,
      "language_loss": 0.77343822,
      "learning_rate": 1.105877026252866e-10,
      "loss": 0.79640919,
      "num_input_tokens_seen": 178898740,
      "step": 8289,
      "time_per_iteration": 2.6987154483795166
    },
    {
      "auxiliary_loss_clip": 0.01175114,
      "auxiliary_loss_mlp": 0.01025437,
      "balance_loss_clip": 1.04835653,
      "balance_loss_mlp": 1.01870728,
      "epoch": 0.996813563398064,
      "flos": 13222740476160.0,
      "grad_norm": 2.2235566466693273,
      "language_loss": 0.72044861,
      "learning_rate": 1.0254778746565663e-10,
      "loss": 0.74245417,
      "num_input_tokens_seen": 178914015,
      "step": 8290,
      "time_per_iteration": 2.502345323562622
    },
    {
      "auxiliary_loss_clip": 0.01317268,
      "auxiliary_loss_mlp": 0.01024002,
      "balance_loss_clip": 1.0458169,
      "balance_loss_mlp": 1.01793671,
      "epoch": 0.9969338062887032,
      "flos": 14647855553280.0,
      "grad_norm": 7.8203287171458475,
      "language_loss": 0.73555142,
      "learning_rate": 9.481125515953259e-11,
      "loss": 0.75896418,
      "num_input_tokens_seen": 178932075,
      "step": 8291,
      "time_per_iteration": 2.6679654121398926
    },
    {
      "auxiliary_loss_clip": 0.01374295,
      "auxiliary_loss_mlp": 0.01025655,
      "balance_loss_clip": 1.04029775,
      "balance_loss_mlp": 1.0182426,
      "epoch": 0.9970540491793423,
      "flos": 25735741142400.0,
      "grad_norm": 1.9637415972133963,
      "language_loss": 0.79970729,
      "learning_rate": 8.737810688064228e-11,
      "loss": 0.82370681,
      "num_input_tokens_seen": 178951910,
      "step": 8292,
      "time_per_iteration": 3.8468434810638428
    },
    {
      "auxiliary_loss_clip": 0.01325393,
      "auxiliary_loss_mlp": 0.01028795,
      "balance_loss_clip": 1.04889655,
      "balance_loss_mlp": 1.02146935,
      "epoch": 0.9971742920699813,
      "flos": 21470236237440.0,
      "grad_norm": 2.0976999052592067,
      "language_loss": 0.79346097,
      "learning_rate": 8.024834375608414e-11,
      "loss": 0.81700283,
      "num_input_tokens_seen": 178970500,
      "step": 8293,
      "time_per_iteration": 2.699910879135132
    },
    {
      "auxiliary_loss_clip": 0.01059351,
      "auxiliary_loss_mlp": 0.01000988,
      "balance_loss_clip": 1.00548577,
      "balance_loss_mlp": 1.0001421,
      "epoch": 0.9972945349606205,
      "flos": 72211223629440.0,
      "grad_norm": 0.8138728572583651,
      "language_loss": 0.62743378,
      "learning_rate": 7.342196686788149e-11,
      "loss": 0.6480372,
      "num_input_tokens_seen": 179023665,
      "step": 8294,
      "time_per_iteration": 4.014256715774536
    },
    {
      "auxiliary_loss_clip": 0.01275454,
      "auxiliary_loss_mlp": 0.01028536,
      "balance_loss_clip": 1.05133569,
      "balance_loss_mlp": 1.02086532,
      "epoch": 0.9974147778512595,
      "flos": 19678293515520.0,
      "grad_norm": 2.121645470943436,
      "language_loss": 0.68304378,
      "learning_rate": 6.689897725142834e-11,
      "loss": 0.70608366,
      "num_input_tokens_seen": 179043140,
      "step": 8295,
      "time_per_iteration": 2.6812031269073486
    },
    {
      "auxiliary_loss_clip": 0.01273591,
      "auxiliary_loss_mlp": 0.01024333,
      "balance_loss_clip": 1.04457998,
      "balance_loss_mlp": 1.01793146,
      "epoch": 0.9975350207418986,
      "flos": 15960821391360.0,
      "grad_norm": 2.14495053396276,
      "language_loss": 0.88326216,
      "learning_rate": 6.067937589615545e-11,
      "loss": 0.90624142,
      "num_input_tokens_seen": 179061215,
      "step": 8296,
      "time_per_iteration": 2.618645191192627
    },
    {
      "auxiliary_loss_clip": 0.01173238,
      "auxiliary_loss_mlp": 0.009987,
      "balance_loss_clip": 1.00645947,
      "balance_loss_mlp": 0.99777621,
      "epoch": 0.9976552636325378,
      "flos": 59961879768960.0,
      "grad_norm": 0.7422660478103639,
      "language_loss": 0.57662177,
      "learning_rate": 5.476316374575241e-11,
      "loss": 0.59834117,
      "num_input_tokens_seen": 179124700,
      "step": 8297,
      "time_per_iteration": 4.100350618362427
    },
    {
      "auxiliary_loss_clip": 0.0117807,
      "auxiliary_loss_mlp": 0.0102415,
      "balance_loss_clip": 1.0519346,
      "balance_loss_mlp": 1.01674128,
      "epoch": 0.9977755065231768,
      "flos": 22487872452480.0,
      "grad_norm": 1.8891234748686514,
      "language_loss": 0.72461593,
      "learning_rate": 4.9150341697723476e-11,
      "loss": 0.74663818,
      "num_input_tokens_seen": 179144590,
      "step": 8298,
      "time_per_iteration": 2.5788798332214355
    },
    {
      "auxiliary_loss_clip": 0.01273367,
      "auxiliary_loss_mlp": 0.0102783,
      "balance_loss_clip": 1.04802251,
      "balance_loss_mlp": 1.02077579,
      "epoch": 0.9978957494138159,
      "flos": 26030280666240.0,
      "grad_norm": 3.797125607120145,
      "language_loss": 0.66769445,
      "learning_rate": 4.384091060338768e-11,
      "loss": 0.69070637,
      "num_input_tokens_seen": 179165060,
      "step": 8299,
      "time_per_iteration": 2.744431495666504
    },
    {
      "auxiliary_loss_clip": 0.01220883,
      "auxiliary_loss_mlp": 0.010309,
      "balance_loss_clip": 1.0481689,
      "balance_loss_mlp": 1.02425373,
      "epoch": 0.998015992304455,
      "flos": 22637835734400.0,
      "grad_norm": 2.1358664017492495,
      "language_loss": 0.73904645,
      "learning_rate": 3.883487126810081e-11,
      "loss": 0.76156431,
      "num_input_tokens_seen": 179184320,
      "step": 8300,
      "time_per_iteration": 2.6785454750061035
    },
    {
      "auxiliary_loss_clip": 0.01216942,
      "auxiliary_loss_mlp": 0.01021535,
      "balance_loss_clip": 1.04495716,
      "balance_loss_mlp": 1.01470459,
      "epoch": 0.9981362351950941,
      "flos": 18223444955520.0,
      "grad_norm": 1.7178017199499926,
      "language_loss": 0.79270673,
      "learning_rate": 3.41322244516995e-11,
      "loss": 0.81509149,
      "num_input_tokens_seen": 179202265,
      "step": 8301,
      "time_per_iteration": 2.580578565597534
    },
    {
      "auxiliary_loss_clip": 0.01366713,
      "auxiliary_loss_mlp": 0.01018442,
      "balance_loss_clip": 1.04535782,
      "balance_loss_mlp": 1.01188254,
      "epoch": 0.9982564780857331,
      "flos": 33474095573760.0,
      "grad_norm": 2.6372342342755903,
      "language_loss": 0.63270956,
      "learning_rate": 2.9732970866946925e-11,
      "loss": 0.65656114,
      "num_input_tokens_seen": 179222145,
      "step": 8302,
      "time_per_iteration": 3.6992685794830322
    },
    {
      "auxiliary_loss_clip": 0.01312546,
      "auxiliary_loss_mlp": 0.01026738,
      "balance_loss_clip": 1.03934407,
      "balance_loss_mlp": 1.01885843,
      "epoch": 0.9983767209763723,
      "flos": 15523465392000.0,
      "grad_norm": 2.4303722289168506,
      "language_loss": 0.78319311,
      "learning_rate": 2.563711118175327e-11,
      "loss": 0.80658597,
      "num_input_tokens_seen": 179239030,
      "step": 8303,
      "time_per_iteration": 2.6560866832733154
    },
    {
      "auxiliary_loss_clip": 0.01318774,
      "auxiliary_loss_mlp": 0.01022629,
      "balance_loss_clip": 1.04506075,
      "balance_loss_mlp": 1.01578021,
      "epoch": 0.9984969638670114,
      "flos": 19974377324160.0,
      "grad_norm": 1.7953875248328464,
      "language_loss": 0.83726948,
      "learning_rate": 2.184464601717728e-11,
      "loss": 0.8606835,
      "num_input_tokens_seen": 179257345,
      "step": 8304,
      "time_per_iteration": 2.73249888420105
    },
    {
      "auxiliary_loss_clip": 0.01225098,
      "auxiliary_loss_mlp": 0.01025899,
      "balance_loss_clip": 1.05012465,
      "balance_loss_mlp": 1.01913428,
      "epoch": 0.9986172067576504,
      "flos": 20375750874240.0,
      "grad_norm": 3.254746145386755,
      "language_loss": 0.77854997,
      "learning_rate": 1.8355575948758585e-11,
      "loss": 0.80105996,
      "num_input_tokens_seen": 179275330,
      "step": 8305,
      "time_per_iteration": 2.5849852561950684
    },
    {
      "auxiliary_loss_clip": 0.01274986,
      "auxiliary_loss_mlp": 0.01025668,
      "balance_loss_clip": 1.04343617,
      "balance_loss_mlp": 1.01828897,
      "epoch": 0.9987374496482896,
      "flos": 23727903724800.0,
      "grad_norm": 2.2461773021851252,
      "language_loss": 0.73311555,
      "learning_rate": 1.5169901505407424e-11,
      "loss": 0.75612211,
      "num_input_tokens_seen": 179292395,
      "step": 8306,
      "time_per_iteration": 2.6472344398498535
    },
    {
      "auxiliary_loss_clip": 0.01268792,
      "auxiliary_loss_mlp": 0.01026516,
      "balance_loss_clip": 1.0458802,
      "balance_loss_mlp": 1.01963139,
      "epoch": 0.9988576925389286,
      "flos": 25044029959680.0,
      "grad_norm": 2.3257123070181205,
      "language_loss": 0.73997712,
      "learning_rate": 1.228762317073695e-11,
      "loss": 0.76293015,
      "num_input_tokens_seen": 179311225,
      "step": 8307,
      "time_per_iteration": 2.6951401233673096
    },
    {
      "auxiliary_loss_clip": 0.01278376,
      "auxiliary_loss_mlp": 0.01024341,
      "balance_loss_clip": 1.05009913,
      "balance_loss_mlp": 1.01773131,
      "epoch": 0.9989779354295677,
      "flos": 31285627637760.0,
      "grad_norm": 3.7833897835449197,
      "language_loss": 0.78899413,
      "learning_rate": 9.70874138195299e-12,
      "loss": 0.81202132,
      "num_input_tokens_seen": 179333135,
      "step": 8308,
      "time_per_iteration": 2.794603109359741
    },
    {
      "auxiliary_loss_clip": 0.0117323,
      "auxiliary_loss_mlp": 0.0101993,
      "balance_loss_clip": 1.04802775,
      "balance_loss_mlp": 1.01255655,
      "epoch": 0.9990981783202069,
      "flos": 19573398823680.0,
      "grad_norm": 1.7461949368624077,
      "language_loss": 0.74768525,
      "learning_rate": 7.433256530076093e-12,
      "loss": 0.7696169,
      "num_input_tokens_seen": 179353090,
      "step": 8309,
      "time_per_iteration": 2.569227933883667
    },
    {
      "auxiliary_loss_clip": 0.01376099,
      "auxiliary_loss_mlp": 0.01024656,
      "balance_loss_clip": 1.04195631,
      "balance_loss_mlp": 1.01818347,
      "epoch": 0.9992184212108459,
      "flos": 17199667514880.0,
      "grad_norm": 4.343984170375211,
      "language_loss": 0.76254237,
      "learning_rate": 5.46116896038562e-12,
      "loss": 0.78654993,
      "num_input_tokens_seen": 179367500,
      "step": 8310,
      "time_per_iteration": 2.71580171585083
    },
    {
      "auxiliary_loss_clip": 0.01276571,
      "auxiliary_loss_mlp": 0.0102613,
      "balance_loss_clip": 1.04806483,
      "balance_loss_mlp": 1.01929939,
      "epoch": 0.999338664101485,
      "flos": 46497853681920.0,
      "grad_norm": 1.981616554235785,
      "language_loss": 0.61754906,
      "learning_rate": 3.792478972197699e-12,
      "loss": 0.64057606,
      "num_input_tokens_seen": 179388085,
      "step": 8311,
      "time_per_iteration": 2.8900208473205566
    },
    {
      "auxiliary_loss_clip": 0.01169913,
      "auxiliary_loss_mlp": 0.01026209,
      "balance_loss_clip": 1.04646969,
      "balance_loss_mlp": 1.01954818,
      "epoch": 0.9994589069921241,
      "flos": 15158253859200.0,
      "grad_norm": 2.1434060406161652,
      "language_loss": 0.70507061,
      "learning_rate": 2.4271868181990895e-12,
      "loss": 0.72703189,
      "num_input_tokens_seen": 179405250,
      "step": 8312,
      "time_per_iteration": 2.543121337890625
    },
    {
      "auxiliary_loss_clip": 0.01223641,
      "auxiliary_loss_mlp": 0.01027369,
      "balance_loss_clip": 1.04652345,
      "balance_loss_mlp": 1.0207653,
      "epoch": 0.9995791498827632,
      "flos": 12531460256640.0,
      "grad_norm": 2.2964391667372563,
      "language_loss": 0.80915999,
      "learning_rate": 1.3652927060014973e-12,
      "loss": 0.83167011,
      "num_input_tokens_seen": 179420845,
      "step": 8313,
      "time_per_iteration": 2.5957930088043213
    },
    {
      "auxiliary_loss_clip": 0.01329106,
      "auxiliary_loss_mlp": 0.01024161,
      "balance_loss_clip": 1.04728079,
      "balance_loss_mlp": 1.01719308,
      "epoch": 0.9996993927734023,
      "flos": 19245175320960.0,
      "grad_norm": 2.177331923082963,
      "language_loss": 0.63700873,
      "learning_rate": 6.067967965872612e-13,
      "loss": 0.66054142,
      "num_input_tokens_seen": 179440455,
      "step": 8314,
      "time_per_iteration": 2.655853033065796
    },
    {
      "auxiliary_loss_clip": 0.01318594,
      "auxiliary_loss_mlp": 0.01025994,
      "balance_loss_clip": 1.04592776,
      "balance_loss_mlp": 1.01957464,
      "epoch": 0.9998196356640414,
      "flos": 62952804518400.0,
      "grad_norm": 1.7899771234736448,
      "language_loss": 0.77230346,
      "learning_rate": 1.5169920497548615e-13,
      "loss": 0.79574937,
      "num_input_tokens_seen": 179465075,
      "step": 8315,
      "time_per_iteration": 3.083948850631714
    },
    {
      "auxiliary_loss_clip": 0.01171628,
      "auxiliary_loss_mlp": 0.01011211,
      "balance_loss_clip": 1.02671599,
      "balance_loss_mlp": 1.00722528,
      "epoch": 0.9999398785546805,
      "flos": 50922375073920.0,
      "grad_norm": 1.1213571057662117,
      "language_loss": 0.54982269,
      "learning_rate": 0.0,
      "loss": 0.5716511,
      "num_input_tokens_seen": 179513955,
      "step": 8316,
      "time_per_iteration": 3.182302474975586
    },
    {
      "epoch": 0.9999398785546805,
      "num_input_tokens_seen": 179513955,
      "step": 8316,
      "total_flos": 6.996749092776837e+17,
      "train_loss": 0.7923422268286756,
      "train_runtime": 25183.1629,
      "train_samples_per_second": 13.209,
      "train_steps_per_second": 0.33
    }
  ],
  "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
}